2015 ACVIM Forum Research Abstract Program

Abstract

2015 ACVIM Forum Research Abstract Program Indianapolis, Indiana, June 3–6, 2015 Index of Abstracts Oral Presentations – Thursday, June 4 Bolded type indicates ACVIM Resident Research Award eligibility Time # Presenting Author Abstract Title SMALL ANIMAL ‐ CARDIOLOGY a 9:00 am C‐1 Junseok Lee Cardaic Specific Calcium Uptake Genes Expressed in Peripheral Blood Effectively Reflect Myocardial Distress Induced By Hemodynamic Change 9:15 am C‐2 Melanie Hezzell Differentiating Cardiac Vs Non‐Cardiac Causes of Pleural Effusion in Cats Using Plasma and Pleural Fluid with a Point‐of‐Care NT‐proBNP Test 9:30 am C‐3 Melanie Hezzell Pre‐Specified Escalation of Medical Therapy Reduces Plasma NT‐proBNP Concentrations in Dogs with Stable CHF Due to Mitral Valve Disease 9:45 am C‐4 Autumn Harris Biological Variability of N‐Terminal Pro‐Brain Natriuretic Peptide in Adult Healthy Cats  BREAK 10:30 am C‐5 SeungWoo Jung Microrna Signaling Networks in a Canine Model of Mitral Regurgitation 10:45 am C‐6 Tatsuyuki Osuga Effect of Acute Volume Overload on the Left Atrial Phasic Function in Healthy Dogs 11:00 am C‐7 Caio Nogueira Duarte Doppler Echocardiographic Assessment Of Left Ventricular +Dp/Dt and ‐Dp/Dt in Dogs with Chronic Mitral Valve Disease 11:15 am C‐8 Bruno Boutet Clinical Characteristics in 35 Dogs Over 5 Years of Age Diagnosed with Patent Ductus Arteriosus 11:30 am C‐9 Lee Chang‐Min Evaluation of the Correlation Between Serum Homocysteine Concentrations and the Severity of Mitral Valve Disease 11:45 am C‐10 Justin Thomason Echocardiographic Diagnosis and Outcome of Constrictive Pericardial Disease in Dogs (18 Cases:  2002‐2013) 12:00 pm C‐11 Catherine Gunther‐Harrington Timolol Ophthalmic Solution for Diastolic Function Testing in Cats: A Pilot Study of Heart Rate and Echocardiographic Effects 12:15 pm C‐12 Ilaria Spalla Changes in the Biomechanics of the Left Ventricle in Healthy Young and Adult Great Danes BREAK 4:30 pm C‐13 Kelly Flynn Plasma L‐Citrulline Concentrations in L‐Arginine‐Treated Dogs: A Pilot Study for the Treatment of Pulmonary Hypertension 4:45 pm C‐14 Fernando Rosa Left Atrial Volume Obtained By Biplane Simpson Method in Healthy Dogs: Body Weight and Body Surface Area 5:00 pm C‐15 Aparecido Camacho Determination of Lactate Threshold in Dogs with Degenerative Mitral Valve Disease Submitted to Incremental Exercise Test 5:15 pm C‐16 Kathyn Meurs Familial Ventricular Arrhythmias and Sudden Cardiac Death in the Young Rhodesian Ridgeback 5:30 pm C‐17 William Tyrrell NT‐proBNP and Troponin I (Ctni) Levels As Screening Biomarkers in Irish Wolfhounds a Also See Small Animal Cardiology abstracts C‐18 ‐ C‐25, Friday, June 5, 8:00 am ‐ 10:00 am. SMALL ANIMAL – NEUROLOGY 9:00 am N‐1 Bruno Torres Neuroprotective Effects of Dantrolene and Mesenchymal Stem Cells on Acute Spinal Cord Injury 9:15 am N‐2 G. Diane Shelton Congenital Myasthenic Syndrome in Jack Russell Terriers Is Caused By a Frameshift Mutation in Chrne 9:30 am N‐3 Kathryn Winger Prognostic Indicators for the Functional Outcome of Deep Pain Negative Dachshunds 9:45 am N‐4 Laura Johnstone Retrospective Evaluation of Horses Diagnosed with Lyme Neuroborreliosis on Postmortem Examination: 13 Cases (2004‐2014) BREAK 10:30 am N‐5 Maria Ines de Freitas Brain Herniation in 113 Cats: A Descriptive Study 10:45 am N‐6 Jill Hicks Evaluation of Temozolomide and Gadolinium Conjugated PLGA Microcylinders in the Canine Brain 11:00 am N‐7 Guillaume Leblond Biomechanical and Radiological Comparison Between Transarticular Screw Fixation and 2 Multi‐Implant Constructs for Ventral Atlantoaxial Stabilization in the Dog 11:15 am N‐8 Nicholas Archambault Computed Tomography Guided Sterotaxic Navigation of the Canine Thoracolumbar Spine 11:30 am N‐9 Jennifer Michaels Pharmacokinetics of Single Dose Rectal Zonisamide in Healthy Dogs 11:45 am N‐10 Andy Shores Fenestration of the Lateral Ventricle in the Management of Obstructive Hydrocephalus in Young Small Animals 12:00 pm N‐11 Liana Granum Clinical Characteristics and Outcome of Cases Diagnosed with Electrographic Seizure in 73 Dogs and 13 Cats 12:15 pm N‐12 Thomas Cave Utility of a Clinical Scoring System and Ocular Pharmacological Tests for Ante‐Mortem Diagnosis of Feline Dysautonomia BREAK 2:15 pm N‐13 Rachel Song Utility of Von Frey Anesthesiometry to Assess Recovery of Sensory Function After Thoracolumbar Intervertebral Disc Extrusion in Dogs 2:30 pm N‐14 Natalya Ulanova Endoscopic Laminectomy Technique in Dogs with Hansen Type I Disc Disease 2:45 pm N‐15 Elizabeth Parsley An Internet Survey of the Approach to Neurosurgical Emergencies By ACVECC Diplomates 3:00 pm N‐16 Bernardo Martins Neuroprotective Effects of Riluzole and Dantrole Association on Acute Spinal Cord Injury SMALL ANIMAL – ONCOLOGY 9:00 am O‐1 Rachel Rasmussen Phase I Lead‐in and Subsequent Randomized Trial Assessing Safety and Modulation of Regulatory T Cell Numbers Following a Maximally Tolerated Dose Doxorubicin and Metronomic Dose Cyclophosphamide Combination Chemotherapy Protocol in Tumor‐Bearing Dogs (VCS Award Winner) 9:15 am O‐2 Tonya Gustafson Inhibition of Hedgehog Signaling Inhibits Proliferation in Canine Transitional Cell Carcinoma (VCS Award Winner) 9:30 am O‐3 Michelle Beehler Adjuvant Alpha‐Tocopherol & Improved Anticancer Efficacy in an Immune Competent Pre‐Clinical Model 9:45 am O‐4 Francisco Clemente‐Vicario Metronomic Antiangiogenic Therapy After Amputation and Suramin/Doxorubicin Chemotherapy Does Not Improve the Disease Free Interval or Survival in Retired Racing Greyhounds with Osteosarcoma BREAK 10:30 am O‐5 Alycen Lundberg Investigating NQOL As a Radiation Inducible Enzyme in Osteotropic Neoplasms and Targeted Therapy with the Novel Drug Deoxynyboquinone 10:45 am O‐6 Carlos Rodriguez Jr. A Prospective, Open‐Label Study Evaluating Treatment of Canine B‐Cell Lymphoma with L‐Asparaginase, Doxorubicin and a Canine Anti‐CD20 Monoclonal Antibody 11:00 am O‐7 Emily Rout Characterization of Immunoglobulin Gene Use and Mutation Status in Canine B Cell Chronic Lymphocytic Leukemia 11:15 am O‐8 Masato Kobayashi Molecular Changes Associated with the Development of Resistance to Imatinib in an Imatinib‐Sensitive Canine Neoplastic Mast Cell Line Carrying a Kit C.1523A>T Mutation 11:30 am O‐9 Crystal Garnett Fixed Dose Rate Gemcitabine in the Cat 11:45 am O‐10 Charlotte Johnston Pharmacokinetics and Bioavailability of Metformin in Dogs 12:00 pm O‐11 Saaya Hiyoshi Genescan Analysis for the Differentiation of Clonal Origin of Cells in Dogs That Developed Two Different Types of Lymphoid Malignancies SMALL ANIMAL ‐ GASTROENTEROLOGY 9:00 am GI‐1 Romy Heilmann Effect of Gastrointestinal Parasites and Viral Agents on Fecal S100A12 Concentrations in Puppies 9:15 am GI‐2 Hiroshi Ohta Prolonged Administration of Prednisolone Alone or in Combination with Azathioprine Alters Serum Canine Pancreatic Lipase Immunoreactivity (Spec cPLTM) in Dogs 9:30 am GI‐3 Blake Guard The Effect of the Probiotic Sivoytm on Fecal Microbial Composition in Healthy Dogs 9:45 am GI‐4 Emily Gould Quantitative and Qualitative Identification of Feline Tritrichomonas Foetus Surface Antigens: Putative Targets for Diagnosis and Treatment BREAK 10:30 am GI‐5 Jenessa Winston Cytopathological Findings of Fecal or Rectal Cytology Samples From Dogs Identified with or without the Presence of Cyniclomyces guttulatus 10:45 am GI‐6 Stefanie DeMonaco Natural History of Biliary Sludge in Dogs 11:00 am GI‐7 Alyssa Sullivant Identification of Histamine Receptors in the Canine Gastrointestinal Tract 11:15 am GI‐8 JB Honneffer Alterations in Fecal Metabolite Profiles of Dogs with Chronic Enteropathy 11:30 am GI‐9 Rosana Lopes Targeted Serum Amino Acid Analysis of Hypocobalaminemic Dogs with Decreased or Increased Serum Folate Concentrations 11:45 am GI‐10 Rosana Lopes Serum Amino Acid Analysis in Hypocobalaminemic Dogs with Exocrine Pancreatic Insufficiency 12:00 pm GI‐11 Sergi Segarra Effects of an Oral Supplement Based on Chondroitin Sulfate, Resistant Starch, and Prebiotics in Dogs with IBD 12:15 pm GI‐12 Jacqueline Whittemore Gastrointestinal Endoscopic Mucosal Lesion Scores in Healthy Dogs Receiving Prednisone, Aspirin, and Omeprazole BREAK 2:15 pm GI‐13 Cecily Bonadio Characterization of Canine Dysphagia in a Veterinary University Hospital 2:30 pm GI‐14 Talia Guttin Ability of Ultrasound to Predict the Presence and Location of Histologic Lesions in the Small Intestine of Cats 2:45 pm GI‐15 Paola Gianella Prognostic Factors for Short‐ and Long‐Term Survival in Dogs with Protein‐Losing Enteropathy 3:00 pm GI‐16 Jennifer Sinclair Evaluating Quality and Adequacy of Gastrointestinal Samples with Reusable and Disposable Forceps 3:15 pm GI‐17 Kelli Bogard Double‐Blind, Placebo‐Controlled Ultrasonographic Evaluation of the Effects of Antiemetic Drugs on Antral Motility and Gastric Emptying in Healthy Dogs 3:30 pm GI‐18 Erin Olson Evaluation of the Effects of a 2‐Week Treatment with Metronidazole on the Fecal Microbiome of Healthy Dogs 3:45 pm GI‐19 Mellora Sharman Gastrokine mRNA Expression Is Up‐Regulated in Gastric Tissue From Dogs with Helicobacter Colinization but without Inflammatory Change BREAK 4:30 pm GI‐20 Yasushi Minamoto Fecal Short‐Chain Fatty Acid Concentrations in Dogs with Chronic Enteropathy 4:45 pm GI‐21 Tomomi Minamoto Assessment of Serum Lipoprotein Profiles in Dogs with Chronic Enteropathy and Healthy Control Dogs 5:00 pm GI‐22 Niels Grützner Serum Homocysteine Concentrations in Greyhounds 5:15 pm GI‐23 Stanley Marks Effect of the Probiotic Enterococcus faecium SF 68 on Presence of Diarrhea in Weanling Kittens 5:30 pm GI‐24 Joseph Parambeth Fecal N‐Methylhistamine Concentrations in Dogs with Exocrine Pancreatic Insufficiency 5:45 pm GI‐25 Giacomo Rossi The Effect of the Probiotic Sivoy™ on Clinical and Histopathological Parameters in Cats  with  Chronic Idiopatic Constipation and Megacolon 6:00 pm GI‐26 Tim Kretzschmar Effects of a Synthetic Serine Protease Inhibitor, Camostat Mesilate (FOY‐305), on Markers of Pancreatic Acinar Cell Damage, Inflammation, and Fibrosis in Dogs with Suspected Naturally Occuring Chronic Pancreatitis SMALL ANIMAL – IMMUNOLOGY 9:00 am IM‐1 Megan Grobman Thymidine Kinase, C‐Reactive Protein and Vitamin D Concentrations in Dogs with Immune‐Mediated Thrombocytopenia and Immune‐Mediated Hemolytic Anemia 9:15 am IM‐2 Lara Barron Comparison of Proliferative and Immunomodulatory Potential of Adipose‐Derived Mesenchymal Stem Cells From Young and Old Cats 9:30 am IM‐3 Steven Friedenberg Association of a Variant Allele in the Dog Leukocyte Antigen Class I Gene DLA‐79 with Multiple Canine Autoimmune Diseases SMALL ANIMAL – HEMATOLOGY 10:30 am HM‐1 Shannon Westgarth The Individual and Combined Effects of Long‐Chain N‐3 Polyunsaturated Fatty Acids and Low‐Dose Acetylsalicylic Acid on Platelet Function in Healthy Dogs 10:45 am HM‐2 Unity Jeffery Dogs Cast Nets Too: Neutrophil Extracellular Traps in Health and Immune‐Mediated Hemolytic Anemia 11:00 am HM‐3 Michelle Foote Effects of Leukoreduction on Coagulation Factors in Canine Fresh Frozen Plasma 11:15 am HM‐4 Stéphanie Goulet Production and Caracterization of Anti‐Dal Antibodies Following Sensitization of a Dal‐Negative Dog 11:30 am HM‐5 Samantha Muro Effects of Leukoreduction and Storage on Phosphatidylserine Expression on Canine Packed Red Blood Cells 11:45 am HM‐6 AM Lynch Pre‐Operative Evaluation of Hemostatic Biomarkers in Stable Dogs Undergoing Splenectomy for Splenic Masses SMALL ANIMAL ‐ INFECTIOUS DISEASE 2:15 pm ID‐1 Edward Breitschwerdt Canine Vector‐Borne Disease Prevalence in Dogs From the Southeastern United States Diagnosed with Immune‐Mediated Hemolytic Anemia and/or Immune‐Mediated Thrombocytopenia 2:30 pm ID‐2 Nancy Sanders Evaluation of the FIP Virus Realpcr® Test and the Impact of Sample Selection on the Detection of the Virulent Strain of Feline Coronavirus in FIP Suspect Cats 2:45 pm ID‐3 Michael Lappin Anaplasma phagocytophilum and Borrelia burgdorferi Infections in Cats Exposed Repeatedly to Ixodes scapularis 3:00 pm ID‐4 Michael Lappin Comparison of Results of 2 Assays Detecting Dirofilaria immitis Antigen and Ehrlichia canis Antibodies Using Samples From High and Low Risk Regions 3:15 pm ID‐5 Michael Lappin Attempted Transmission of Bartonella henselae to Cats By Ingestion of Infected Ctenocephalides felis 3:30 pm ID‐6 Erin Burton Multiple Drug Resistance in Canine Urinary Tract Pathogens 3:45 pm ID‐7 Filippo Ferri Splenitis in Dogs: A Review of 33 Cases (2005‐2013) BREAK 4:30 pm ID‐8 Jenessa Winston Cyniclomyces guttulatus From Domestic Dogs and Laboratory Rabbits Have Distinct Genotypes 4:45 pm ID‐9 Madeline Anna Comparison of 2 Commercially Available PCR Assays for the Amplification of Ehrlichia Spp. DNA From Blood of Naturally Exposed Dogs in Oklahoma 5:00 pm ID‐10 Jennifer Thomas Infection of Babesia conradae in Hunting Greyhounds From Oklahoma 5:15 pm ID‐11 Christian Leutenegger Cross‐Species Transmission of Canine Parvovirus 2 to Healthy Shelter Kittens 5:30 pm ID‐12 Diana Scorpio Serologic and Molecular Surveillance to Predict Prevalence of Canine Vector‐Borne Diseases on the Island of St. Kitts 5:45 pm ID‐13 JE Slovak Surveillance of Feral Cats in Central Iowa for Subclinical Infection with Cytauxzoon felis 6:00 pm ID‐14 John Prescott A Novel Pore‐Forming Toxin in Type A Clostridium perfringens Associated with Fatal Canine Hemorrhagic Gastroenteritis and Neonatal Foal Necrotizing Enterocolitis 6:15 pm ID‐15 Stacie Summers Effect of Parenteral Administration of Modified Live or Inactivated Feline Herpresvirus 1, Calicivirus, and Panleukopenia Vaccines on Clinical Signs in a Day 7 Feline Herpesvirus 1 Challenge Model SMALL ANIMAL ‐ NUTRITION/METABOLISM 9:00 am NM‐1 Tsz Hong Law The Effects of a Medium Chain Triglycerides Diet Trial on Dogs with Idiopathic Epilepsy 9:15 am NM‐2 Laura Tonkin Evaluation of Serum Fatty Acids in Cats with Chronic Kidney Disease 9:30 am NM‐3 Yuanlong Pan Effects of Protein: Carbohydrate Ratios on Weight Loss in Overweight Cats 9:45 am NM‐4 Patrick Robbins Predictors of Core Temperature During Interval Exercise in Dogs BREAK 10:30 am NM‐5 Linda Rhodes Safety of the Ghrelin Agonist, Capromorelin, Administered Daily to Beagle Dogs for One Year 10:45 am NM‐6 Reut Tenne Palatability and Clinical Effects of an Oral Recuperation Fluid During the Recovery of Dogs with Parvoviral Enteritis SMALL ANIMAL ‐ PHARMACOLOGY 11:15 am P‐1 Claire Fellman Effects of Aspirin and Cyclosporine on Canine T‐Cell Cytokines and Regulatory T‐Cells 11:30 am P‐2 Esther Bijsmans Measurement of Plasma Amlodipine Concentration Using LC/MS/MS at the Point of Blood Pressure Control in Hypertensive Cats SMALL ANIMAL ‐ ENDOCRINOLOGY 2:15 pm EN‐1 Christopher Scudder The Feline AIP Gene: the Key to Hypersomatotropism Tumorigenesis? (ESVE Award Winner) 2:30 pm EN‐2 Mark Peterson Evaluation of Serum Thyroid Stimulating Hormone Concentration As a Diagnostic Test for Hyperthyroidism in Cats 2:45 pm EN‐3 John Lucy Efficacy of Low‐Dose (2 Millicurie) Versus Standard‐Dose (4 Millicurie) Radioiodine (131) Treatment for Cats with Mild‐to‐Moderate Hyperthyroidism 3:00 pm EN‐4 Heather Vaske Assessment of Renal Function in Hyperthyroid Cats Managed with Hill's® Prescription Diet® Y/D® Feline 3:15 pm EN‐5 Lisa Gorman Increases in Serum Beta Hydroxybutyrate in Cats with Chronic Kidney Disease, Hyperthyroidism, and Hepatic Lipidosis 3:30 pm EN‐6 Tim Williams Investigation of Dysproteinemia in Hyperthyroid Cats As a Biomarker of Concurrent Chronic Kidney Disease (CKD) 3:45 pm EN‐7 Cynthia Ward Efficacy of Prozincr Insulin in Naïve and Insulin‐Established Cats Using Continuous Interstitial Glucose Monitoring BREAK 4:30 pm EN‐8 Susan Gottlieb Metabolic Profiling of Diabetic Cats in Remission 4:45 pm EN‐9 Margaret Scuderi Safety and Efficacy Assessment of a GLP‐1: Glargine Combination for Feline Diabetes 5:00 pm EN‐10 Bianca Lourenço Diagnosis of Canine Hyperadrenocorticism Is Associated with Gender, Age, Breed, and Comorbid Conditions 5:15 pm EN‐11 Chen Gilor A Novel Protocol for Glucagon Stimulation Test in Cats Using Recombinant DNA Glucagon 5:30 pm EN‐12 Chen Gilor Day‐to‐Day Variability of Serial Blood Glucose Curves in Diabetic Dogs Treated with Twice Daily NPH or Insulin Detemir 5:45 pm EN‐13 Ellen Behrend Effect of Glucosuria on Urine Specific Gravity (USG) in Dogs 6:00 pm EN‐14 Laura Cosgrove A Novel Cortisol Based Method for Monitoring Trilostane Therapy in Dogs with Hyperadrenocorticism 6:15 pm EN‐15 Danielle Davignon Effect of Non‐Thyroidal Illness on Serum Concentrations of T4, Free T4 and Thyroid Stimulating Hormone in Cats SMALL ANIMAL ‐ NEPHROLOGY/UROLOGY a 9:00 am NU‐1 Jewel Milo A New Method for Measuring Glomerular Fitlration Rate (GFR) in Cats with Comparison to Two Standard Methods 9:15 am NU‐2 Chloe Wormser Perioperative Complications, Mortality, and Long‐Term Outcome of Ureteral Surgery and Ureteral Stenting in Cats: 117 Cases 9:30 am NU‐3 Jonathan Foster Effect of Dialysate Sodium Concentration on Serum Sodium Concentration During Intermittent Hemodialysis in Dogs 9:45 am NU‐4 Jack Lawson Urinary Active TGF‐ Β1 in Feline Chronic Kidney Disease BREAK 10:30 am NU‐5 Alicia Dudley Platelet Function in Dogs with Naturally Occurring Chronic Kidney Disease 10:45 am NU‐6 Ji Hye Hwang Evaluation of Effectiveness and Possibility of Nafamostat Mesilate As an Alternative Anticoagulant During Intermittent Hemodialysis on Healthy Beagle Dogs 11:00 am NU‐7 Michael Wood Modifying Urine Glycosaminoglycan Concentrations By Intramuscular Injection of Polysulfated Glycosaminoglycans in Dogs 11:15 am NU‐8 Yann Queau Effects of Gradual Increase of Dietary Sodium Chloride on Urinary Parameters and Relative Supersaturation in Healthy Cats 11:30 am NU‐9 Cinthia Martorelli Serum Magnesium Concentration in Dogs with Naturally Occurring Chronic Kidney Disease 11:45 am NU‐10 William Culp A Comparison of Perioperative Outcome in Cats Undergoing Ureterotomy or Ureteral Stenting for the Treatment of Benign Ureteral Obstructions 12:00 pm NU‐11 Julie Byron Signalment and Diagnosis of 205 Neutered Female Dogs with Urinary Incontinence in the United States a Also See Small Animal Nephrology/Urology abstracts NU‐12 ‐ NU‐15, Friday, June 5, 9:00 am ‐ 10:00 am. SMALL ANIMAL – OTHER 2:15 pm OT‐1 Romy Heilmann Evaluation of Serum Alpha 1‐Proteinase Inhibitor Concentrations in Dogs with Systemic Inflammatory Response Syndrome or Sepsis 2:30 pm OT‐2 Stuart Walton In Vitro Determination of the Utility of Therapeutic Plasma Exchange (TPE) in the Treatment of Ibuprofen Toxicity 2:45 pm OT‐3 Jeffrey Mitchell Prognostic Indicators in Dogs with Septic Peritonitis: 54 Cases (2004‐2014) 3:00 pm OT‐4 Keriann Cabral Does Do Not Resuscitate (DNR) Always Mean DNR? Exploring DNR Orders in Veterinary Medicine 3:15 pm OT‐5 Kevin Le Boedec Impact of Statistical Methods on Analyte Reference Interval Accuracy 3:30 pm OT‐6 Inigo Sanz Acute Phase Reaction  & Cardiac Biomarkers in Feline Non‐Cardiac Diseases EQUINE a 9:00 am E‐1 Natalie Heliczer Accuracy of Non‐Invasive Oscillometric Blood Pressure in Normo‐ Hyper and Hypotensive Adult Standing Horses 9:15 am E‐2 Tiago Afonso Attenuation of the Blood Pressure Response to Exogenous Angiotensin I After Oral Administration of Benazepril to Healthy Adult Horses 9:30 am E‐3 Ellen de Graaf‐Roelfsema Effect of Diet‐Induced Weight‐Gain on Oral Glucose Tolerance and Insulin Response in Shetland Ponies 9:45 am E‐4 Concetta Amato Hormonal Changes: Predictive Factors of Chronic Fatigue in Show Horses BREAK 10:30 am E‐5 Katarzyna Dembek Multiple Adrenocortical Steroid and Steroid Precursores Response to ACTH Stimulation Test in Hospitalized Foals 10:45 am E‐6 Nicole Arana‐Valencia The Effect of Ph on Large Intestinal Ion Transport ad Transepithelial Resistance in Suspected Insulin Resistance Horses 11:00 am E‐7 Steven Grubbs Epidemiological Characteristics of Horses with PPID at Initial Diagnosis 11:15 am E‐8 Melissa Restifo Effects of Fasting and the Oral Sugar Test on Thyrotropin‐Releasing Hormone Stimulation Test Results in Horses 11:30 am E‐9 Laura Dunbar Evaluation of Agreement of Four Common Diagnostic Tests for Insulin Resistance in Adult Light‐Breed Horses 11:45 am E‐10 Ahmed Kamr Vitamin D Metabolites and Their  Association with Severity of Illness and Mortality in Hospitalized Foals 12:00 pm E‐11 Sarah Jacob Use of Metabolomic Approaches for Insight Into Metabolic Perturbations in Welsh Ponies with Insulin Dysregulation 12:15 pm E‐12 Emil Olsen Can Motion Capture Detect Ataxia in Horses? BREAK 2:15 pm E‐13 Eleonora Po Intravenous Plasma Administration Evokes a Balanced Systemic Inflammatory Response in Healthy Newborn Foals 2:30 pm E‐14 Rachel Liepman A Comprehensive Evaluation of Traditional and Novel IgG Cut‐Off Values and Their Association with Mortality in Neonatal Foals 2:45 pm E‐15 Katherine Delph Comparison of Immunologic Responses Following Intranasal and Oral Vaccination with a USDA Approved, Live‐Attenuated Streptococcus equi Vaccine 3:00 pm E‐16 Sian Durward‐Akhurst Major Histocompatibility Complex Expression in Muscle of Horses with Immune Mediated Myositis 3:15 pm E‐17 Gaelle Hirsch Characterization of the Respiratory Tract Microbiota of Healthy Horses 3:30 pm E‐18 Daniela Bedenice The Relationship Between Serum 25‐Hydroxy Vitamin D Levels and Inflammatory Airway Disease (IAD) in Horses 3:45 pm E‐19 Sandra Taylor Septic Pleuropneumonia in Horses BREAK 4:30 pm E‐20 Gayle Hallowell Factors Affecting Complication Rates with Subpalpebral Lavage Catheter Use in Horses 4:45 pm E‐21 Igor Canisso Estradiol 17β and Estrone Sulfate in Mares with Experimentally Induced Ascending Placentitis 5:00 pm E‐22 Frank Andrews IGM and IgG Responses in Horses and Pony Foals After Vaccination for West Nile Virus and Eastern Equine Encephalitis 5:15 pm E‐23 Brandy Burgess Factors Associated with Equine Shedding of Multi‐Drug Resistant Salmonella enterica and its Impact on Health Outcomes 5:30 pm E‐24 Sarah Shaw Estimating the Sensitivity and Specificity of Real‐Time Quantitative PCR of Fecal Samples for Diagnosis of Rhodococcus equi Pneumonia in Foals 5:45 pm E‐25 Laurie Beard Bayesian Geostatistical Analysis and Environmental Risk Factors of Pigeon Fever 6:00 pm E‐26 Sara Higgins Risk Factors for Equine Infectious Anemia in Canada a Also See Equine E‐27 ‐ E‐34, Friday, June 5, 8:00 am ‐ 10:00 am. FOOD ANIMAL 9:00 am F‐1 Julie Berman Lung Consolidation in Veal Calves Metaphylactically Treated with Tildipirosin: Predictive Factors and Impact on Weight Gain 9:15 am F‐2 AbdElMonem AbdAllah Systematic Review of the Diagnostic Accuracy of Haptoglobin, Serum Amyloid A and Fibrinogen Versus Clinical Reference Standard for the Diagnosis of Bovine Respiratory Disease (BRD) 9:30 am F‐3 Danielle Doyle Agreement Between Sample Collection Methods Used for the Diagnosis of Bovine Respiratory Disease in Dairy Calves 9:45 am F‐4 Katharine Simpson Frontal Sinusitis in Adult Beef Bulls: A Retrospective Analysis of 18 Cases (1999‐2014) BREAK 10:30 am F‐5 Jane Woodrow Comparative Plasma Pharmacokinetics of Ceftiofur Sodium and Ceftiofur Crystalline Free Acid in Neonatal Calves 10:45 am F‐6 Ameer Megahed Changes in Plasma Calcium Concentration in Periparturient Dairy Cattle Fed an Acidogenic Diet in the Late Dry Period 11:00 am F‐7 Ameer Megahed Evaluation of the Precision Xtra® Hand‐Held Meter for the Rapid Point of Care Measurement of Blood and Plasma Glucose Concentration in Periparturient Dairy Cattle 11:15 am F‐8 Sarah Raabis Health Benefits of IL‐10 Egg Yolk Antibodies Administered to Milk‐Fed Dairy Calves 11:30 am F‐9 Emily Barrell Presumptive Fenbendazole Toxicosis in Three Holstein Calves 11:45 am F‐10 Pamela Adkins The Relationship Between Body Site Colonization and Intramammary Infection with Non‐aureus Staphylococcal Species in Dairy Heifers 12:00 pm F‐11 Philippa Gibbons Visual Estimation of Goat Liveweight By Veterinary and Animal Science Students BREAK 5:30 pm F‐12 Thibaud Kuca Chronic (Vagal) Indigestion in Adult Dairy Cattle: A Retrospective Study of 52 Cases (2004‐2013) 5:45 pm F‐13 Diego Gomez Evaluation of Endotoxin in Plasma of Diarrheic Calves and Its Association with Demeanor, Hyper‐L‐Lactatemia and Hypoglycemia 6:00 pm F‐14 Eugene Nwaokorie Epidemiologic Evaluation of Silica Urolithiasis in Goats (106 Cases): Perspective  From the Minnesota Urolith Center Oral Presentations – Friday, June 5 Bolded type indicates ACVIM Resident Research Award eligibility. Time # Presenting Author Abstract Title SMALL ANIMAL – CARDIOLOGY 8:00 am C‐18 Christian Weder Multi‐Dose Pharmacokinetics and Pharmacodynamics of the Commercially Available Formulation of Oral Apixaban in Cats: A Pilot Study 8:15 am C‐19 Claudia Vollmar Long Term Outcome of Irish Wolfhounds with Lone Atrial Fibrillation 8:30 am C‐20 Claudia Vollmar Irish Wolfhounds with Dilated Cardiomyopathy: Causes of Death 8:45 am C‐21 Jorge Prieto‐Ramos Echocardiographic Findings in an Aged Population of Cavalier King Charles Spaniels 9:00 am C‐22 Lois Wilkie Comparison of Methods for Measuring Left Ventricular Wall Thickness and Left Atrial Size in Cats 9:15 am C‐23 Veronica Strömberg iPhone ECG: A New Simple Method to Detect Hypertrophic Cardiomyopathy (HCM) in Cats? 9:30 am C‐24 Lauren Wiley Morphology of Tricuspid Valve Apparatus in Feline Hypertropic and Dilated Cardiomyopathy 9:45 am C‐25 Marisa Ames The Urine Aldosterone to Creatinine Ratio (UALDO:C) Determined By Radioimmunoassay (RIA) in Normal Dogs Greater Than 5 Years of Age SMALL ANIMAL – HEPATOLOGY 8:00 am HP‐1 Sara Wennogle Measurement of Plasma Fibrinogen in Dogs with Hepatobiliary Disease 8:15 am HP‐2 Alexandra Hamilton Serum Markers of Vitamin B Metabolism in Cats with Hepatic Lipidosis 8:30 am HP‐3 Medora Pashmakova Association Between Biliary Cytology and Microbiology in Dogs and Cats: 52 Cases (2004‐2014) 8:45 am HP‐4 Allison Bradley Histologic Scoring and Clinical Outcomes of Hepatitis in 42 Standard Poodles 9:00 am HP‐5 Jody Gookin Possible Association of Gallbladder Mucocele Histologic Diagnosis with Selected Preventative Drug Use in Dogs: A Matched Case‐Controlled Study 9:15 am HP‐6 Alexis Lecoindre A New Combination of Blood Parameters for Accurate Non‐Invasive Diagnosis of Liver Fibrosis in Dogs SMALL ANIMAL ‐ RESPIRATORY 8:00 am R‐1 Megan Grobman Maropitant (Cerenia®) Ameliorates Clinical Signs but Not Inflammation Associated with Canine Chronic Bronchitis 8:15 am R‐2 Michal Crha Comparison of Fatigue Characteristics of Two Commercially Available Veterinary Tracheal Stents 8:30 am R‐3 Kevin Le Boedec Meta‐Analysis of the Association Between Mycoplasma Species and Feline Respiratory Tract Diseases SMALL ANIMAL ‐ NEPHROLOGY/UROLOGY 8:45 am NU‐12 Crystal Cooley Consumption of Viyo Recuperation Renal – Advanced Formula in Cats with Stable Chronic Kidney Disease and its Effects on Biochemical Parameters 9:00 am NU‐13 Jack Quinn Effect of Blood Contamination in Canine Canine Urine on The Performance of a Commercial Immunoassay for the Acute Kidney Injury (AKI) Marker: Urinary Clusterin (uCLUS) 9:15 am NU‐14 Aida Vientós‐Plotts Effect of Hematuria on Urine Protein:Creatinine Ratio in Cats 9:30 am NU‐15 Carrie Palm Acute Evaluation of Urethral Thermoplasty in Normal Dogs EQUINE** 8:00 am E‐27 Patricia Dowling Depletion of Phenylbutazone Residues From Equine Urine and Tissues 8:15 am E‐28 Amy Stieler Effect of Rifampin on Erythromycin‐Induced Anhidrosis in Foals 8:30 am E‐29 Rosemary Cuming Ex Vivo Equine Subconjucntival Injection Characteristics and In Vitro Release Profiles of Voriconazole From a PLGA‐PEG‐PLGA Thermogel 8:45 am E‐30 Joy Tomlinson Post‐Transfusion Survival Time of Cross‐Match Incompatible Red Blood Cells in Adult Horses 9:00 am E‐31 Rachel Liepman Temporal Effects of Intravenous Antibiotics on the Equine Fecal Microbiome 9:15 am E‐32 Emilie Setlakwe Prospective Evaluation of the Accuracy of the MFlaSh Ultrasound Protocol and Transrectal Abdominal Palpation for Predicting Specific Surgical Diagnoses in Cases of Equine Acute Colic 9:30 am E‐33 Sarah Pedersen Prevalence of and Risk Factors for Gastric Ulceration in Showjumping Warmbloods 9:45 am E‐34 Angelika Schoster The Effect of Transport, Fasting and Anesthesia on the Composition of the Equine Microbiota Poster Presentations On Display: Thursday, June 4, 7:00 am – 5:00 pm Friday, June 5, 7:00 am – 5:00 pm Attended by ALL Authors Eligible for ACVIM Resident Research Awards: Thursday, June 4, 9:50 am – 10:30 am Friday, June 5, 9:50 am – 10:30 am Attended by ALL Authors – Wine & Cheese Reception: Friday, June 5, 6:00 pm – 7:30 pm Bolded type indicates ACVIM Resident Research Award eligibility. # Presenting Author Abstract Title SMALL ANIMAL – CARDIOLOGY C‐26 Jonathan Lichtenberger Preliminary Assessment of a Novel 14‐Day Ambulatory ECG Monitor (Zio® Patch) in Dogs C‐27 Jonathan Lichtenberger Cardiac Biomarker Profiles in Dogs with Naturally‐Occurring Pre‐Capillary Pulmonary Hypertension C‐28 Masami Uechi Detection of Periodontopathic Bacteria in Dogs with Mitral Regurgitaion C‐29 Kathryn Meurs Cardiac Regenerative Potential of Cardiosphere‐Derived Cells From Adult Dog Hearts C‐30 Aparecido Camacho Correlation Between Echocardiographic Parameters and IL‐1β, IL‐6, IL‐10, TNF‐Α and C‐Reactive Protein in Bitches with Sepsis C‐31 Caio Nogueira Duarte Assessment of QTC Interval and QT Dispersion in Dogs with Pyometra C‐32 Aparecido Camacho Subclassification of Stage 2 in Congestive Heart Failure in Dogs with Myxomatous Mitral Valve Disease C‐33 Aparecido Camacho Heart Rate Variability in Dogs with Myxomatous Mitral Valve Disease Undergoing Treadmill Exercise C‐34 Lisa Freeman Plasma Omega‐3 Fatty Acid (Eicosapentaenoic Acid) Concentrations Are Related to Survival in Cats with Hypertrophic Cardiomyopathy C‐35 Wonjung Kim Clinical Efficacy and Safety of Imidapril in Dogs with Congestive Heart Failure C‐36 Yunhye Kim Evaluation of Left Ventricular Function in Dogs with Different Stages of Myxomatous Mitral Valve Disease: Tissue Doppler and Strain Imaging Study C‐37 Dianne Mawby Correlation Between Radial and Coccygeal Artery Indirect Doppler Blood Pressure Measurements in Cats C‐38 Giulio Menciotti Three‐Dimensional Echocardiographic Assessment of Normal Canine Mitral Valve C‐39 Arine Pellegrino Genetic Evaluation of Persian Cats with Hypertrophic Cardiomiopathy: Screening of Mutations at ACTC1 and MYBPC3 Genes C‐40 Arine Pellegrino Cardiovascular Evaluation of Persians Cats with Polycystic Kidney Disease C‐41 Fernando Rosa Echocardiographic Assessment of the Effects of Carvedilol, Cyclosporin‐A and Sildenafil Citrate on Doxorubicin‐Induced Cardiomyopathy in an Experimental Model with Rabbits C‐42 Fernando Rosa Left Ventricular Circumferential Strain and Strain Rate By Two‐Dimensional Imaging Velocity Vector Imaging in Healthy Beagles Dogs C‐43 Elizabeth Rozanski The Effect of Catheter Location During Simulated CPR on the Time for a Peripheral Injection to Reach the Heart C‐44 Inigo Sanz Cardiac Biomarkers and Acute Phase Proteins in Feline Cardiac Disease C‐45 Ingrid van Hoek Does Size Matter? A Significant Relation Between Head Width and Left Ventricular Free Wall Growth Rates in Young Cats SMALL ANIMAL – NEUROLOGY N‐17 Rennie Waldron Comparison of Clinical Outcome with Zonisamide, Levetiracetam or Phenobarbital Monotherapy in Dogs with Idiopathic Epilepsy N‐18 Kathryn Bray Efficacy of Midazolam Continuous Rate Infusion for Emergent Seizure Therapy N‐19 Kathryn Bray MRI Characteristics of Atlanto‐Axial Subluxation in 42 Dogs N‐20 Guillaume Leblond Determination of Ideal Implant Placement in Canine Atlantoaxial Vertebrae – Description of a New Method Using Osirix N‐21 Akinori Nomura Clinical Efficacy and Safety of Zonisamide Monotherapy in Dogs with Newly Diagnosed Idiopathic Epilepsy N‐22 Mario Dolera Canine Meningioma: Comparison of Palliative Therapy, Surgery and Stereotactic Radiosurgery N‐23 Mario Dolera Suspected Hydrated Nucleus Pulposus Extrusions Associated with the Dorsal Longitudinal Ligament in Dogs N‐24 Luca Malfassi The Use of Magnetic Resonance Spectroscopy in Differential Diagnosis of Canine Brain Diseases N‐25 Go Togawa Development of F‐Response Method Using Multiple‐Point Stimulation for Motor Unit Number Estimation Technique in Normal Dogs Small Animal – Oncology O‐12 Butch KuKanich Analysis of Lomustine Content in FDA Approved and Compounded Lomustine Capsules O‐13 Silvia Lucas Expression of Multiple Drug Resistance Markers ABCB1 and LRP  in Cats with Alimentary Lymphoma O‐14 Valter Winkel Metallothionein and KI‐67 in Cats with Alimentary Lymphoma O‐15 Sabrina Rodigheri Response of Il‐2 and Il‐6 After Unilateral Total Mastectomy Alone or in Association with Ovariosalpingohisterectomy in Bitches with Mammary Tumors O‐16 Sabrina Rodigheri Stress Response to Surgical Trauma of Unilateral Total Mastectomy Alone or in Association with Ovariosalpingohisterectomy in Bitches with Mammary Tumors O‐17 Jeanne Larson Pharmacokinetics of Orally Administered Low‐Dose Rapamycin in Healthy Dogs (VCS Award Winner) O‐18 Christine Mullin Minoxidil Inhibits the Metastatic Phenotype of Osteosarcoma (VCS Award Winner) SMALL ANIMAL – ENDOCRINOLOGY EN‐16 Andrew Woolcock Evaluation of the Use of Baseline Cortisol to Monitor Twice‐Daily Trilostane Therapy in Dogs with Spontaneous Hyperadrenocorticism EN‐17 Anthony Carr Blood Pressure in Hyperthyroid Cats in Comparison to Age Matched Controls Using High Definition Oscillometry EN‐18 Viviani De Marco Epidemiologic Study of Adrenal Tumors in Dogs: 49 Cases (2012‐2014) EN‐19 Eric Januário Association of Canine Acute Pancreatitis with Hypertriglyceridemia and Adrenomegaly: A Retrospective Study (2012‐2014): 91 Cases EN‐20 Hakhyun Kim Association Between Endocrinopathies and Hyperlipidemia in 104 Dogs with Acute Pancreatitis: A Retrospective Study EN‐21 Hye‐Sun Kim Serum Adipokine Concentrations in Dogs with Myxomatous Mitral Valve Disease EN‐22 Melania Osto Evaluation of the Isolation and Functionality of Pancreatic Islets in Cats EN‐23 Bo Hye Shin Effect of Administration with Mirtazapine on the Hypothalamic‐Pituitary‐Adrenocortical Axis in Healthy Beagle Dogs SMALL ANIMAL – GASTROENTEROLOGY GI‐27 Frances Hurrell Beta‐Hydroxybutyrate Concentrations in Dogs with Acute Pancreatitis and without Diabetes Mellitus GI‐28 Allyson Berent Endoscopic Retrograde Cholangiography and Biliary Stent Placement for Biliary Tract Obstruction in Dogs and Cats GI‐29 Sarah Cooper Effect of Bifidobacterium animalis AHC7 and Tylosin on the Microbiome During the Quarantine Period of Military Working Dogs GI‐30 Niels Grützner Association of Two Unstable HAS2 Gene Duplications with Cobalamin Deficiency in the Chinese Shar‐Pei GI‐31 Julia Honneffer The Fecal Microbiome of Coyotes (Canis latrans) and a Comparison with Healthy Pet Dogs (Canis canis) GI‐32 Alexis Lecoindre Focal Intestinal Lipogranulomatous Lymphangitis in 10 Dogs GI‐33 Sue Yee Lim Serial Evaluation of Specific Canine Pancreatic Lipase Immunoreactivity and C‐Reactive Protein in Dogs with Cerulein‐Induced Acute Pancreatitis GI‐34 Jessica Rigling Validation of a Urinary 5‐Hydroxyindoleacetic Acid Assay, and Preliminary Use in Dogs with Enteropathies GI‐35 Panpicha Sattasathuchana Analytical Validation of a Gas Chromatography/Mass Spectrometry Method for the Quantification of 3‐Bromotyrosine Concentrations in Fecal Samples From Dogs GI‐36 Panpicha Sattasathuchana Serum 3‐Bromotyrosine Concentrations in Dogs Infected with Spirocerca lupi GI‐37 Chick Weisse Prospective Evaluation of a One‐Stage Esophageal Balloon Dilation Feeding Tube for Benign Esophageal Strictures in Dogs: Initial Results SMALL ANIMAL ‐ HEMATOLOGY HM‐7 Marie Holowaychuk Activated Clotting Time in Healthy Sedated Cats Using the Max‐Act System HM‐8 A. Courtney Crane Cytopenias Associated with Procarbazine Therapy in 8 Dogs HM‐9 Lyndsay Kong Evaluation of the Enterprise Point‐of‐Care Blood Gas Analyzer HM‐10 Christiane Weingart Osmotic Fragility Test – A Useful Diagnostic Test in Cats with Anemia? SMALL ANIMAL ‐ HEPATOLOGY HP‐7 Gideon Daniel Thromboelastography in Cats with Cholestatic Liver Disease HP‐8 Whitney Fry Thromboelastography in Dogs with Chronic Hepatitis HP‐9 Alexis Lecoindre The Use of Hepatic Non‐Invasive Markers for Monitoring Fibrogenic Progression in 49 Scottish Terriers with Vacuolar Hepatopathy SMALL ANIMAL ‐ IMMUNOLOGY IM‐4 Oliver Garden Towards the Development of a Flow Cytometric Assay for Canine Interleukin‐10 IM‐5 Tracy Webb In Vitro Evaluation of the Effects of Pre‐Conditioning and Gender on Feline Adipose‐Derived Mesenchymal Stem Cell Cytokine Production SMALL ANIMAL ‐ INFECTIOUS DISEASE ID‐16 Ramaswamy Chandrashekar Efect of Treatment on Borrelia burgdorferi and Anaplasma phagocytophilum Test Results ID‐17 Emily Kirkpatrick Prevalence of Dirofilaria immitis in Non‐Endemic Northern Colorado Animal Shelters ID‐18 Sanae Kubota Prevalence of Novel Helicobacter Strains in Japan ID‐19 Michael Lappin Prevalence of Giardia duodenalis and Cryptosporidium Spp. Amongst Senior Veterinary Students and Their Dogs and Cats ID‐20 Michael Lappin Detection of Feline Immunodeficiency Virus Antibodies in Serum of Vaccinated Cats Using a Commercially Available Kit ID‐21 Mary Marcondes Gene Expression of Interleukines in the Spleen of Cats Naturally Infected By Leishmania infantum chagasi ID‐22 Nancy Moran Canine and Livestock Husbandry Practices Among Ohio Farmers and Potential for Zoonotic Disease Transmission ID‐23 Maria‐Dolores Tabar Evaluation of Selected Canine Vector Borne Pathogens in Blood and Pericardial Samples From Dogs with Pericardial Effusion ID‐24 Sahatchai Tangtrongsup Molecular Detection of Selected Vector‐Borne Pathogens in Dogs with and without Tick or Flea Infestation ID‐25 Brendon Thatcher Performance of the Anigen Rapid Caniv‐4 Test Kit Using Characterized Samples ID‐26 Mandy van Koppen Prevalence of Cyclospora cayetaensis and Other Enteric Parasites in Shelter Cats with Diarrhea in Colorado, USA SMALL ANIMAL ‐ NEPHROLOGY/UROLOGY NU‐16 Aida Vientós‐Plotts Effects of Hematuria on Dipstick Evaluation in Dogs and Cats NU‐17 Giosi Farace Feline Serum Cystatin C Does Not Differentiate Healthy Cats From Those with Chronic Kidney Disease (CKD) NU‐18 Hye young Kim Pharmacokinetic Study of Meropenem in Healthy Beagle Dogs on Intermittent Hemodialysis NU‐19 Daniel Patch High Throughput Imuunoassay That Coreelates to Gold Standard Liquid Chromatography‐Mass Spectrometry (LC‐MC) Assay for the Chronic Kidney Disease (CKD) Marker Symmetricaldimethyl Arginine (SDMA) NU‐20 Jessica Quimby Retroperitoneal Delivery of Mesenchymal Stem Cells for the Treatment of Feline Chronic Kidney Disease NU‐21 Jodi Westropp Safety Evaluation of the Live Biotheratpeutic Product, Asymptomatic Bacteriuria E. coli 2‐12, in Healthy Dogs SMALL ANIMAL ‐ NUTRITION/METABOLISM NM‐7 Eri Onozawa Effect of Single‐Dose Treatment with Synthetic Ghrelin on Canine Anorexic Patients: A Preliminary Study NM‐8 Jullie Allen Safety of the Ghrelin Agonist, Capromorelin, Administered Daily to Cats for 91 Days at an Oral Dose of 6mg/kg NM‐9 Linda Rhodes Capromorelin, an Orally Active Ghrelin Agonist, Caused Sustained Increases in IGF‐1, Increased Food Intake and Body Weight in Cats NM‐10 Camille Torres‐Henderson Effect of Enterococcus faecium Strain SF68 on the Gastrointestinal Clinical Signs of Healthy Cats Administered Amoxicillin‐Clavulanate NM‐11 Carolina Zaghi Cavalcante Weight Loss Program Associated with Vibrating Platform in Obese Dogs and Overweight SMALL ANIMAL ‐ OTHER OT‐7 Katherine Nadolny Idiopathic Immune Mediated Polyarthritis in Dogs: 67 Cases (2006‐2014) OT‐8 Colin Reich Evaluation of the Effects of Time, Temperature, and Storage Conditions on In Vitro Lactate Concentrations OT‐9 Elizabeth Schmidt Acute Phase Protein Concentrations in Cats with Urinary Tract Obstruction OT‐10 Karina Seferian Monoclonal Antibodies Specific to Serum Amiloid A From Different Species SMALL ANIMAL ‐ PHARMACOLOGY P‐3 Sarah Guess The Effect of Concurrent and Delayed Sucralfate Administration on the Relative Bioavailability of Fluoroquinolones in Greyhounds P‐4 Kellyi Benson Pharmacodynamics of Transdermal Mirtazapine in Healthy Client‐Owned Cats P‐5 Naris Thengchaisri Comparison of Tolfenamic Acid and Robenacoxib for Postoperative Analgesia in Cats After Ovariohysterectomy P‐6 Hitoshi Kakiuchi Safety and Efficacy of Intravenous Tranexamic Acid As an Emetic in Canine Patients P‐7 Jessica Quimby Pharmacokinetics and Pharmacodynamics of Transdermal Mirtazapine in Purpose‐Bred Cats P‐8 Lesley Rausch‐Derra Safety and Toxicokinetic Profiles in Cats Administered Grapiprant, a Selective Prostaglandin‐Receptor Antagonist P‐9 Hye Bin Wang Pharmacokinetic Comparison of Body Surface Area and Body Weight Based Prednisolone Dose on Small and Large Breed Dogs SMALL ANIMAL ‐ RESPIRATORY R‐4 Amy Brida Severe Pre‐Capillary Pulmonary Hypertension in 11 Shih Tzu Dogs: (2007‐2014) R‐5 Modest Vengust Effect of Furosemide on Transvascular Fluid Fluxes in Pulmonary Circulation in Dogs Under General Anesthesia R‐6 Hyuck Joo Yang The Pharmacokinetics of Β2‐Adrenoceptor Agonist in Tulobuterol on Beagle Dogs After Transdermal and Intravenous Administration EQUINE E‐35 Olivia Lorello White Coat Hypertension, Interday Variation of Blood Pressure and Interday and Interobserver Variation of Splenic Volume E‐36 Natalie Heliczer Correlations of Blood Pressure, Splenic Volume and Hematocrit After Phenylephrine and Acepromazine Administration E‐37 Laura Dunbar Comparison of Morphometric Data and Proxy Measurements of Insulin Resistance to Minimal Model Analysis in Adult Light‐Breed Horses E‐38 Heidi Banse Hair Cortisol in Horses with PPID E‐39 Katarzyna Dembek Dynamics of Adrenal Steroids, Steroid Precursors and Neurosteroids in Neonatal Foals During Hospitalization E‐40 Nicholas Frank Repeatability of the Oral Sugar Test and Glucagon‐Like Peptide‐1 and High Molecular Weight Adiponectin Concentrations in Horses with Equine Metabolic Syndrome E‐41 Melanie Fratto Cortisol Binding Globulin Concentrations in Septic Foals, Healthy Foals, and Healthy Horses E‐42 Steven Grubbs Clinical Signs Associated with PPID Status in a Large Population of Horses E‐43 Sarah Mercer‐Bowyer Evaluation of Wellbeing Parameters in New York City Carriage Horses E‐44 Frank Andrews Prospective Evaluation of a Colic Prediction Model E‐45 Mark Bowen Diagnostic Value of Gastric Mucosal Biopsies in Horses with Glandular Disease E‐46 Michael Katz Water in Ingesta Influences Blood Glucose Concentration After Eating in the Horse E‐47 Heath MacLeod Prevalence of and Risk Factors for Gastric Ulceration in Polo Horses E‐48 Benjamin Sykes A Comparison of Gastric Ulcer Prevalence in Feral and Domesticated Horses in the UK: An Abattoir Study E‐49 Benjamin Sykes The Use of an Indwelling Percutaneous Gastrotomy Tube for the Measurement of Intra‐Gastric Ph in the Horse E‐50 Rachel Liepman Severe Myelosuppression and Alopecia in Horses Associated with Trichothecene Ingestion E‐51 Lorena Feijó Plasma Lactate As a Marker of Healthy in Foals Born From Mares with Placentitis E‐52 Joseph Bertone Humoral Response to Borrelia burgdorferi Outer Surface Protein A (OspA) Vaccine via the Subcutaneous and Transdermal Route in Horses E‐53 Brett Sponseller Development of Pulmonary Immunocompetency in the Foal E‐54 Marta Barba Corynebacterium Pseudotuberculosis Seroprevalence in Healthy Horses in a Non‐Endemic State E‐55 Margaret Brosnahan Genome Wide Association Study of Equine Herpesvirus Type 1‐Induced Myeloencephalopathy E‐56 Steven Grubbs WNV IgG Antibody Response in Foals (with and without Maternal Antibody) Following Vaccination with a Combination WNV Vaccine E‐57 Angelika Schoster Therapeutic Monitoring of Plasma Aminoglycoside Concentrations and Their Relation to Minimum Inhibitory Concentrations of Local Gram Negative Bacteria In Horses and Foals E‐58 Sara Azarpeykan Effect of Blanketing on Vitamin D Status in Horses at Pasture E‐59 Joanne Hewson Equine Referring Veterinarians' Expectations of Equine Veterinary Specialists and Equine Referral Care Centers E‐60 Anne Wooldridge Endothelial Progenitor Cell Isolation in Horses E‐61 Amy Stieler A Comparison of the Effects on Sweating of Three Macrolide Antibiotics Used in Foals E‐62 Michela Bullone Endobronchial Biopsy Scores Reliably Assess Disease Severity in Horses with Heaves E‐63 Nicolas Herteman Evaluation of Angiogenesis in the Airways of Horses with Heaves FOOD ANIMAL F‐15 Laureline Lecoq Evaluation of the Cardiomyotoxic Effects of Doxycycline Overdose in Calves Using 2‐Dimensional Speckle Tracking F‐16 Laureline Lecoq Two‐Dimensional Speckle Tracking Echocardiography in Calves: Feasibility, Repeatability and Variability Study F‐17 Hasan Guzelbektes Venous Lactate, Ph and pCO2 Levels As Mortality Indicators in Premature Calves F‐18 Ismail Sen Compare Effect of Combinations of Intravenous and Oral Electrolyte Solution on Treatment of Calf Diarrhea with Mild–Moderate Dehydration and Metabolic Acidosis F‐19 Diego Gomez Cholangiohpatitis in Cattle: 10 Cases F‐20 Shymaa El badawy Comparison of the Pharmacokinetic Values of Cefquinome Following Intramammary Infusion in Healthy Lactating Goats and Goats with Experimentally Induced Staphylococcal aureus mastitis F‐21 Shymaa El badawy Comparison of Cefquinome Concentrations and Pharmacokinetic Values in Plasma and Milk From Healthy and Mastitic Goats Using a Microbiological Assay and High Performance Liduid Chromatography C01 CARDAIC SPECIFIC CALCIUM UPTAKE GENES EXPRESSED IN PERIPHERAL BLOOD EFFECTIVELY REFLECT MYOCARDIAL DISTRESS INDUCED BY HEMODYNAMIC CHANGE Junseok Lee, Kayoko Harada, Arane Takahashi, Hiroshi Takano, Takeshi Mizuno, Masashi Mizuno, Masami Uechi JASMINE Veterinary Cardiovascular Medical Center, Yokohama, Kanagawa, Japan Down‐regulated gene expression, encoding sarcoplasmic reticulum (SR) Ca2 + ‐ATPase and its regulatory proteins such as SR Ca2 + ‐ATPase isoform 2α (SERCA2α) and phospholamban (PLN), plays a crucial role in myocardial decompensation due to the impairment of intracellular Ca2+ cycling. It has been reported that SERCA2α and PLN expressed in peripheral blood mononuclear cells (PBMC) not only are translatable to the myocardial setting, but also significantly decrease in dogs with degenerative mitral valve disease. The aim of this study is to clarify how effectively the target genes reflect hemodynamic overload, a primary determinant of myocardial distress. Healthy laboratory beagles (n = 24), very young dogs with patent ductus of arteriosus (PDA; n = 8) and with pulmonic stenosis (PS; n = 5) were enrolled in this study. None of patients have been treated previously with cardiovascular medication. All PDA patients were in subclinical stage, whereas PS patients had severe stenosis. The mRNA levels of SERCA2α and PLN in PBMC were evaluated before and one week after surgical intervention with using quantitative real‐time PCR. Plasma concentration of NT‐proBNP was also measured as a reference of cardiac biomarkers. Compared with control group, the fold‐changes of the both target genes were significantly low in PDA (0.50 ± 0.17 for SERCA2α and 0.41 ± 0.21 for PLN) as well as in PS (0.47 ± 0.20 for SERCA2α and 0.35 ± 0.10 for PLN). In paired comparison between pre‐ and post‐operation, their expression levels significantly increased one week after surgical correction in both PDA (0.74 ± 0.22 for SERCA2α and 0.68 ± 0.24 for PLN) and PS (0.74 ± 0.18 for SERCA2α and 0.65 ± 0.20 for PLN). Meanwhile, plasma levels of NT‐proBNP between before and after surgery were not different in both PDA (before surgery, 761.0 ± 579.2 pmol/L; one week after surgery, 679.9 ± 437.2 pmol/L) and PS (before surgery, 867.2 ± 560.2 pmol/L; one week after surgery, 863.4 ± 523.1 pmol/L). Additionally, the expression levels of SERCA2α and PLN were significantly correlated to hemodynamic indices, such as fractional shortening and LA/Ao ratio in PDA, and peak pressure gradient in PS (P < 0.05). Furthermore, receiver‐operating characteristic analysis showed high values of the area under the curve in SERCA2α (0.95 ± 0.04 in PDA; 0.95 ± 0.04 in PS) and PLN (1.00 ± 0.0 in PDA; 0.98 ± 0.03 in PS). To conclude, SERCA2α and PLN expressed in the PBMC may be effectively able to reflect myocardial distress determined by hemodynamic change as cardiac biomarkers. C02 DIFFERENTIATING CARDIAC VS NON‐CARDIAC CAUSES OF PLEURAL EFFUSION IN CATS USING PLASMA AND PLEURAL FLUID WITH A POINT‐OF‐CARE NT‐PROBNP TEST Melanie Hezzell 1, John Rush2, Elizabeth Rozanski2, Suzanne Cunningham2, Mark Oyama1 1University of Pennsylvania, Pennsylvania, USA, 2Tufts University, Massachusetts, USA Measurement of NT‐proBNP in plasma and pleural fluid samples using a traditional ELISA has previously been shown to differentiate cardiac and non‐cardiac causes of pleural effusion in feline patients. In emergent patients, rapid results obtained using a point‐of‐care test could be of significant clinical benefit. The purpose of this study was to investigate whether use of a point‐of‐care NT‐proBNP test in plasma and pleural fluid samples would have similar utility to the traditional ELISA. Blood and pleural fluid were prospectively collected from cats presenting with pleural effusion to the emergency rooms of two university teaching hospitals. The cause of the pleural effusion was determined in each case by the attending board‐certified cardiologist on the basis of the echocardiogram, history, physical examination and results of any additional diagnostic tests performed. The cardiologist was blinded to all measurements of NT‐proBNP. NT‐proBNP was measured using both second‐generation quantitative plate (BNP‐quant) and point‐of‐care (BNP‐POC) versions of a feline‐specific ELISA. Continuous data are reported as median [interquartile range]. Groups were compared using Mann‐Whitney U tests and Fisher's exact test, as appropriate. Receiver operator characteristic (ROC) curves were constructed to derive cut‐offs for differentiation of cardiac from non‐cardiac cause of pleural effusion. Diagnostic accuracy was assessed by calculation of sensitivity, specificity and positive and negative likelihood ratios. Thirty‐eight cats were studied. Pleural effusion was determined to be of cardiac origin in 21 cats and non‐cardiac origin in 17 cats. No significant differences in weight (P = 0.742), age (P = 0.862), heart rate (P = 0.268) or respiratory rate (P = 0.170) were detected between groups. Plasma (1500 [790–1500] vs 58 [31–174.5], P < 0.001) and pleural fluid (1500 [918–1500] vs 108 [56–324.5], P < 0.001) NT‐proBNP (BNP‐quant), left atrial to aortic root ratio (LA/Ao) (2.59 [2.25–2.82] vs 1.3 [1.2–1.4], P < 0.001) and the proportions of male cats (80% vs 35%, P = 0.008) and cats with murmurs (71% vs 29%, P = 0.021) and gallop rhythms (62% vs 6%, P = 0.001) were higher in the cardiac vs non‐cardiac group. A cut‐off of 452 pmol/L for plasma BNP‐quant yielded a sensitivity of 90.5% (95% CI 71.1–97.4%) and a specificity of 88.2% (95% CI 65.6–96.7%, AUC=0.96) %) with a positive likelihood ratio of 7.29 (95% CI 2.08–28.48) and a negative likelihood ratio of 0.11 (95% CI 0.03–0.41). A cut‐off of 537 pmol/L for pleural fluid BNP‐quant yielded a sensitivity of 81.0% (95% CI 60.0–92.3) and a specificity of 88.2% (95%CI 65.6–96.7%, AUC=0.96) with a positive likelihood ratio of 6.88 (95% CI 1.84–25.71) and a negative likelihood ratio of 0.22 (95% CI 0.09–0.53). Plasma BNP‐POC yielded a sensitivity of 95.2% (95% CI 77.3 – 99.2%) and a specificity of 87.5% (95% CI 64.0 – 96.5%) with a positive likelihood ratio of 7.62 (95% CI 2.08 – 27.95) and a negative likelihood ratio of 0.05 (95% CI 0.01 – 0.37). Pleural fluid BNP‐POC yielded a sensitivity of 100% (95% CI 84.5 – 100%) and a specificity of 64.7% (95% CI 41.3 – 82.7%) with a positive likelihood ratio of 2.83 (95% CI 1.49 – 5.39) and a negative likelihood ratio that could not be calculated (denominator of 0). In conclusion, this is the first study to demonstrate that plasma BNP‐POC differentiates cardiac from non‐cardiac causes of pleural effusion with sensitivity (95.2%) and specificity (87.5%) similar to those of BNP‐quant. However, the pleural fluid BNP‐POC test is of limited clinical utility, since a positive result is a non‐diagnostic finding due to low specificity, although a negative result helps rule out cardiac causes of pleural effusion. C03 PRE‐SPECIFIED ESCALATION OF MEDICAL THERAPY REDUCES PLASMA NT‐PROBNP CONCENTRATIONS IN DOGS WITH STABLE CHF DUE TO MITRAL VALVE DISEASE Melanie Hezzell, Chloe Thorn, Danielle Laughlin, Mark Oyama University of Pennsylvania, Pennsylvania, USA There is evidence in human patients with congestive heart failure (CHF) that therapy designed to achieve a certain reduction in N‐terminal pro‐B‐type natriuretic peptide (NT‐proBNP) improves outcomes. In one previous study, dogs with CHF that experienced lower NT‐proBNP following conventional therapy survived longer than those with persistently high measurements. It is unknown whether it is possible to purposely change plasma NT‐proBNP concentration by adjusting medical therapy in dogs. The aim of the present study was to determine whether a pre‐specified therapeutic algorithm would result in reductions in plasma NT‐proBNP concentration in dogs with chronic CHF secondary to MMVD. Dogs were recruited within 21 days of initial diagnosis of CHF secondary to MMVD. All dogs were clinically stable at the time of recruitment and were receiving at least 2 mg/kg furosemide per day, 0.5 mg/kg enalapril BID and 0.25 mg/kg pimobendan BID. Dogs were excluded for any of the following: more than 1 episode of CHF requiring medical therapy, receipt of diuretics prior to the onset of CHF, history of chronic kidney disease, systemic hypertension, supraventricular tachyarrhythmia or significant other disease. Dogs were examined on up to 3 occasions at 7–10 days intervals. At the first visit, blood pressure measurement, radiography, echocardiography and measurement of NT‐proBNP, blood urea nitrogen (BUN) and creatinine was performed. If NT‐proBNP was <1500pmol/L no adjustments to existing medications were made and only a physical examination and measurement of circulating markers was performed at the next visit (group 0). If NT‐proBNP was ≥1500pmol/L and creatinine was ≤3.0 mg/dL, (group 1) therapy was escalated as follows; if the current furosemide dose was <6 mg/kg/day, this dose was increased by 50%; if the current furosemide dose was ≥6 mg/kg/day, Aldactazide (1 mg/kg SID) was added; if the current dose of furosemide was ≥6 mg/kg/day and the dog was already receiving Aldactazide, the daily pimobendan dose was increased by 50–100%. If any dose adjustments were made, all diagnostic tests were repeated at the next visit. If creatinine was ≥3.0 mg/dL the dog was withdrawn from the study. Repeated measures mixed models were used to evaluate changes in variables over time. Fifteen dogs were recruited. At the first visit, 8 dogs were assigned to group 0 and 7 dogs to group 1. At the second visit, NT‐proBNP was <1500pmol/L in two dogs in group 1, and no further dose adjustments were made. Thus, in 5 dogs, treatment was again escalated based on NT‐proBNP≥1500pmol/L. Plasma NT‐proBNP decreased significantly over time in group 1 (B = ‐156.1pmol/L/day (95% confidence interval (CI) ‐245.8 to ‐66.5), P = 0.001) but did not change in group 0 dogs (B = 51.7pmol/L/day (95% CI ‐30.7 to 134.1), P = 0.209); these rates of change were significantly different between groups (P < 0.001). Serum BUN (B = 1.27 mg/dL/day (95% CI 0.42 to 2.12), P = 005) and creatinine (B = 0.018 mg/dL/day (95% CI 0.004 to 0.033), P = 0.016) increased significantly over time in group 1 but did not change in group 0 dogs (P = 0.376 and 0.928, respectively); these rates of change were not significantly different between groups (P = 0.449 and 0.350, respectively). Serum creatinine was ≥3.0 mg/dL in one dog at its third visit. Vertebral heart size decreased significantly over time in group 1 (P = 0.027), whereas left atrial to aortic ratio did not change in this group (P = 0.879). In conclusion, a pre‐specified treatment escalation algorithm in dogs with stable CHF due to MMVD and with NT‐proBNP≥1500pmol/L resulted in significant decreases in plasma NT‐proBNP concentration over time. Serum BUN and creatinine measurements significantly increased over baseline in these dogs, suggesting that careful monitoring of renal function is necessary during therapeutic escalation. Further study is warranted to determine whether targeted reductions in NT‐proBNP result in improved outcomes in dogs with CHF secondary to MMVD. C04 BIOLOGICAL VARIABILITY OF N‐TERMINAL PRO‐BRAIN NATRIURETIC PEPTIDE IN ADULT HEALTHY CATS Autumn Harris 1, Amara Estrada1, Alexander Gallagher1, Brandy Mincey1, Kenneth Lamb2, Mary Bohannon1, Jancy Hanscom3, Celine Mainville3 1University of Florida, Gainesville, Fl, USA, 2Lamb Consulting, West Saint Paul, MN, USA, 3IDEXX Laboratories, Inc., Westbrook, ME, USA N‐terminal proBNP is increased in humans, dogs and cats with various cardiac disorders. It has been shown to distinguish healthy cats from those with subclinical disease and congestive heart failure from respiratory disease in dyspneic cats. The biological variability of NT‐proBNP and impact on the diagnostic utility of the biomarker is unknown in both healthy cats and those with cardiac disease. The purpose of this study was to determine the biological variation of NT‐proBNP within day and week‐to‐week in healthy adult cats. Adult cats were prospectively evaluated by CBC, biochemistry, total T4, echocardiography, ECG and blood pressure to exclude underlying systemic or cardiac disease. Thirteen adult healthy cats were enrolled and blood samples were obtained at 11 time points over a 6‐week period (0, 2 hr, 4 hr, 6 hr, 8 hr, 10 hr and at weeks 2, 3, 4, 5 and 6). The intra‐individual (CVI) and inter‐individual (CVG) biological variation along with reference change values (RCV) were calculated. The intra‐assay (CVA) variation has previously been determined and is 6.3%. The median daily and weekly NT‐proBNP concentrations for the population were 36.9 pmol/L (range, 23–257) and 24.5 pmol/L (range, 23–113). The median daily and weekly CVI for the population were 11.2% (range, 0–40.4) and 21.5% (range, 0–68.4). The median daily and weekly RCV for the population were 39.6% (range, 17.3‐ 112) and 69.6% (range, 17.3–188). Thus, a change in individual NT‐proBNP concentration greater than 70% would be needed in healthy cats to denote a true change and not biological or assay variation. C05 MICRORNA SIGNALING NETWORKS IN A CANINE MODEL OF MITRAL REGURGITATION SeungWoo Jung, Ray Dillon Auburn University, Auburn, AL, USA MicroRNAs (miRNAs) negatively regulate downstream target genes in heart disease and may play a role as novel molecular biomarkers of valvular degeneration. Regulatory networks of miRNAs in the mitral valve of dogs with mitral regurgitation need to be established to further enhance our understanding of valvular pathogenesis. The study aim was to test the hypothesis that unique miRNA profiles regulate downstream signalling networks that determine molecular pathways in the mitral valve of dogs with early stage mitral regurgitation. Six healthy Beagle dogs were recruited. Baseline examination confirmed normal cardiovascular status based on physical examination, CBC, serum biochemistry, systemic blood pressure, echocardiography, and electrocardiography. Following hemodynamic measurement under general anesthesia, a mild degree of mitral regurgitation was created in three dogs (MR group) through transvenous surgical disruption of the mitral valve chordae tendineae. Three dogs were used as a control group. After 4 weeks of post‐operative monitoring, dogs underwent humane euthanasia, and harvested mitral valve leaflets were immediately processed for miRNA isolation and subjected to miRNA sequencing via a high‐throughput next generation sequencing (NGS) platform (Illumina Hiseq‐2000). miRNA‐seq data from MR dogs were compared to the control group. Quantitative reverse transcriptase polymerase chain reaction (qRT‐PCR) was utilized for validating NGS data. Bioinformatic analysis of NGS data was conducted to identify a subset of downstream genes targeted by differentially expressed miRNAs. Median length of the small RNA reads was 22 bp (range: 10–46 bp). Principal component analysis showed a significant difference in expression profiles between MR dogs and controls (P < 0.01). Using a paired sample design, the expression profiles of 164 miRNAs were found differentially regulated between MR dogs and controls. Cfa‐miR‐196b (involved in stem cell differentiation) was the most highly expressed and cfa‐miR‐208a (associated with cardiac hypertrophy) the most negatively regulated miRNAs on the mitral valve of MR dogs as quantified by RNA sequencing (P < 0.001). Cfa‐mir‐128 and cfa‐mir‐218 were up‐regulated (P < 0.001), and their predicted target gene, phosphodiesterase 3a (PDE3a), was down‐regulated via qRT‐PCR validation (P = 0.038). Gene ontology analysis revealed that differentially expressed miRNAs regulate a set of genes involved in the signalling pathways of myxomatous mitral valve degeneration including TGF‐β and Wnt. Our results suggest that miRNA expression profiles and their downstream target genes may characterize early molecular events in dogs with mild mitral regurgitation. C06 EFFECT OF ACUTE VOLUME OVERLOAD ON THE LEFT ATRIAL PHASIC FUNCTION IN HEALTHY DOGS Tatsuyuki Osuga, Kensuke Nakamura, Tomoya Morita, Sue Yee Lim, Nozomu Yokoyama, Khoirun Nisa, Keitaro Morishita, Hiroshi Ohta, Mitsuyoshi Takiguchi Department of Veterinary Clinical Sciences, Graduate School of Veterinary Medicine, Hokkaido University, Sapporo, Hokkaido, Japan A recent study has shown that the left atrial phasic function, assessed based on the left atrial phasic areas, is impaired during disease progression and its deterioration is an adverse prognostic factor in canine myxomatous mitral valve disease. This study was performed to elucidate the relation between volume overload and the left atrial phasic function in healthy dogs. Six healthy Beagles were anesthetized and intubated. A Swan‐Ganz catheter was placed to measure mean pulmonary capillary wedge pressure (mPCWP). Preload was increased by intravenous infusion of lactated Ringer's solution at 150 ml/kg/h for 90 minutes. Transthoracic echocardiography was performed before (0 minute) and at 15, 30, 45, 60, 75, and 90 minutes after starting infusion. At each timepoint the apical four‐chamber images were recorded and analysed with software based on two‐dimensional speckle tracking echocardiography to derive time‐left atrial area curves. Left atrial total, passive, and active fractional area changes as indicators of reservoir, conduit, and booster pump functions, respectively, were calculated from left atrial phasic areas in the curves. Acute volume overload significantly increased mPCWP at 15 to 90 minutes compared to 0 minute. All fractional area changes (total, passive, and active) at 15 to 90 minutes were significantly increased compared to 0 minute. Significant positive correlations were observed between mPCWP and all fractional area changes. These results indicate that the left atrial phasic function enhanced during volume loading. Loading condition should be considered when evaluating the left atrial phasic function by indices derived from left atrial phasic sizes. C07 DOPPLER ECHOCARDIOGRAPHIC ASSESSMENT OF LEFT VENTRICULAR +DP/DT AND ‐DP/DT IN DOGS WITH CHRONIC MITRAL VALVE DISEASE André Gimenes, Jacqueline Castro, Matheus Mantovani, Caio Duarte, Mariana Ueda, Lilian Petrus, Guilherme Goldfeder, Maria Helena Larsson, Denise Schwartz University of Sao Paulo, São Paulo, SP, Brazil This study aimed to evaluate noninvasive values of +dP/dt and ‐dP/dt in dogs in different stages of chronic mitral valve disease (CMVD), and to correlate these values with systolic and diastolic echocardiographic indexes of left ventricular function. A cross‐sectional prospective observational clinical study of 36 dogs with CMVD was performed; 11 patients were in stage B1 and 9 in stage B2 without congestive heart failure (CHF) and 16 in stage C with CHF. Physical exam, systolic arterial blood pressure, thoracic radiography, electrocardiogram and transthoracic echocardiography were obtained. Dogs in CHF presented lower +dP/dt (2461 mmHg/s, P25‐P75 = 2065–3177, P = 0.0292) and ‐dP/dt (931.7 mmHg/s, P25‐P75 = 742.2–1141, P = 0.0053) compared to those without CHF (+dP/dt=3085 mmHg/s, P25‐P75 = 2560–4309; ‐dP/dt=1393 mmHg/s, P25‐P75 = 974.3–2144), despite the fact that ejection fraction and fractional shortening were not different. Values of ‐dP/dt ​​were correlated with diastolic indexes, as IVRT (r = 0.3821; P = 0.0235), E/IVRT (r = ‐0.5093, P = 0.0019), E wave (r = ‐0.4548; P = 0.0060) and %LVDd (r = 0.6679; P < 0.0001). Both ‐dP/dt and +dP/dt were negatively correlated with LA/Ao ratio (r = 0.6476; P < 0.0001 and r = ‐0.4619; P = 0.006). It was concluded that ‐dP/dt can be a complementary index to assess diastolic function in dogs with CMVD. In this study, +dP/dt was more efficient detecting impairment of systolic function than the routinely used indexes, possibly due to a lower load dependence. C08 CLINICAL CHARACTERISTICS IN 35 DOGS OVER 5 YEARS OF AGE DIAGNOSED WITH PATENT DUCTUS ARTERIOSUS Bruno G. Boutet, Ashley B. Saunders, Sonya G Gordon Texas A&M University College of Veterinary Medicine and Biomedical Sciences, College Station, Texas, USA Patent ductus arteriosus (PDA) is a common congenital heart defect usually diagnosed and closed in young dogs due to a poor long‐term prognosis. Herein we describe the clinical characteristics of 35 dogs at least 5 years of age diagnosed with PDA between November 1997 and October 2014. Data including signalment, physical exam, concurrent cardiac disease, heart size and treatment were obtained from medical records. Dogs presented for clinical signs (respiratory, weakness, syncope) of heart disease (n = 14, 40%) or evaluation of an asymptomatic murmur (n = 21, 60%). A left basilar continuous murmur was present in 32 (91.4%). Dogs were predominately female (n = 23, 65.7%), with median age of 7.4 years (range, 5.1–12.3) and median weight of 9.6 kg (range, 1.0–44.9). Dogs were classified as small (n = 26) or large (n = 9) breed; most common were mixed (n = 7) and German Shepherd (n = 4). Concurrent heart disease was diagnosed in 22 (62.9%) and was predominately degenerative valve disease (n = 13) and pulmonary hypertension in (n = 9). Concurrent congenital heart disease was rare. 33 were left‐to‐right shunting with 2 bidirectional. 10 (28.5%) presented on diuretic therapy or required it in hospital prior to closure. Arrhythmias were identified in 14; most common ventricular (n = 9) and atrial fibrillation (n = 4). Mean ± SD (range) at presentation for the following: vertebral heart size (n = 24) was 13.1 ± 1.59(10.7–18.2); left ventricular internal dimension in diastole and systole indexed to body weight (LVIDdN, LVIDsN) was 2.36 ± 0.57(1.14–4.68) and 1.49 ± 0.44(0.79–3.05), respectively; fractional shortening was 34.17%±8.55 (17.61–49.63). Left atrial enlargement was documented in 28 based on mmode and 2D measurements. PDA closure consisted of transcatheter occlusion in 20 (13 Amplatz ® canine duct occluder (ACDO), 5 coil, 2 vascular plug); surgical ligation in 6. Complications included femoral artery surgical site hemorrhage (n = 2); intraoperative ACDO dislodgment (n = 1). LVIDdN and LVIDsN at post closure recheck were 2.05 ± 0.54(1.29–3.36) and 1.49 ± 0.39(0.91–2.34); respectively. Echocardiographic residual flow was documented within 24 hrs in 8/26 dogs; resolved in all but 1 and developed in 1 by 1–150 days. Diuretic therapy was discontinued in 9/10 dogs after closure. Closure was not performed in 9 because owners declined the recommendation (n = 7) or it was not immediately indicated due to concurrent disease (n = 2). Dogs over the age of 5 years can present with hemodynamically significant PDA, concurrent heart disease is common, and clinical improvement is possible following closure. C09 EVALUATION OF THE CORRELATION BETWEEN SERUM HOMOCYSTEINE CONCENTRATIONS AND THE SEVERITY OF MITRAL VALVE DISEASE IN DOGS Jeong Da‐Min, Lee Chang‐Min, Kang Min‐Hee, Park Hee‐myung Department of Veterinary Internal Medicine, College of Veterinary Medicine, Konkuk University, Seoul, Korea Homocysteine levels are commonly measured to monitor cardiovascular, renal and other diseases in human medicine. However, only a few studies have measured homocysteine and have addressed the relationship between homocysteine and disease in veterinary medicine. The purpose of this study was to estimate the serum homocysteine concentration in dogs and investigated the relationship with myxomatous mitral valve disease (MMVD). Fifty‐three MMVD dogs were enrolled. Patients with systemic problems, including neurologic, renal, hepatic, neoplastic, diabetic, adrenal and thyroid diseases were excluded from the study. Ten healthy beagle dogs were selected as the control group for comparison with the MMVD dogs. The MMVD dogs were classified according to the American College of Veterinary Internal Medicine scheme. Serum homocystein and Tropoin‐I levels were measured by enzyme and chemiluminescence immunoassay, respectively. The serum homocystein levels of MMVD dogs were significantly different from those of controls. Our results demonstrated a significant correlation between homocysteine and the severity of heart failure (controls: 6.72 ± 1.651; ACVIM B: 13.37 ± 4.163; ACVIM C: 18.86 ± 6.736; ACVIM D: 28.26 ± 4.485, P = 0.001). In addition, the homocysteine levels were correlated with troponin‐I (r = 0.336, P = 0.007), serum creatinine (r = 0.460, P = 0.000), systolic blood pressure (r = 0.571, P = 0.000) and La/Ao ratio (r = 0.280, P = 0.042), which have positive correlation with cardiovascular disease. This study demonstrates a positive relationship between homocysteine and the severity of heart failure. Thus, we infer that elevated homocysteine level can be used as a biomarker of heart failure risk prediction in dogs with MMVD. C10 ECHOCARDIOGRAPHIC DIAGNOSIS AND OUTCOME OF CONSTRICTIVE PERICARDIAL DISEASE IN DOGS (18 CASES: 2002–2013) Justin Thomason 1, Marc Kraus3, MaryAnn Radlinsky2, Karen Cornell2, Elizabeth Howerth2, Tiffany Fallaw2, Clay Calvert2 1Kansas State University, Manhattan, KS, USA, 2University of Georgia, Athens, GA, USA, 3Cornell University, Ithaca, NY, USA Constrictive pericarditis (CP) is the result of inflammation/fibrosis. CP can be secondary to neoplasia, infection, or can be idiopathic. The purposes of this retrospective study were to describe the echocardiographic features and outcomes of dogs with CP. The study population was comprised of 18 client‐owned dogs. In addition to standard echocardiography, the physiology function was used to record respiratory cycles so that M‐mode recording of septal motion and blood flow‐velocities measured by pulsed‐wave Doppler could be correlated to phase of respiration. There were 10 males and 8 females. The mean ages and body weights (median; range) were 7 (6.5; 2–11) years and 24.4 (24.0; 2.3–59.5) kg, respectively. Pleural effusion was present in 15 of 18 (83%) and was accompanied by ascites in 5 (28%). Pericardial effusion was absent in 11. Echocardiography was consistent with normal systolic function and absence of dilated chambers. Ventricular interdependence was demonstrated by M‐mode in 18. Findings consistent with constrictive physiology were present in all 14 dogs examined by pulsed‐wave Doppler. Fifteen dogs were treated by pericardiectomy. In 9 dogs with the etiology of idiopathic pericarditis, 7 (78%) were alive and overtly healthy after at least 2 years. Survival was less than 5 months post‐pericardiectomy in 6 dogs with suspected or proven neoplasia. In conclusion, CP should be considered in dogs with pleural effusion in the absence of or with minimal pericardial effusion and absence of dilated chambers. This suspicion should increase with characteristic echocardiographic findings. The prognosis for non‐malignant etiologies is favorable with pericardiectomy. C11 TIMOLOL OPHTHALMIC SOLUTION FOR DIASTOLIC FUNCTION TESTING IN CATS: A PILOT STUDY OF HEART RATE AND ECHOCARDIOGRAPHIC EFFECTS Catherine Gunther‐Harrington, Eric Ontiveros, Timothy Hodge, Joshua Stern Department of Medicine & Epidemiology, School of Veterinary Medicine, University of California Davis, Davis, CA, USA Echocardiographic assessment of diastolic function can prove challenging in cats, partially due to transmitral flow pattern fusion (EAfusion) secondary to elevated heart rates in the clinic. With reduced heart rate, transmitral flow waveforms may separate. Detection of diastolic dysfunction can aid in the early detection of cardiomyopathies. Timolol, an ophthalmic, non‐selective beta‐blocker used in cats with glaucoma is demonstrated to be safe and transiently decrease heart rate in clinical trials. Here we hypothesize that topical timolol ophthalmic suspension will decrease heart rate and allow E and A wave separation during echocardiogram, without significant adverse effects. Additionally, timolol may rapidly demonstrate the effect of beta‐blockade in cats where oral beta‐blockade is being considered. Fourteen healthy cats were enrolled in the study. Six‐lead electrocardiograms and complete echocardiograms were evaluated before, and 20 minutes after, one drop of timolol 0.5% ophthalmic solution was administered to the right eye. Cats with respiratory disease, ophthalmic disease, and cardiac arrhythmias were excluded. Wilcoxon matched‐pairs signed‐rank test revealed a significant median heart rate reduction of 25 bpm after timolol (P = 0.0005). Of 10 cats with EAfusion, 6 separated after timolol (P = 0.01). Four cats without initial EAfusion remained separated after timolol and E:A ratio was not significantly different (P = 0.23). No bradyarrhythmias were noted after administration. Resolution of dynamic right ventricular obstruction was observed in 1 of 1 cats. Ocular timolol safely and reliably reduces heart rate in cats, generates reproducible changes in echocardiographic parameters, may facilitate diastolic assessment, and may rapidly elucidate a cat's response to beta blockade. C12 CHANGES IN THE BIOMECHANICS OF THE LEFT VENTRICLE IN HEALTHY YOUNG AND ADULT GREAT DANES Ilaria Spalla 1, Salvatore Alonge2, Chiara Locatelli1, Monica Melandri1, Paola Brambilla1, Claudio Bussadori3 1Università degli Studi di Milano, DIVET, Milano, Italy, 2Università degli Studi di Milano, VESPA, Milano, Italy, 3Clinica Veterinaria Gran Sasso, Milano, Italy The helical ventricular myocardial band is the anatomic basis of left ventricular (LV) fiber architecture and has helped to redefine cardiac motion and function. Recently, the evaluation of myocardial motion can be quantified by means of non‐invasive advanced echocardiography (speckle‐tracking echocardiography‐STE), which provides deformation (strain) and deformation rates (strain rate) in the longitudinal, circumferential and radial planes. LV twisting motion along the long axis is another important key feature in LV systolic function. In human cardiology there has been evidence of postnatal cardiac maturation of the myofibers as infants grow, with a difference in the contraction profiles of children and adults, both in strain/strain rate values and LV twist. We sought to evaluate the changes in the biomechanics of the left ventricle in a giant dog breed by performing standard and advanced echocardiography in 2 months old Great Dane puppies and in adult Great Danes older than 1 year. The population comprised 20 puppies and 14 adult Great Danes (mean age: 29 months). No dog presented cardiac murmur. Dogs older than one year of age underwent 24 hours Holter recording and no dog presented any arrhythmia. All standard echocardiographic parameters (end‐diastolic/systolic volumes indexed to body surface area both in M‐mode and B‐mode area length, allometric scaling, LA/Ao ratio and EPSS) were within normal ranges for published dog breed. STE parameters included longitudinal, circumferential and radial strain and strain rate, basal rotation and apical rotation, twist (net difference between apical and basal rotation) and torsion (net twist/ LV diastolic length). Great Danes puppies and adult had statistically different heart rate (177 bpm vs 108 bpm) and selected standard echocardiographic parameters (P < 0.001). Circumferential and radial strain and strain rate and longitudinal strain rate were not different between the two groups, but longitudinal strain, net twist and torsion were significantly different between the two groups (P < 0.001). Puppies had a less negative basal rotation as compared to adults (‐3.6 ± 1.8 vs ‐5.3 ± 1.2, P < 0.03), while no difference was found in apical rotation between the two groups (6.9 ± 2.8 vs 7.7 ± 2.2, NS). In conclusion, our study identifies an age dependency effect on longitudinal strain values, twist and torsional mechcanics of the LV in Great Danes, which appears to reflect at least partially the maturation of the myocardial architecture, as already established in human pediatric cardiology. Standard LA/Ao EPSS EDVI B ESVI B AlloD AlloS EF B Puppies 1.16 1.7a 31.4a 12.1a 1.3a 0.7a 63a Adult 1.17 4.9a 65.6a 32.1a 1.5a 0.9a 56a Advanced Circ St Circ Sr Long St Long Sr Rad St Rad Sr Twist Puppies ‐19.8 ‐2.77 ‐12.2a ‐1.54 32.9 3.17 10.5a Adult ‐20.9 ‐2.39 ‐15.0a ‐1.50 34.9 2.63 13.1a a P < 0.05 statistical significance (t‐test). C13 PLASMA L‐CITRULLINE CONCENTRATIONS IN L‐ARGININE‐TREATED DOGS: A PILOT STUDY FOR THE TREATMENT OF PULMONARY HYPERTENSION Kelly Flynn, Heidi Kellihan, Lauren Trepanier University of Wisconsin, Madison, WI, USA The purpose of this study was to determine whether oral supplementation with L‐arginine could reach plasma concentrations of L‐citrulline, as a surrogate of nitric oxide generation, that are associated with pulmonary vascular hemodynamic improvement in humans with pulmonary hypertension. Three clinically healthy, staff owned dogs were treated in a pilot study with 50 mg/kg q 8 hrs of L‐arginine PO for one week. Plasma samples were obtained at pre‐treatment baseline (BL), at steady state trough (TR) and 0.5, 1, 1.5, 2, 4, 6 and 8 hours after dosing. Plasma L‐citrulline and L‐arginine were analyzed by HPLC. Peak concentrations of both plasma L‐arginine (median 404.7 uM, range 298.8–552.1) and plasma L‐citrulline (median 140 uM, range 92.1–165.4) were consistently seen at 4 hours post dosing (T4 hrs). However, a 33% increase in plasma L‐citrulline from BL, which is the target in humans, was not reached in any dogs (median 10.9%, range −9.5 to 25.5%). Eleven clinically healthy, staff owned dogs were then treated with 100 mg/kg q8 hrs of L‐arginine PO for one week. Plasma L‐arginine concentrations increased from a median (range) of 160.1 uM (100.2–231.4) at BL to 412.7 uM (206.5–807.3) at T4 hrs and plasma L‐citrulline concentrations increased from a median (range) of 82.1 uM (59.1–117.1) at BL to 102.2 uM (47.4–188.3) at T4 hrs. While only 2/11 dogs showed a > 33% increase in plasma L‐citrulline concentrations, all dogs dosed with L‐arginine 100 mg/kg q8 hrs had plasma L‐citrulline concentrations greater than those (mean 44 uM) associated with pulmonary vascular hemodynamic improvement in people. Median systemic systolic, diastolic and mean blood pressures were not significantly different between BL (SBP 150 mmHg, 93–173; DBP 93 mmHg, 45–106; MBP 119 mmHg, 57–128), TR (SBP 145 mmHg, 112–179; DBP 85 mmHg, 52–111; MBP 108 mmHg, 86–138) and T4 hrs (SBP 155 mmHg, 121–180; DBP 90 mmHg, 75–112; MBP 115 mmHg, 96–138) (SBP P = 0.47, DBP P = 0.163 and MBP P = 0.256). Median heart rate did not significantly change between the BL (100 bpm, 79–154), TR (112 bpm, 84–132) and T4 hrs (97 bpm, 74–144) (P = 0.163). Chemistry profiles were analyzed in 7/11 dogs and there were no changes seen at T4 hrs when compared to BL. The results of this study show that BL plasma L‐citrulline concentrations are apparently higher in dogs (median 160.1 uM, range 100.2–231.4) compared to humans (mean 33 uM, SEM ± 3) and that an L‐arginine dosage of 100 mg/kg PO q8 hrs exceeded peak human target concentrations of 44 uM in all dogs treated. Studies are underway to dose L‐arginine in client‐owned dogs with pulmonary hypertension to see whether peak L‐citrulline concentrations, or a specific % change in L‐citrulline concentrations, are associated with clinical and hemodynamic improvement in pulmonary hypertension. C14 LEFT ATRIAL VOLUME OBTAINED BY BIPLANE SIMPSON METHOD IN HEALTHY DOGS: BODY WEIGHT AND BODY SURFACE AREA Rodrigo P Franco1, Evandro Zacché2, Rafael R Camacho2, Fabrício A Marinho2, Fernando A Rosa 2, Aparecido A Camacho2 1University of Marília, Marília, São Paulo, Brazil, 2São Paulo State University, Jaboticabal, São Paulo, Brazil The left atrial volume (LAV) can be obtained echocardiographically determined by biplane Simpson method and is considered a prognostic marker in the evaluation of left atrial enlargement in heart diseases. In dogs, the Simpson method for atrial measurement is still rarely used and therefore there are no reference values, in most cases, the assessment of left atrial enlargement is by the left atrial Performed‐to‐aortic diameter ratio (LA: Ao). Thus, the aim of this study was determined to LAV as the body weight and the body surface area (BSA), using the Simpson biplane method in healthy dogs. For this purpose, were evaluated 107 healthy dogs, subdivided into four groups, being G1 (≤10Kg), G2 (>10 Kg and ≤ 15Kg), G3 (>15Kg and ≤20 kg) and G4 (>20Kg). Dogs were considered healthy when showed no abnormalities in clinical features, including cardiac auscultation, systolic blood pressure and also in complete blood count, serum biochemistry, electrocardiographic and echocardiography evaluation. LAV en diastole (d) and systole (s) was obtained by atrial measurement using the biplane Simpson method obtained by left parasternal long‐axis four and two chamber view. Values from the groups were compared by analysis of variance (ANOVA) and Tukey post hoc test. Subsequently, values from LAV and body weight were submitted to Pearson correlation test and in the presence of statistical significance, these values were indexed to BSA (VAE‐Dixd and LAV‐Sixd). The results are shown in Table 1. Table 1. Mean and standard deviation of LAV obtained from healthy dogs (n = 107), divided into 4 groups based on body weights. PARAMETERS (n = 107) G1 ‐ (n = 35) G2 ‐ (n = 39) G3 ‐ (n = 19) G4 ‐ (n = 14) p LAV‐d (ml) 5.61 ± 0.4 A 8.9 ± 2.4B 13.0 ± 1.7C 13.8 ± 2.4C 0.0001 LAV‐s (ml) 2.80 ± 0.9 A 5.5 ± 1.7B 7.8 ± 0.6C 7.7 ± 1.0C 0.001 Because of the significant differences observed and the presence of significant correlation (LAV‐d = r > 0.77 and LAV‐s = r > 0.73) with body weight, the values of VAE were corrected by BSA identifying LAV‐Dixd = 16.3 ± 3.4 ml/m2 and LAV‐Sixd = 9.5 ± 2.0 ml/m2, both approved in normality test. Thus, it was possible to determine the LAV based on different body weights, as well as the values indexed in canine BSA. C15 DETERMINATION OF LACTATE THRESHOLD IN DOGS WITH DEGENERATIVE MITRAL VALVE DISEASE SUBMITTED TO INCREMENTAL EXERCISE TEST Ana Paula P A Tristão1, Felipe K Adams1, Wilmer A Z Restan1, H G Moranza1, P S Innocente1, Marlos G Sousa2, Guilherme C Ferraz1, Aparecido A Camacho 1 1São Paulo State University, Jaboticabal, São Paulo, Brazil, 2Federal University of Paraná, Curitiba, Paraná, Brazil Exercise became widely used as a diagnostic tool, prognostic and therapeutic in human cardiology. Studies in dogs are scarce and the determination of the lactate threshold in dogs with heart disease is nonexistent in the literature. The aim of this study was determined the lactate threshold in incremental exercise test in dogs with degenerative mitral valve disease (DMVD). The animals were divided in groups based on left atrial to aortic root ratio (LA/Ao): Group 1 (LA/Ao < 1.4 – n = 5); group 2 (LA/Ao 1.4 ‐ 1.8 – n = 5) and group 3 (LA/Ao > 1.8 – n = 5). The test was performed in motorized treadmill with slope of 5%, with increasing speeds of 0.5 m/s in each increment, lasting 5 minutes. Blood samples were collected 90 seconds after each step through a venous catheter placed into the jugular vein. The test was finished when the animal showed signs of fatigue. Exponential regression was used to determine the v1.5 and v2.0 (velocity at which blood lactate concentration reached 1.5 and 2.0 mmol/L). Data were submitted to normality test and analysis between groups was performed by using repeated measure one‐way ANOVA (P < 0.05). There was interaction between the groups (P = 0.002). For v1,5 the groups 2 and 3 had lower values ​​than in group 1 (P = 0.012 and 0.002), with values ​​of 2.0 ± 0.3, 1.6 ± 0.2 and 3.1 ± 0.4 m/s, respectively. Groups 2 and 3 had lower values in ​​v2.0 than in group 1 (P = 0.013 and 0.002), with values of 2.4 ± 0.3 m/s, 2.0 ± 0.2 m/s and 3.6 ± 0.5 m/s. It′s possible determinate the lactate threshold in dogs and the later stages of DMVD need more of the anaerobic pathway contribution to support the physical effort. C16 FAMILIAL VENTRICULAR ARRHYTHMIAS AND SUDDEN CARDIAC DEATH IN THE YOUNG RHODESIAN RIDGEBACK Kathyn Meurs 1, Jess Weidman2, Steve Rosenthal2, Keven Lahmers3, Steve Friedenberg1 1North Carolina State University, Raleigh, NC, USA, 2CVCA Cardiac Care, Annapolis, MD, USA, 3Virginia Tech, Blacksberg, VA, USA Sudden cardiac death in the young dog has been previously reported in the German Shepherd Dog and English Springer Spaniel. We identified four related young Rhodesian Ridgebacks who died suddenly between 7 and 12 months of age. Two of them died while sleeping, and two died after very brief periods of exercise or excitement. The objective of this study was to evaluate for a possible familial relationship and an arrhythmic cause for sudden death in the Rhodesian Ridgeback. We collected pedigrees as well as post‐mortem reports on three of the four dogs that had died. Additionally we collected clinical information including 24‐hour Holter monitor readings and echocardiograms (when possible) on related surviving dogs. In Family 1, two male dogs died suddenly at 9 and 12 months of age. Fourteen surviving dogs including the parents and littermates of the deceased dogs were evaluated with Holter monitoring. Three additional dogs (all female) were determined to be affected based on increased ventricular ectopy (656, 792 and 3912 VPCs per 24 hours, respectively). Both parents of all affected dogs had zero VPCs on 24 hour Holter monitoring. Echocardiography on one affected dog, and post‐mortem evaluation on the two dogs that died of sudden death did not identify any structural lesions. A single female with zero VPCs was either the mother (two dogs) or grandmother (three dogs) to all of the sudden death and affected dogs. Additionally, four of the five affected shared a single father with zero VPCs. In Family 2, two littermates (1 male, 1 female) died suddenly within 11 days of each other at 7 months of age. Three littermates were found to have frequent ventricular ectopy by Holter monitoring (1200–8700 VPCs) and four littermates were normal by Holter monitoring (0 VPCs). The mother had 90 VPCs. The mother, one affected and one unaffected littermate were available for echocardiography and no structural lesions were observed. Post mortem evaluation was performed on one of the sudden death dogs and no structural lesions were observed. The father of this family was unavailable for evaluation. Overall, eleven dogs (6 female, 5 male) were determined to be affected based on increased ventricular ectopy or sudden death. All four dogs were from two families and both families were distantly related through several dogs. The pattern of inheritance appears to be consistent with an autosomal recessive pattern based on the presence of a fairly equal gender distribution and at least in Family 1, the presence of affected dogs with unaffected parents. However, an autosomal dominant mode with incomplete penetrance cannot be completely ruled out. In conclusion, we describe a familial form of ventricular ectopy and sudden cardiac death in the young Rhodesian Ridgeback. An autosomal recessive pattern of inheritance is most likely, and inbreeding should be strongly discouraged. C17 NT‐PROBNP AND TROPONIN I (CTNI) LEVELS AS SCREENING BIOMARKERS IN IRISH WOLFHOUNDS William Tyrrell1, Steven Rosenthal2, Jesse Buch3, Melissa Beall3, Jancy Hanscom3, Frances Abrams4, Mariellen Dentino5 1Chesapeake Veterinary Cardiology Associates CVCA, Leesburg, VA, USA, 2Chesapeake Veterinary Cardiology Associates CVCA, Towson, MD, USA, 3Idexx Laboratories, Inc, Westbrook, ME, USA, 4Irish Wolfhound Foundation, Yellow Springs, OH, USA, 5Nephrology Associates, Evansville, IN, USA A variant of dilated cardiomyopathy (DCM) combined with atrial fibrillation (AF) is a common cardiac ailment within the Irish Wolfhound (IWH) breed. We sought to evaluate normal levels of NT‐proBNP and high sensitivity cTnI in the IWH and whether these biomarker levels could be used as a screening tool for AF with maintained left ventricular systolic function combined with left atrial enlargement (IWH type DCM). Ninety Irish Wolfound dogs were screened at two separate Irish Wolfhound Club of America sanctioned shows in March and May of 2014. Focused cardiac physical examinations, electrocardiograms (ECG), Doppler and 2D echocardiography were performed on all dogs. Medical history, family history of cardiac disease and concurrent medications were recorded on each patient. A biochemical panel, NT‐proBNP level and high sensitivity cTnI was run on each patient. Thirty two males and 58 females were screened. Mean age was 4.61 years (1.2–11.1 years) and mean body weight was 66.4 kg (49.4–103.9 kg). Each dog was classified into one of eight major groups: Normal (n = 35); inconsequential mitral and/or aortic valve insufficiencies (n = 28); IWH type DCM with AF (n = 10); AF with equivocal IWH Type DCM (n = 1); equivocal elevations in left ventricular outflow tract (LVOT) velocities (1.7–2.25 m/sec.) (n = 7); presence of ventricular arrhythmias (VPCs) with normal echocardiogram (n = 7); lone AF (n = 1); and chemodectoma (n = 1). NT‐proBNP and cTnI levels were run on all 90 dogs and the results are summarized below. Diagnosis n NTproBNP, pmol/L High Sensitivity cTnI, ng/ml Median IQR Median IQR Normal 35 685 519–1079 0.03 0.02–0.04 Inconsequential Valve insufficiency 28 898 680–1262 0.05 0.038–0.128 Equiv. LVOT 7 471 387–559 0.02 0.0 IWH DCM/AF 10 5090 3863–5668 0.12 0.058–0.218 EquivDCM w/AF 1 3676 – 0.14 – VPCs 7 976 867–1223 0.06 0.035–0.120 Lone AF 1 4562 – 0.13 – Chemodectoma 1 1329 – 0.08 – Based on 35 normal dogs in this study, reference interval upper limits for NT‐proBNP and cTnI were calculated at 2937pmol/L and 0.99 ng/mL, respectively. NT‐proBNP and cTnI correlated well with cardiac disease in a small sampling of cardiac diseased Irish Wolfhounds and showed significant differences between groups by Kruskal‐Wallis test (P < 0.0001 and P = 0.0007, respectively). NT‐proBNP shows promise as a screening tool for differentiating the presence of equivocal or overt DCM/AF from the absence of DCM/AF in the Irish Wolfhound (AUC = 0.967 by ROC) but is likely to be significantly increased in the absence of DCM for dogs with lone AF. Lesser potential was exhibited by cTnI (AUC = 0.783 by ROC) as a screening tool for IWH‐type DCM. Follow‐up studies with both echocardiography and ECG on the Irish Wolfhounds found to have either elevated NTproBNP and/or cTnI along with normal echocardiography/ECG will be required to determine if NT‐proBNP may have a predictive factor for the development of overt cardiomyopathy and/or atrial fibrillation in this breed of dog. C18 MULTI‐DOSE PHARMACOKINETICS AND PHARMACODYNAMICS OF THE COMMERCIALLY AVAILABLE FORMULATION OF ORAL APIXABAN IN CATS: A PILOT STUDY Christian Weder, Christine Olver, Luke Wittenburg, Petya Trifonova, Jessica Quimby, Janice Bright Colorado State University, Fort Collins, CO, USA Arterial thromboembolism (ATE) is a common and devastating complication of feline cardiovascular disease. There is limited evidence supporting the efficacy of traditional strategies for ATE prophylaxis in cats. Apixiban is a novel oral factor Xa inhibitor that is superior to warfarin in preventing stroke or systemic embolism in people with nonvalvular atrial fibrillation. Apixaban is FDA approved for this indication as well as for the treatment and secondary prophylaxis of deep vein thrombosis (DVT) and pulmonary embolism (PE) in people. Preliminary pharmacokinetic (PK) and pharmacodynamic (PD) data for a single oral dose of pure apixaban in cats were obtained during a previous study at Colorado State University. However, multi‐dose PK and PD data using the commercially available tablet are needed. The purposes of this study were to obtain PK and PD data after multi‐dose oral administration of the commercially available apixaban tablet and to establish appropriate oral dose and dosing intervals for this formulation. Additionally, side effects associated with multi‐dose administration were documented. Three apparently healthy purpose bred cats were used in this 5‐day pilot study. Based on previous studies, all cats were administered 0.625 mg (1/4 of commercially available tablet) PO BID. Following placement of an indwelling jugular catheter on day 0, initial PK data was obtained on day 1 following the first oral dose. Apixaban was then administered orally twice daily on days 1, 2, and 3 and peak and trough plasma concentrations determined. PD activity was assessed on days 1, 2, 3, and 4 by determining the effect of apixaban on factor Xa (FXa) activity using a commercially available assay. A single dose was administered on day 4, and plasma concentrations determined during a 12‐hour washout period. All cats were monitored for side effects associated with multi‐dose administration of the drug. Apixaban had a predictable inhibitory effect on FXa activity in all cats resulting in FXa activity within the therapeutic range (11–42% compared to baseline) used in human patients. This range corresponds to a therapeutic international normalized ratio (2.0–3.5) and was noted in all cats at multiple time points. Furthermore, PD data in these cats showed cyclical inhibition of FXa activity that is typical of novel oral anticoagulant drugs and a property that is considered important for their safety profile. PK data from this pilot study showed a lack of drug accumulation and a consistent elimination half‐life between the first and last oral doses at approximately 2.5 hours. The cyclical inhibition of FXa activity appeared to correlate directly with plasma concentration‐time curves. One cat became lethargic and vomited once on day 3 and was removed from the study. Interestingly, while peak plasma concentrations in this cat were not significantly different, trough concentrations were on average 3‐fold higher. This cat's signs resolved after discontinuing the drug. Bleeding was not noted in any cat. Results from this pilot study show that apixaban is a safe and effective anticoagulant in cats and should be considered for use in the treatment and prevention of feline ATE. The PK and PD data obtained from this pilot study are necessary for subsequent clinical trials in feline patients with heart disease. C19 LONG TERM OUTCOME OF IRISH WOLFHOUNDS WITH LONE ATRIAL FIBRILLATION Claudia Vollmar 1, Bruce Keene2, Barbara Kohn1, Andrea Vollmar3 1Small Animal Clinic, Freie Universität, Berlin, Germany, 2North Carolina State University CVM, Raleigh, NC, USA, 3Small Animal Clinics Dr. Vollmar, Bonn and Wissen, Germany Irish wolfhounds (IW) are commonly affected with atrial fibrillation (AF) and dilated cardiomyopathy (DCM). This study reports the long‐term outcome of 52 IW with apparently “lone” AF, compared to an age and gender matched control cohort. IW from Western Europe (n = 1552) underwent physical examination, standard echocardiography and electrocardiography between 5/1990–10/2014 (AV). Dogs were followed longitudinally, and owners reported the date and circumstances of death. Kaplan‐Meier survival curves were constructed for both groups, and compared using the Log‐Rank (Mantel‐Cox) test. Of the 52 dogs with lone AF, 55% developed DCM after a median of 2.5 ± 1.5 yrs. Overall, 47 deaths occurred in the AF group, with cardiac deaths accounting for approximately 45% (20% died suddenly 2.6 ± 1.4 yrs after diagnosis; 25% developed CHF and died after 3.5 ± 2.2 yrs.). More than 53% died from non‐cardiac reasons 2.8 ± 1.5 yrs after diagnosis; of these, 5 had developed DCM prior to death. The IW control cohort without lone AF survived significantly longer than dogs with lone AF (P = 0.01, median of 1349 vs 1014 days). Lone AF is associated with increased mortality and development of DCM in IW, and represents a strong predictor of cardiac death. C20 IRISH WOLFHOUNDS WITH DILATED CARDIOMYOPATHY: CAUSES OF DEATH Claudia Vollmar 1, Bruce Keene2, Barbara Kohn1, Andrea Vollmar3 1Small Animal Clinic, Freie Universität, Berlin, Germany, 2North Carolina State University CVM, Raleigh, NC, USA, 3Small Animal Clinics Dr. Vollmar, Bonn and Wissen, Germany Dilated cardiomyopathy (DCM) is common in Irish wolfhounds (IW). The purpose of this study was to compare the survival of IW with preclinical DCM and atrial fibrillation (AF) or sinus rhythm (SR) versus dogs with congestive heart failure (CHF) and to report the incidence of cardiac death (sudden cardiac death (SD), or death due to CHF). IW from Western Europe (n = 1552) were examined by physical examination, standard echocardiography and electrocardiography between 5/1990–10/2014 (AV). Dogs were longitudinally followed, and owners instructed to report date and circumstances of death. Kaplan‐Meier survival curves were constructed from date of diagnosis until date of death and compared by Log‐Rank (Mantel‐Cox) test. DCM and/or AF were diagnosed in 29%. Long‐term follow‐up until death was possible in 134 (80 m,54f) dogs. The most common reasons for non‐cardiac death were osteosarcoma, pneumonia, soft tissue cancer, and paraparesis in old IW. DCM+AF+CHF DCM+AF DCM+ Sinus (n = 29) (n = 76) (n = 29) SD (n=) 7 19 6 CHF (n=) 19 18 4 non‐cardiac death (n=) 3 39 19 mean age at diagnosis (years±SD) 5.5 ± 2.6 yrs 4.7 ± 1.8 yrs 5.0 ± 2.2 yrs Median survival (months) 8 48 49 Survival curves for both cardiac and all‐cause mortality for dogs with and without CHF were compared to those from dogs with DCM and AF or sinus rhythm without CHF. Survival in dogs with CHF was significantly shorter than dogs without CHF, regardless of their underlying rhythm at diagnosis. In dogs without CHF, there was no difference in survival between dogs with AF and those with SR. C21 ECHOCARDIOGRAPHIC FINDINGS IN AN AGED POPULATION OF CAVALIER KING CHARLES SPANIELS Jorge Prieto‐Ramos 1, Simon Swift3, Andrea Corda4, Brendan Corcoran2, Kim Summers2, Iñigo Sanz1, Anne French1 1University of Glasgow, Glasgow, Scotland, UK, 2University of Edinburgh, Edinburgh, Scotland, UK, 3University of Florida, Gainesville, Florida, USA, 4University of Sassari, Sassari, Italy Cavalier King Charles Spaniels (CKCS) are predisposed to degenerative atrioventricular valve disease. The prevalence of echocardiographic changes in aged CKCS has not been reported. In this study 126 CKCS (35% males, 65% females) aged 8.0–16.2 years (mean 10.6) underwent echocardiography. Animals ≥ 8 years were recruited from UK regional breed clubs. Echocardiographic examinations were performed by a boarded‐certified cardiologist or a cardiology resident under supervision of a boarded‐certified cardiologist. Echocardiographic parameters measured/calculated comprised: left ventricular diameter diastole/systole (LVDd/s), left atrial (LA) diameter long‐axis view, LA to aorta ratio (LA:Ao) in 2D short‐axis, mitral valve (MV) thickening, MV prolapse, MV regurgitation, tricuspid valve regurgitation (TR), pulmonary hypertension (TR velocity> 2.8 m/s) and fractional shortening. Markers of remodelling were defined as increased LVDd, LVDs, LA size, LA:Ao ratio (>1.5). All dogs had mitral valve thickening and regurgitation. MV prolapse was found in 111/126 (88%), of which 20/111 (18%) were severe. TR was seen in 98%. Pulmonary hypertension was identified in 19/109 dogs (17%). LA:Ao was increased in 49/126 (39%) whereas LA by long‐axis view was enlarged in 59/126 (47%). LVDd was increased in 46/126 (37%) and LVDs was increased in 8/126 (6%). 14%, 21%, 20%, 3% had one, two, three and four markers of remodelling increased respectively. One or more markers of remodelling were reported in 73/126 (58%). In this population of aged CKCS, 100% showed echocardiographic evidence of degenerative atrioventricular valve disease however a low prevalence of markers of disease severity were found. C22 COMPARISON OF METHODS FOR MEASURING LEFT VENTRICULAR WALL THICKNESS AND LEFT ATRIAL SIZE IN CATS Lois Wilkie, Jessie Payne, Tsumugi Kurosawa, Sean Allers, Julia Sargent, David Connolly, Virginia Luis Fuentes Royal Veterinary College, Hertfordshire, UK Aims: to compare inter‐observer agreement between different methods for measuring left ventricular (LV) wall thickness and left atrial (LA) size and evaluate the magnitude of difference between these measurement methods. Two‐dimensional (2D) and M‐mode (MM) echocardiography was recorded in 10 cats of variable phenotype. Four trained observers agreed measurement techniques then independently measured LV septal (IVSd) and free wall (LVFWd) thickness in MM and 3 2D views. IVSd was measured by leading edge to leading edge (LL) and to trailing edge (LT). LA size was measured from MM (LAmax), long axis 2D (LAD) and short axis 2D (LA:Ao). LA:Ao was measured at end‐systole (ES) and end‐diastole (ED). Inter‐observer repeatability was calculated using intraclass correlation coefficients (ICC) and 95% confidence intervals (CI), and paired measurements by the Bland‐Altman method. ICC ranged from 0.31 to 0.86. Agreement was best for MM LAmax (ICC 0.86 (0.68–0.96)) and worst for LA:Ao ES (0.31 (0.05–0.67)). Bias for IVSd was +0.5 mm for LT versus LL; LA:Ao bias was ‐0.25 at ED versus ES; IVSd bias was ‐0.23 mm for short axis 2D versus MM; and LVFWd bias was ‐0.28 mm for 2D versus MM. Inter‐observer agreement varies with measurement method. The magnitude of difference between LL or LT methods for measuring LV wall thickness is sufficient to change the diagnosis between LV hypertrophy and equivocal (or equivocal and normal). Results of LA:Ao measured at ES or ED are not interchangeable. Measurement methods for LV wall thickness and LA should be specified, or else a consensus reached on standardization. C23 IPHONE ECG: A NEW SIMPLE METHOD TO DETECT HYPERTROPHIC CARDIOMYOPATHY (HCM) IN CATS? Clarence Kvart, Veronica Strömberg Anatomy and Physiology Vet. College, Uppsala, Sweden Hypertrophic cardiomyopathy (HCM) in the cat is difficult to easily and rapidly detect in clinical practice by physical examination in both sub clinically and clinically affected cats. Auscultation of the heart is insensitive for the diagnosis even in clinically affected cats. A rapid and reliable diagnosis is desirable before decisions on e.g.: sedation/anesthesia, therapy of acute dyspnea, or screening for breeding purposes. Standard diagnostic tools such as echocardiography, radiography or sampling of blood for analysis of biomarkers are time consuming, not always affordable, or unavailable. A simple, rapid and easily available method to diagnose HCM would be beneficial for many clinicians and clients. Electrocardiographic abnormalities are reported in both subclinical and clinically affected cats. A simple electrocardiographic instrument would simplify the detection of these. The purpose of this pilot study was to evaluate if iPhone ECG (iPECG) using an “AliveCor” veterinary heart monitor could be used to detect HCM in cats. The unit is a small plate containing a microprocessor, and has 2 flat electrodes on it. It attaches as a cover to an iPhone and is controlled by an app. Seventeen normal cats and six cats with HCM were examined with this recorder, together with standard ECG limb lead II (StECG) and echocardiography. ECG was recorded with the iPhone on the left side of thorax over the cardiac region. The (iPECG) was recorded on the left side of thorax with the telephone parallel to the long axis of the heart. The heart was located by auscultation and the middle of the telephone placed over the point of maximal heart sounds. This to ensure that the electrodes were placed one cranial to the heart base and one distal to the heart apex. All ECGs recorded in the iPhone (iPECG) were stored and read blindly later after magnification on the screen of the iPhone. Cats with QRS duration >0.04 sec on both iPECG and StECG were categorized as abnormal. Comparison of iPECG and StECG revealed that only 2 of 17 of the normal cats had QRS duration >0.04s on ECG, producing two false positives using iPECG. On iPECG 8 of 23 cats had QRS duration >0.04s, including all 6 cats positive for HCM on echocardiography (sensitivity = 1.0, specificity = 0.83, κ = 0.8). StECG lead II with cut off QRS duration of 0.04s to detect HCM produced 1 false negative and 1 false positive (sensitivity = 0.83, specificity = 0.94, κ = 0.77). A statistically significant difference (P < 0.05) was found for QRS duration and amplitude between normal cats and cats with HCM using iPECG. iPECG was easy and rapid to record compared to StECG. Despite the limited number of HCM cats the result of this pilot study suggest that screening for HCM and selection of cases to refer for echocardiography could be accomplished with iPhone ECG. Further studies with expanded number of cases are warranted. C24 MORPHOLOGY OF TRICUSPID VALVE APPARATUS IN FELINE HYPERTROPHIC AND DILATED CARDIOMYOPATHY Lauren Wiley 1, Emily Herrold2, Philip Fox1 1The Animal Medical Center, New York, NY, USA, 2Cornell University, Ithaca, NY, USA The tricuspid valve apparatus (TVA) plays a key role in maintaining valvular‐ventricular interaction. Little information describing its morphology exists. We prospectively evaluated 16 archived feline hearts to characterize TVA (HCM‐n = 8; [median, range] [14 yrs, 2–17 yrs]; DCM‐n = 8; [13.5 yrs, 5–17 yrs]; P = 0.66). Using steriomicroscopy, gross structures were examined, photographed, and measured from long axis, four‐chamber sections. The TV comprised two mural leaflets (anterior, anteriolateral) and a septal leaflet. Myxomatous degeneration occurred in 1/8 HCM and 4/8 DCM TVs. RV pectination was hypoplastic (6/8 DCM), normal (6/8 HCM, 2/8 DCM), and hyperplastic (2/8 HCM). Broad, ‘strut’ pectinate muscles (median 4/valve) inserted at the annular junction of mural leaflets in all hearts. Distinctive apical nexuses of obliquely directed pectinate muscles occurred in 2/8 HCM and 7/8 DCM (P = 0.041). At least 1 mural PM occurred in all HCM and 6/8 DCM hearts. Numbers of mural or septal PMs between HCM vs DCM respectively, were NS different (P > 0.05); mural PM diameter (HCM [2.3, 1.1–4.4 mm] vs DCM [1.9, 1.1–3.5 mm] was NS [P = 0.41]); septal PM diameter (HCM [1.1, 0.4–3.4 mm] vs DCM [1.4, 0.4–3.3 mm] was NS [P = 0.21]). Chordal trunks originating from PMs (range, 1–13 trunks/PM) arborized into marginal or ventricular chordae tendineae (CT) subservient to one or more leaflets. Number of chordal trunks from mural PMs (3.5;1–13) were greater than from septal PMs (1, 1–9) (P < 0.001). Thirty‐sixty percent of septal leaflet CT trunks originated directly from ventricular septum, whereas all mural leaflet CT trunks originated from PMs. TVA morphology was variable. Alterations of these structures might influence phenotypic heterogeneity of feline cardiomyopathy and merits further study. C25 THE URINE ALDOSTERONE TO CREATININE RATIO (UALDO:C) DETERMINED BY RADIOIMMUNOASSAY (RIA) IN NORMAL DOGS GREATER THAN 5 YEARS OF AGE Marisa Ames 1, Clarke Atkins2, Kyle Webb2 1College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, CO, USA, 2North Carolina State University College of Veterinary Medicine, Raleigh, NC, USA The urine aldosterone to creatinine ratio (UAldo:C), using a ‘spot’ sample has been correlated to the 24‐hour urine aldosterone excretion in the dog. Evaluation of the UAldo:C can be used to assess the status of a patient's renin‐angiotensin‐aldosterone system (RAAS). Aldosterone, a major effector molecule of RAAS, has been associated with perpetuation and progression of heart failure. Monitoring of the UAldo:C may help guide pharmacotherapy and determine which patient's may benefit from mineralocorticoid receptor blockade. The UAldo:C in 34 normal, 1–6 year old dogs in the authors’ laboratory is 0.40 ± 0.21 μg/g. Only one dog had a UAldo:C greater than 1.0. The prospective study, reported herein, examined the UAldo:C in normal dogs that greater than 5 years of age, which better approximates the cohort of dogs with acquired heart disease. Student and staff‐owned dogs with no history of systemic illness/disease were prospectively recruited and evaluated via a complete history, physical examination, brief chemistry panel (BUN, creatinine, ALP, ALT), PCV/TS, and UA. Five milliliters of urine were collected by free catch between 10am and 2 pm from each dog. The urine aldosterone assay was performed at the Michigan State University Diagnostic Center for Population and Animal Health, using a radioimmunoassay (Siemens Coat‐a‐Count), which measures both free aldosterone and one of its more abundant metabolites, aldosterone glucuronide. Twenty‐six dogs were evaluated and five were excluded due to an abnormal physical, historical, and/or clinicopathological finding. Twenty‐one dogs were included. There were 14 spayed females and 7 castrated males, mean age was 9.2 ± 2.9 (range, 5.4 to 14.6) years, and mean body weight was 25.6 ± 11.1 kg. The mean UAldo:C was 0.49 μg/g, ± 0.27 and no dog had a UAldo:C greater than 1.0. The difference between the UAldo:C in this group of older dogs was not different from the younger, laboratory dogs (P = 0.2). These data were normally distributed. The reference interval was calculated to be (0.05 to 0.92 μg/g), (2.5 and 97.5% percentiles). The 90% confidence interval for the lower limit was (0 – 0.084 μg/g), and for the upper limit (0.882 – 1.23 μg/g). Although evaluation of more dogs is required to establish an exact reference interval, it can be seen that the UAldo:C rarely exceeds 1.0 μg/g in normal dogs and the UAldo:C does not appear to change significantly with age in normal dogs. C26 PRELIMINARY ASSESSMENT OF A NOVEL 14‐DAY AMBULATORY ECG MONITOR (ZIO® PATCH) IN DOGS Jonathan Lichtenberger 1, Kathryn Meurs2, Etienne Côté1 1Atlantic Veterinary College, Charlottetown, PE, Canada, 2North Carolina State College of Veterinary Medicine, Raleigh, NC, USA Cardiac arrhythmias often are transient and might not be detected using conventional electrocardiography or 24‐hour Holter monitoring. The ZIO® Patch (ZP; iRhythm Technologies, Inc, San Francisco, California) is a single‐lead, lightweight, adhesive, 14‐day ambulatory ECG monitor. This device, made for human patients, does not have external leads or wires, is single‐use, has a low‐profile design, is water‐resistant, and has a button that can be pressed to mark clinical events. This study aimed to prospectively assess the usability of this monitor in 4 Boxer dogs considered to be healthy or affected with arrhythmogenic right ventricular cardiomyopathy. The study consisted of 3 parts: Optimal placement of electrodes was investigated using a standard cardiac event monitor. The amplitude of QRS complexes was measured with electrodes placed in 8 different positions on the thorax. The position associated with the tallest QRS complexes was chosen to place the ZP. A ZP was placed on each dog's thorax according to the manufacturer's recommendations. Maximum recording times, quality of ECG tracings during different everyday situations, and complications were documented. During the first 24 hours of the study period, each dog also wore a Holter monitor. Ventricular ectopic beats were manually counted by the same observer (JL) on full disclosure reports obtained with both devices. Optimal recording was obtained by placing the ZP on the left side of the animal's thorax, at the 5th intercostal space, slightly dorsal to the costochondral junction (mid‐ventricle), and oriented either vertically (perpendicular to the spine) or diagonally with a 45° angle from the vertical position, parallel to the long axis of the heart. The concave edge of the ZP was always directed caudally. In 3 dogs, the ZP remained attached for the entire study period (14 days). One dog removed the ZP after 59 hours and developed signs of stress colitis. Mild to moderate skin irritation was documented in all dogs at the time of ZP removal and resolved spontaneously. In all dogs, the analyzable time was > 93% of the total wear time and recordings provided an unambiguous rhythm diagnosis at rest. Walking, running or playing caused intermittent motion artifact that could impair accurate ECG interpretation. ZP results were comparable to those obtained with Holter monitoring during the first 24 hours of the study period: median numbers of premature ventricular complexes, couplets, triplets and runs of ventricular tachycardia obtained from the ZP/Holter monitor were 216/223, 1/1, 0/0 and 0/0, respectively. In conclusion, extended monitoring with the ZP for up to 14 days is feasible in dogs and provides interpretable recordings. C27 CARDIAC BIOMARKER PROFILES IN DOGS WITH NATURALLY‐OCCURRING PRE‐CAPILLARY PULMONARY HYPERTENSION Jonathan Lichtenberger 1, Etienne Côté1, Stephen Ettinger2 1Atlantic Veterinary College, Charlottetown, PE, Canada, 2VetCorp, Beverly Hills, CA, USA The diagnosis of pre‐capillary hypertension (PH) can be challenging in dogs due to the non‐specific nature of associated clinical signs, and the cost and technical requirements for direct and indirect evaluations of pulmonary pressure. Cardiac biomarkers have been evaluated in humans with PH, but limited data are available regarding cardiac biomarker profiles in dogs with naturally‐occurring PH. Based on existing knowledge, PH could lead to right‐sided cardiac remodeling and/or damage, causing the release of N‐terminal pro‐B‐type natriuretic peptide (NT‐proBNP), N‐terminal pro‐A‐type natriuretic peptide (NT‐proANP) and cardiac troponin I (cTnI) into the circulation in dogs. We hypothesized that plasma concentrations of these 3 biomarkers would be higher in dogs with PH compared to dogs with clinical signs compatible with PH but no echocardiographic evidence of PH. PH was considered to be present when a peak tricuspid regurgitation pressure gradient (TRG) ≥ 40 mm Hg was identified echocardiographically in the absence of pulmonic stenosis. Moreover, we hypothesized that plasma concentrations of all 3 biomarkers would positively correlate with TRG. Blood samples were prospectively obtained from dogs over a 27‐month period for determining plasma concentrations of NT‐proBNP, NT‐proANP and cTnI. Dogs were categorized post‐hoc into two groups: PH and no evidence of other cardiovascular disease (affected group). No PH, no evidence of cardiovascular disease and presence of respiratory clinical signs compatible with PH (control group). Twenty‐six dogs met the criteria for categorization: affected group (n = 17) and control group (n = 9). Age, sex, weight, and body condition score were not significantly different between groups. In the affected group, 2 dogs had mild PH (TRG < 51 mmHg), 9 dogs had moderate PH (TRG = 51–75 mmHg) and 6 dogs had severe PH (TRG > 75 mmHg). Plasma concentrations of NT‐proBNP, NT‐proANP and cTnI were significantly higher in the affected group than in the control group (P = 0.018, P = 0.011 and P = 0.016, respectively). There was a moderate positive correlation between NT‐proBNP, NT‐proANP and cTnI levels and TRG (r = 0.56, r = 0.49, r = 0.67, respectively). At a cutoff of 530 pmol/L, sensitivity and specificity of NT‐proBNP to distinguish dogs with PH from dogs with respiratory signs but no PH were 69% and 100%, respectively. At a cutoff of 1025 fmol/L, sensitivity and specificity of NT‐proANP to distinguish dogs with PH from dogs with respiratory signs but no PH were 85% and 86%, respectively. At a cutoff of 0.6 ng/mL, sensitivity and specificity of cTnI to distinguish dogs with PH from dogs with respiratory signs but no PH were 59% and 100%, respectively. Dogs with PH and echocardiographic signs of cardiac remodeling did not have significantly higher biomarker levels than dogs with PH and an echocardiographically normal heart morphology. These 3 biomarkers could be useful tools for detecting PH in dogs and distinguishing dogs affected by this condition from dogs with respiratory signs but no PH. C28 DETECTION OF PERIODONTOPATHIC BACTERIA IN DOGS WITH MITRAL REGURGITAION Masami Uechi 1, Marina Funayama2, Kiyoko Watanabe3, Kumiko Yamashiro2, Toshinobu Kuramatsu2, Masashi Mizuno1, Takeshi Mizuno1, Kayoko Harada1, Arane Takahashi1, Hiroshi Takano1 1JASMINE Veterinary Cardiovascular Medical Center, Yokohama, Japan, 2Nihon University, Fujisawa, Japan, 3Kanagawa Dental University, Yokosuka, Japan Relationship between periodontal pathogens and cardiovascular diseases in dogs is unknown. The purpose of this study was to examine the presence of the DNA of periodontopathic bacteria in heart specimen in dogs with mitral regurgitation. A total 81 dogs with mitral regurgitation were referred to the Nihon University and 196 specimens from left atrium, left ventricle, and mitral valve were harvested during mitral valve repair. The distribution of an animal periodontopathic organism Porphyromonas gulae (P. gulae), in addition to 7 human periodontopathic bacteria, including Porphyromonas gingivalis, Prevotella intermedia, Treponema denticola, Tanneralla forsythia, Campylobacter rectus, Aggregatibacter actinomycetemcomitans and Fusobacterium nucleatum, were examined by polymerase chain reaction (PCR) with species‐specific sets of primers. Oral swab specimens from the same dogs were also collected and analyzed the distribution of periodontopathic bacterial species. Pocket depth and bleeding on probing on the designated tooth were measured as clinical parameters. PCR analysis identified periodontopathic bacteria in 46/81 study animals (56.8%). Among eight periodontopathic bacteria studied, four species, P. gulae, T. forsythia, F. nucleatum, and C. rectus, were detected in heart specimen. The detection rates for P. gulae, T. forsythia, F. nucleatum, and C. rectus in heart samples were 46.9%, 23.9%, 23.9%, and 4.3%, respectively. As for oral specimens from the same subjects, the detection frequencies for P. gulae, T. forsythia, F. nucleatum, and C. rectus, were 84.0%, 82.7%, 75.3%, and 79.0%, respectively. These results suggest that periodontopathic bacteria are related to bacteremia and may be contribute the development of cardiovascular diseases. C29 CARDIAC REGENERATIVE POTENTIAL OF CARDIOSPHERE‐DERIVED CELLS FROM ADULT DOG HEARTS Michael Hensley1, James de Andrade1, Junnan Tang1, Bruce Keene2, Kathryn Meurs 2, Teresa DeFrancesco3, Ke Cheng1 1North Carolina State University. College of Veterinary Medicine, Department of Molecular Biomedical Sciences. Bio‐therapeutics Lab., Raleigh, NC, USA, 2North Carolina State University. College of Veterinary Medicine. Department of Clinical Sciences., Raleiigh, NC, USA, 3Colorado State University, Denver, CO, USA The regenerative potential of cardiosphere‐derived cells (CDCs) for ischemic heart disease has been demonstrated in mice, rats, pigs, and a recently‐completed clinical trial. The regenerative potential of CDCs from dog hearts have yet to be tested. Here we show that canine CDCs can be produced from adult dog hearts. These cells display similar phenotypes in comparison to previously studied CDCs derived from rodents and human beings. Canine CDCs can differentiate into cardiomyocytes, smooth muscle cells, and endothelial cells in vitro. In addition, conditioned media from canine CDCs promote angiogenesis but inhibit cardiomyocyte death. In a doxorubicin‐induced mouse model of dilated cardiomyopathy, intravenous infusion of canine CDCs improves cardiac function and decreases cardiac fibrosis. Histology revealed that injected canine CDCs engraft in the mouse heart and increase capillary density. Out study demonstrates the regenerative potential of canine CDCs in a mouse model of dilated cardiomyopathy. C30 CORRELATION BETWEEN ECHOCARDIOGRAPHIC PARAMETERS AND IL‐1β, IL‐6, IL‐10, TNF‐α AND C‐REACTIVE PROTEIN IN BITCHES WITH SEPSIS Evandro Zacche, Marina Brito, Ana Paula Simões, Ana Paula Tristão, Felipe Adams, Jorge Silva Filho, Aparecido Camacho Unesp, Jaboticabal, São Paulo, Brazil Sepsis is defined as a systemic inflammatory response which occurs during an infection, and myocardial dysfunction is one of its most important features. Although the cardiovascular alterations in human patients with sepsis being studied for over 50 years, data from small animals are minimal. Thus, the aim of this study was to correlate serum concentrations of IL‐1β, IL‐6, IL‐10, TNF‐α and C‐reactive protein with echocardiographic parameters in bitches with sepsis due to cystic endometrial hyperplasia‐pyometra complex. For this purpose, were included in the study 17 bitches with sepsis due to cystic endometrial hyperplasia‐pyometra complex. Prior to hysterectomy (T‐0), 15 days (T‐15) and 30 days (T‐30) after, animals underwent blood sample collection and a complete echocardiographic examination, with left ventricular internal diameter corrected by body surface area. Biomarkers evaluated were C‐reactive protein (CRP), tumor necrosis factor alpha (TNF‐α), interleukin 1 beta (IL‐1β), 6 (IL‐6) and 10 (IL‐10). Biomarkers results were logarithmic transformed and submitted to Spearman's correlation test with echocardiographic parameters (P < 0.05). Results are summarized in Table 1. Table 1. Spearman1's correlation between echocardiographic parameters and TNF‐α, IL‐1β, IL‐6, IL‐10 and CRP in bitches with sepsis (n = 17). IL‐1log IL‐6log IL‐10log TNF‐alog Prot. C Reat.log LVIDdiastole – index NS −0.356a NS −0.308a NS LVIDsystole – index NS NS NS NS NS FS (%) NS NS NS NS −0.324a EF (%) NS NS NS NS −0.332a E/A NS NS −0.289a NS NS IVRT (ms) NS NS NS NS NS mitral EDT (ms) NS NS NS NS NS a : P < 0.05. According to the results of this study, it was possible to observe an inverse correlation between some of the echocardiographic parameters and IL‐6Log, IL‐10Log, TNF‐αLog and CRPLog, confirming myocardial involvement in dogs with naturally occurring sepsis. C31 ASSESSMENT OF QTC INTERVAL AND QT DISPERSION IN DOGS WITH PYOMETRA Thais Pereira, Caio Duarte, Jacqueline Castro, André Gimenes, Matheus Mantovani, Clair Oliveira, Denise Schwartz University of Sao Paulo, São Paulo, SP, Brazil A prospective cross‐sectional study was conducted to evaluate the duration of the corrected QT interval (QTc) and QT dispersion from 58 electrocardiographic tracings, from 31 bitches with pyometra compared to 27 healthy bitches, of same breed or size, and similar age, in order to evaluate possible changes, considering that prolonged QT can be related to arrythmia predisposition in humans and animals. QTc was calculated using Bazett's (QTc= QT/√RR) and Fridericia's (QTc=QT/3√RR) equations, with RR and QT intervals obtained from lead II. QT dispersion (Qtmax‐QTmin) was calculated among 10 leads (I, II, III, aVR, aVL, aVF, rV2 (CV5RL), V2 (CV6LL), V3(CV6LU) and V10 at 50 mm/s). QTc was significantly prolonged in pyometra dogs when calculated by Bazett′s equation (QTccontrol=264.5 ± 24.1 ms vs QTcpyom=277.4 ± 21.6 ms; P = 0.038), while QTc calculated by Fridericia′s equation and QT dispersion were not significantly different. The prolonged QTc in pyometra dogs, evaluated by Bazett′s equation, suggests increased susceptibility to arrhythmias in this group. The lack of difference when using Fridericia′s equation must be better evaluated in larger samples, and these variables should be assessed in a prospective study, including the outcome of arrhythmia occurence during and after surgery. C32 SUBCLASSIFICATION OF STAGE 2 IN CONGESTIVE HEART FAILURE IN DOGS WITH MYXOMATOUS MITRAL VALVE DISEASE Roberto A Navarrete, A S Ferraudo, Aparecido A Camacho São Paulo State University, Jaboticabal, São Paulo, Brazil Dogs with myxomatous mitral valve disease (MMVD) in Class C of the congestive heart failure (CHF) show clinical signs along the progression of the disease, changes in the size of the heart chambers, rhythm cardiac, heart rate, participation of the autonomic system. Multivariate analysis was used to subdivide the class C, using clinical data, echocardiographic, and Holter examinations of 15 dogs with MMVD. Through cluster analysis of the variables described were determined three different groups within class C. Three factors determined from the principal component analysis. The first factor (F1) contains information of 34% of the total variance (TV), which are related to the autonomic nervous system and the variance of the heart rate obtained by Holter‐ECG; the second factor (F2) contains information of 16% of the TV and represents the chambers walls and heart function obtained from the echocardiography; the third factor (F3) includes the information of the 11% of the TV and corresponds to the inner diameter of the right ventricle obtained by echocardiography and the degree of heart murmur in the tricuspid focus during the physical examination. Factors 1 and 2 when plotted on a Cartesian plane (graphic Biplot), represent on the horizontal axis the participation of the autonomic nervous system and on the vertical axis the size of the heart chambers, getting four quadrants, allowing to obtain three groups different of dogs with MMVD in Class C of the CHF. C33 HEART RATE VARIABILITY IN DOGS WITH MYXOMATOUS MITRAL VALVE DISEASE UNDERGOING TREADMILL EXERCISE Felipe K. Adams1, Ana P. P. A. Tristão1, Wilmer A. Z. Restan1, Fernando A. Rosa2, Guilherme C. Ferraz1, Marlos G. Sousa2, Aparecido A. Camacho 1 1São Paulo State University, Jaboticabal, São Paulo, Brazil, 2Federal University of Paraná, Curitiba,Paraná, Brazil Exercise has been widely used as a diagnostic tool, prognostic indicator, as well as therapy adjuvant in human cardiology. Changes in autonomic cardiovascular response, including increased parasympathetic activity, were confirmed in human beings with heart disease after physical activity was initiated, leading to an increased patient survival. However, despite existing evidences in people, limited clinical trials in dogs with cardiac diseases have been performed to assess the effects of exercise on autonomic nervous system. Therefore, this study was aimed at evaluating the changes in autonomic nervous system in dogs with myxomatous mitral valve disease undergoing treadmill exercise for 8 weeks. Fifteen adult Beagle dogs with mitral valve disease were used. Animals were enrolled in three groups according to LA/Ao ratio: LA/Ao<1.4 (n = 5), LA/Ao 1.4‐1.8 (n = 5) and LA/Ao>1.8 (n = 5). Exercise was performed 3 times a week for 30 minutes. The treadmill velocity for each dog was based in 70% of lactate threshold obtained in a previous incremental exercise test. Heart rate variability (HRV) indices were calculated using 24‐hour Holter digital recordings performed prior to exercise (T‐0) and after 8 weeks of treadmill training (T‐8). Time‐domain HRV variables included SDNN, SDANN, SDNNidx, rMSSD, pNN50. Data underwent a normality test, followed by the Student T test to check for differences between T‐0 and T‐8, as well as an ANOVA to verify whether differences existed between groups. No differences were determined to exist between moments, as well as among groups. In spite of the visibly improved aerobic condition, no improvement in HRV indices was documented in dogs with varying degrees of myxomatous mitral valve disease, suggesting that no change occurred in the autonomic nervous system of these animals after the 8‐week treadmill training. However, further studies with more animals trained over a longer period of time might clarify whether exercise might play a beneficial role in sympathetic‐parasympathetic balance in dogs with cardiac diseases. C34 PLASMA OMEGA‐3 FATTY ACID (EICOSAPENTAENOIC ACID) CONCENTRATION S ARE RELATED TO SURVIVAL IN CATS WITH HYPERTROPHIC CARDIOMYOPATHY Lisa M. Freeman, Daniel J. Hall, John E. Rush Cummings School of Veterinary Medicine at Tufts University, North Grafton, MA, USA Circulating omega‐3 fatty acid concentrations are decreased in dogs with congestive heart failure (CHF) and supplementation of omega‐3 fatty acids has some benefits in people and dogs with cardiac disease. Cats with hypertrophic cardiomyopathy (HCM), however, have higher serum concentrations of docosahexanoic acid (DHA) and total omega‐3 fatty acids. This raises the question of whether omega‐3 fatty acid supplementation would be beneficial in cats with cardiac disease. Therefore, the goal of this study was to determine if serum omega‐3 fatty acid concentrations are related to survival time in cats with HCM. Cats with HCM from which serum fatty acid profiles were available and outcome was known were eligible for the study. Cats were excluded if they had significant concurrent disease (e.g., cancer, diabetes, chronic kidney disease) or were receiving fish oil supplementation. Echocardiography was performed on all cats. Serum fatty acid analysis was performed using chloroform methanol extraction and gas chromatography. Fatty acids were categorized as “high” if they were ≥ the median for the group and “low” if they were < the median for the group. Date of death was determined by contacting the owner or the primary care veterinarian. This study was approved by the Cummings School of Veterinary Medicine at Tufts University Clinical Studies Review Committee. Kaplan‐Meier curves and log‐rank tests were used to compare survival times between cats with high vs low omega‐3 fatty acid concentrations using commercial statistical software. 25 cats with HCM were eligible for the study. Median age was 8 years (range, 1‐16 years), with 20 castrated males and 5 spayed females. Breeds included 14 domestic shorthairs, 5 domestic longhairs, 3 Persians, 2 Ragdolls, and 1 Himalayan. Initial presentation was CHF (n = 16), arterial thromboembolism (ATE; n = 2), and asymptomatic (n = 7). Median survival time was 688 days (range, 2‐4043 days). 9 cats were still alive at the time of analysis. Serum eicosapentaenoic acid (EPA) and DHA concentrations were not significantly different between cats with and without CHF or cats with and without ATE. Cats with CHF (P = 0.007) or ATE (P = 0.02) had significantly shorter survival times compared to asymptomatic cats. Cats with a high EPA concentrations had a significantly longer survival time compared to cats with a low serum EPA concentration (P = 0.04). There was no significant difference in survival times between cats with high vs low DHA concentrations. Results of this retrospective study suggest that supplementation with omega‐3 fatty acids, particularly EPA, may have benefits in cats with HCM, but prospective studies are needed. C35 CLINICAL EFFICACY AND SAFETY OF IMIDAPRIL IN DOGS WITH CONGESTIVE HEART FAILURE Wonjung Kim, Heemyung Park Konkuk University, Seoul, Republic of Korea The aim of this study was to compare the clinical efficacy and safety between angiotensin converting enzyme inhibitor (ACEI) imidapril and ramipril in dogs with myxomatous mitral valve disease (MMVD). The medical history of canine patients with MMVD which had been treated with imidapril or ramipril was reviewed. The dogs were treated with either an imidapril solution containing 2.5 mg per ml or ramipril tablets containing 1.25 or 2.5 mg per tablet. Both treatments were administered orally once a day. The dose of imidapril and ramipril was 0.25 mg/kg and 0.125 mg/kg, respectively. Treatment efficacy was evaluated with the clinical efficacy scale (5 stages). Overall clinical efficacy 28 days after the initiation of the treatment was shown to be very good or good in 15/16 (94%) dogs in the imidapril group versus 10/15 (67%) dogs in the ramipril group. However, there was no statistically significant difference in overall clinical efficacy between the treatment groups rated as very good or good (p = 0.056) or in the distribution over the 5 efficacy score levels (p = 0.096). No adverse events associated with treatment in both groups were observed during the study period. In conclusion, this study demonstrated that the ACEI imidapril has a good clinical efficacy in the treatment of dogs with congestive heart failure (CHF) caused by MMVD. Moreover, the results indicate that imidapril is not inferior to the earlier ACEI ramipril, and treatment with the imidapril is well‐tolerated. C36 EVALUATION OF LEFT VENTRICULAR FUNCTION IN DOGS WITH DIFFERENT STAGES OF MYXOMATOUS MITRAL VALVE DISEASE: TISSUE DOPPLER AND STRAIN IMAGING STUDY Yunhye Kim, Dae‐seung Baek, Seungkuk Oh, Namsoo Kim, Jonghoon Kim, Chul Pakr College of Veterinary Medicine, Chonbuk National University, Jeonju, Republic of Korea Myxomatous mitral valve disease (MMVD) is the most common acquired canine heart disease, characterized by various severities of clinical signs from asymptomatic state to sudden death. Because of potential deleterious consequences and high prevalence of MMVD in veterinary medicine, it is important to detect progression of the disease early and accurately. Of tools for evaluation of myocardial function, tissue Doppler imaging (TDI) and strain imaging were relatively recent technique and independent of load conditions compared with conventional echocardiography. Although valuations of TDI and strain imaging have been performed in MMVD dogs, the classification was quite broad to examine changes with progression of the disease. Therefore, we sought to investigate the alterations of TDI and strain imaging in MMVD dogs with different stages divided into 4 groups. In addition, we aimed to investigate the correlation between variables of conventional echocardiographic, and TDI and strain imaging examination. In this study, dogs consisted of 20 healthy dogs and 31 dogs with MMVD. The classification of MMVD dogs used in the study was in the basis of the ISACH scale; class I (n = 10), class II (n = 11), class IIIa (n = 6), and class IIIb (n = 4). 2D, M‐mode echocardiography, and Doppler examinations were performed in all dogs with unsedated state for conventional echocardiography. 2D color TDI examinations were conducted in short‐axis view for radial and circumferential motion, and in longitudinal view for IVS and LV motion. Radial strain and strain rate seemed decreased with severity. Longitudinal motions showed systolic and diastolic TDI and strain variables increased to moderate stage (class II or IIIa) and decreased in severe stage. Intraventricular dyssynchrony, measured as difference in time‐to‐peak between LV and IVS, was observed in MMVD dogs with high prevalence. Dyssynchrony among segments in each wall was also revealed in MMVD dogs. Left atrial to aortic ratio was positively correlated to mitral E wave to peak early diastolic velocity in the LV basal segment (E/Em), and fractional shortening (FS) and ejection fraction (EF) was positively correlated to radial strain. Mitral E wave velocity was negatively correlated to LV systolic time‐to‐peak and strain time‐to‐peak. Circumferential variables had no significant correlations to conventional echocardiographic variables. Table1. Median values of longitudinal left ventricular free wall tissue Doppler and strain imaging variables in MMVD dogs (31) and control dogs (n = 20). Control (n = 20) classI (n = 10) class II (n = 11) class IIIa (n = 6) class IIIb (n = 4) Basal systolic velocity (cm/s) 4.23 (3.85 to 5.5) 4.13 (2.65 to 5.85) 3.85 (2.51 to 4.3) 3.56 (2.39 to 5.14) 2.87 (2.15 to 3.23) Basal early diastolic velocity(cm/s) 3.35 (3.12 to 4.05) 3.3 (2.27 to 6.27) 3.62 (2.15 to 5.81) 4.24 (2.49 to 6.13) 2.85 (2.68 to 4.02) Basal late diastolic velocity(cm/s) 3.01 (2.87 to 3.56) 3.57 (1.66 to 5.7) 3.14 (3.01 to 4.84) 1.34 (0.54 to 2.97) 1.18 (0.24 to 1.78) Systolic time‐to‐peak (ms) 65 (38 to 89) 61 (45 to 137) 49 (32 to 139) 56.5 (36 to 91) 52 (25 to 53) Starin (%) −29.31 (−23.6 to 34.6) −27.06 (−25.32 to 44.31) −22.44 (−21.59 to 31.81) −33.61 (−27.94 to 54.10) −19.98 (−19.05 to 20.33) Strain time‐to‐peak (ms) 223 (176 to 238) 199(177 to 227) 181(150 to 273) 183(124 to 226) 222(117 to 241) Strain rate(s−1) −2.56 (−2.31 to 3.25) −3.75 (−2.23 to 5.54) −2.74 (−1.79 to 3.21) −4.08 (−3.49 to 5.61) −2.12 (−1.8 to 2.71) Strain rate time‐to‐peak (ms) 56 (44 to 72) 49 (35 to 124) 63 (51 to 123) 38 (25 to 67) 46 (36 to 72) C37 CORRELATION BETWEEN RADIAL AND COCCYGEAL ARTERY INDIRECT DOPPLER BLOOD PRESSURE MEASUREMENTS IN CATS Dianne Mawby, Michael Nystrom, Jacqueline Whittemore University of Tennessee, College of Veterinary Medicine, Knoxville, TN, USA Coccygeal arterial blood pressure measurement is better tolerated than radial arterial measurement in cats, but discordant results are common. The purpose of this study was to assess the impact of age, body condition score (BCS), and muscle condition score (MCS) on radial and coccygeal blood pressure measurements. Privately‐owned cats were enrolled between May and December 2010. BCS and MCS were determined by 2 investigators. Blood pressure was measured per ACVIM consensus guidelines, with the order of measurement site randomized prior to the study. The cat was positioned in right lateral or sternal recumbency, respectively, for radial or coccygeal pressure measurement. Associations between blood pressure measurements and other variables were assessed by correlation coefficients and ANCOVA. Sixty‐six cats were enrolled. Inter‐rater reliability for BCS and MCS was high (Inter‐observer correlation coefficients=0.95 and 0.83). There was no significant effect of order of pressure collection (radial versus coccygeal first). Coccygeal and radial artery pressure were correlated (ρ=0.45, P < 0.001); differences were not consistently positive or negative (see Figure). Radial artery pressure was correlated with age (ρ=0.48, P < 0.001) and MCS (ρ=0.30, P = 0.016). Based on partial correlation analysis, the association between radial artery pressure and MCS reflected the confounding influence of age. Age was positively correlated with radial artery results (F value=8.0, p = 0.006). Coccygeal artery measurement was not significantly correlated with age, BCS, or MCS. Use of the coccygeal artery for blood pressure measurement is recommended to reduce the impact of age and sarcopenia on interpretation of results in cats. C38 THREE‐DIMENSIONAL ECHOCARDIOGRAPHIC ASSESSMENT OF NORMAL CANINE MITRAL VALVE Giulio Menciotti 1, Michele Borgarelli1, Sunshine M. Lahmers1, Michael Aherne1, Jens Haggstrom2, Jonathan A. Abbott1 1Department of Small Animal Clinical Sciences, Virginia‐Maryland College of Veterinary Medicine, Blacksburg, VA, USA, 2Department of Clinical Sciences, Swedish University of Agricultural Sciences, Uppsala, Sweden We sought to evaluate the feasibility and repeatability of quantitative assessment of the canine mitral valve by transthoracic 3D echocardiography (3DTEE). Data from examination of healthy dogs were used to define reference intervals for dimensions of the normal canine mitral valve (MV) measured in mid‐systole. Three‐dimensional datasets were acquired from 49 normal dogs from a left‐parasternal apical view, and analyzed with dedicated software. Normal canine mitral valve has a sphericity index 0.92 ± 0.06 and an annulus‐height‐to‐commissural‐width‐ratio of 0.21 ± 0.05, characteristics that confer an elliptical‐ saddle shape. Following logarithmic transformation, annulus diameters, circumference and area, anterior leaflet area and length, and posterior leaflet area were significantly related to BW (p < 0.05), in agreement with an allometric assumption. Overall, datasets were analyzable in 34 of 49 dogs (69%); a higher feasibility characterized the second half of the study period, where 90% of the datasets were analyzable. Interobserver and intraobserver repeatability of the analysis had a coefficient of variation (CV) <20% for all measurements. CV from two different acquisitions of the same individual at two different times were >20% for annulus area, anterior leaflet area and length, tenting height, area and volume. In conclusion, normal dogs have an elliptical and saddle‐shaped annulus and an allometric relationship exists between MV size and BW. 3DTTE is a feasible and generally repeatable method for assessing the MV in normal dogs. The feasibility improved over time as a result of experience. The higher CV derived from measurement of two datasets from the same dog. C39 GENETIC EVALUATION OF PERSIAN CATS WITH HYPERTROPHIC CARDIOMIOPATHY: SCREENING OF MUTATIONS AT ACTC1 AND MYBPC3 GENES Arine Pellegrino, Alexandre Gonçalves Teixeira Daniel, Juliana Mariotti Guerra, Rebecca Bastos Pessoa, Gabriel Garone de Lucca, Bruno Cogliati, Maria Helena Matiko Akao Larsson School of Veterinary Medicine and Animal Sciences – University of São Paulo, Sao Paulo, Sao Paulo, Brazil Hypertrophic cardiomyopathy (HCM) is the most important feline heart disease and it is characterized by ventricular hypertrophy in absence of dilated left ventricle. In humans, the prevalence is 1 to 500 individuals and the familial HCM occurs in at least 60% of cases. There are more than 1400 mutations in more than 11 sarcomeres genes related to HCM. In some families of cats, HCM is an autosomal dominant genetic disease very similar to the human HCM. A mutation in myosin binding protein C gene (MYBPC3) is observed in Maine Coon cats with HCM. In Ragdoll cats, HCM is associated with a mutation in the same gene, but in a different codon highly conserved in feline species. In other breeds such as Persian, British Shorthair and Norwegian Forest there is also evidence of familial HCM, but the type of genetic inheritance is unknown. Persians cats presents high incidence of HCM and there are reports of familial hereditary component, but the mutation and the genetic inheritance have not been discovered yet. In this study, a population of 100 Persian cats was assessed by: echocardiography, electrocardiography, laboratorial tests and blood pressure determination. The animals were classified according to the presence or not of HCM. Blood samples from experimental groups (20 older cats without HCM and 22 cats with HCM) were subjected to DNA extraction, genotyping by PCR and sequencing of cardiac alpha‐actin gene (exon 5 of ACTC1) and myosin binding protein C gene (exon 27 of MYBPC3) with subsequent correlation with the presence of mutations and HCM. In the evaluated population, HCM was more prevalent in older and male cats; it occurred in 22 animals; and the asymmetric hypertrophy at basal region of septum was the most common. One single nucleotide polymorphism (SNP) at position 890 of exon 5 of the gene ACTC1 and three SNP in intron 5‐6 of the same gene were identified. No polymorphism, addition or deletion was observed in other regions of the gene ACTC1 or MYBPC3 gene. Despite the SNP observed in the study, they do not fit the criteria of HCM causal mutation because they do not cause changes in amino acids and do not occurred exclusively in animals with HCM. Thus, a causal mutation of HCM in Persians cat has not been elucidated and mutations in these two exons of cardiac genes do not seem to be the cause of HCM in this breed. Additional screening of cardiac genes is necessary to identify the genetic and molecular cause of this feline disease in Persian cats. C40 CARDIOVASCULAR EVALUATION OF PERSIANS CATS WITH POLYCYSTIC KIDNEY DISEASE Arine Pellegrino, Alexandre Gonçalves Teixeira Daniel, Juliana Mariotti Guerra, Mariana Ferreira Freitas, Natália Cardoso, Gabriel Garone de Lucca, Rebecca Bastos Pessoa, Luiz Fernando Onuchic, Bruno Cogliati, Maria Helena Matiko Akao Larsson School of Veterinary Medicine and Animal Sciences – University of São Paulo, Sao Paulo/ Sao Paulo, Brazil The autosomal dominant polycystic kidney disease (PKD) is characterized by the presence of multiple cysts in the renal parenchyma and occasionally in liver and pancreas, and it is caused by a punctual mutation at exon 29 of the PKD‐1 gene. It is an important cause of chronic kidney disease and the most prevalent genetic disease in cats, affecting mainly Persians cats or correlated breeds. Cats with PKD presents clinical course very similar to PKD disease in humans. Cardiovascular complications are the main causes of morbidity and mortality in human patients with PKD. There are reports of an association between PKD and heart diseases such as hypertension and consequent ventricular hypertrophy, diastolic dysfunction, valve morphological changes, valve prolapse and development of aneurysms. Few studies have evaluated the incidence of hypertension in PKD cats. However, no studies have correlated the presence of ventricular hypertrophy, valve abnormalities or systolic and diastolic changes with feline PKD. In a prospective and observational study, 82 Persians cats were evaluated by electrocardiography, echocardiography, ultrasound, laboratory tests and blood pressure measurement. The PCR was performed to identify the mutation in exon 29 of the PKD1 gene and the animals were classified as PKD positive group (n = 9) and control group (n = 73). No differences in heart rate, weight, age, sex distribution, and electrocardiographic parameters were observed between the groups. The mean of the systolic blood pressure was 135.78 + /‐12.47 mmHg in cats with PKD and 137.48 + /‐20.66 mmHg in control cats, with no differences between the groups. At echocardiographic evaluation, no differences were observed regarding diastolic function, appearance and movement of the leaflets, as well as the valve flows. Myocardial hypertrophy was observed in some animals of the PKD positive group and in some cats of the control group. However, the proportion of hypertrophy was statistically higher in animals with PKD (p = 0.0001). All cats with hypertrophy presented normal blood pressure, normal values of urea, creatinine and total T4, that ruling out other causes of myocardial hypertrophy. Persians cats with PKD mutation presented a higher prevalence of myocardial hypertrophy, even before changes in arterial blood pressure and renal function. Hypertrophic cardiomyopathy (HCM) is also a prevalent condition in Persian cat and should be considered in the differential diagnosis. Further studies are necessary to evaluate the causal relationship between PKD and hypertrophy or the genetic association between PKD and HCM. As a conclusion, it is necessary the cardiovascular evaluation of PKD positive cats (as some may have cardiovascular abnormalities); and it is important to include the PDK (even in early stages) as a differential diagnosis of HCM in Persians cats. C41 ECHOCARDIOGRAPHIC ASSESSMENT OF THE EFFECTS OF CARVEDILOL, CYCLOSPORIN‐A AND SILDENAFIL CITRATE ON DOXORUBICIN‐INDUCED CARDIOMYOPATHY IN AN EXPERIMENTAL MODEL WITH RABBITS Fernando A Rosa 1, Felipe K Adams1, Murillo D Kirnew1, Fabio N Gava2, Aparecido A Camacho1 1São Paulo State University, Jaboticabal, São Paulo, Brazil, 2Unicastelo University, Descalvado, São Paulo, Brazil Doxorubicin is one of the most effective chemotherapeutic agents currently available. However, its use has been limited by its cardiotoxic effect, especially when used for a long time, leading to iatrogenic cardiomyopathy related to the cumulative dose administered. Because of the refractoriness of its cardiotoxic effects, the prevention plays a singular role. With this purpose, carvedilol, cyclosporin A and sildenafil citrate were tested to assess its cardioprotective action in an experimental model of doxorubicin‐induced cardiomyopathy with rabbits. Forty Botucatu rabbits were randomized in four experimental groups. Animals of all groups received doxorubicin (1 mg/kg) intravenously twice a week for 6 weeks. Groups were named G1 (n = 10): doxorubicin, G2 (n = 10): doxorubicin + carvedilol (6.25 mg/Kg), G3 (n = 10): doxorubicin + cyclosporin A (4 mg/Kg) and G4 (n = 10): doxorubicin + sildenafil (1.8 mg/Kg). The drugs other than doxorubicin were given orally, once a day. Echocardiographic evaluations were performed before the first and after the last doxorubicin administration. The statistical analysis was performed by analysis of variance followed by Tukey′s test. Tissue Doppler evaluation in rabbits of all groups demonstrated relationship between peak early diastolic velocity of the mitral annulus (Em) to peak annular diastolic velocity during atrial contraction (Am) that was less than one (Em:Am<1) after the last doxorubicin administration, which was not observed before doxorubicin treatment. Also, significant increase in left ventricle systolic diameter and reduction of ejection and shortening fraction indicated significant left ventricular systolic dysfunction in all groups. The results showed that doxorubicin induced left ventricular diastolic and systolic dysfunction in rabbits, which could not be prevented by concomitant treatment with carvedilol, cyclosporine‐A or sildenafil citrate. C42 LEFT VENTRICULAR CIRCUMFERENTIAL STRAIN AND STRAIN RATE BY TWO‐DIMENSIONAL IMAGING VELOCITY VECTOR IMAGING IN HEALTHY BEAGLES DOGS Fernando A Rosa 1, Edna M G Ortiz1, Ana Paula P A Tristão1, Wilmer A Z Restan1, Murillo D Kirnew1, Evandro Zacche1, Marlos G Sousa2, Aparecido A Camacho1 1São Paulo State University, Jaboticabal, São Paulo, Brazil, 2Federal University of Paraná, Curitiba, Paraná, Brazil Two‐dimensional (2D) echocardiography has been widely used for years to assess left ventricle (LV) systolic function. However, there are some limitations by using the conventional measurement of LV ejection fraction (EF). Two‐dimensional speckle tracking imaging is a simplified and angle‐independent echocardiographic tool which has been recently used for the evaluation of the quantitative global and regional ventricular functions. Syngo Velocity Vector Imaging (VVI) is a speckle tracking‐based method for direct analysis of myocardial motion. The purpose of this study was to investigate the left ventricle systolic function (septal, free wall and global) by using VVI in healthy awake Beagle dogs. For this purpose, we used 12 clinically healthy Beagle dogs, weighing between 10.2 and 12.5Kg. Animals were restrained in right lateral recumbency and echocardiographic evaluation was performed (Acuson X300 Premium Edition, Siemens Medical Solutions USA, Inc.). Echocardiographic clips were obtained in right parasternal short‐axis LV with papillary muscles view and evaluated offline at VVI software (Syngo® Velocity Vector Imaging™, Siemens Medical Solutions). Once a reliable endocardial tracing over a single frame was manually drawn, endocardial borders were automatically tracked throughout the cardiac cycle. Accuracy of border tracking was visually confirmed by viewing the cardiac cycle in slow‐motion while only border information was displayed. Systole was manually selected by moving the left and the right cursor to the onset of the QRS complex and the end of T wave, respectively. For each dog, the mean result of 3 consecutive cardiac cycles was used. Besides LV global peak systolic circumferential strain (St) time to peak (TPK) and mean value of peak systolic circumferential strain rate (SR), we also considered values obtained at the ventricular septum and the LV free wall. Table 1. Mean ± SD of LV global, ventricular septum and LV free wall peak systolic Strain (St) Strain rate (SR) PK (%) TPK (ms) PK (s−1) Mean ± SD Mean ± SD Mean ± SD LV Global −13.4669 ± 4.1730 184 ± 26 −1.3147 ± 0.4010 Ventricular septum −10.4655 ± 6.9863 190 ± 33 −1.0438 ± 0.6030 LV free wall −13.8322 ± 6.2101 179 ± 27 −1.3136 ± 0.4820 Circumferential strain, time to peak and strain rate in 12 awake healthy Beagle dogs. This study aids in the understanding of the LV global, the ventricular septum and the LV free wall systolic function and synchronicity in awake healthy Beagle dogs. VVI demonstrated to be a feasible tool to assess these parameters in these dogs. C43 THE EFFECT OF CATHETER LOCATION DURING SIMULATED CPR ON THE TIME FOR A PERIPHERAL INJECTION TO REACH THE HEART Elizabeth Rozanski 1, Amanda Abelson1, Daniel Fletcher2, John Rush1 1Cummings School of Veterinary Medicine, North Grafton MA, USA, 2Cornell University College of Veterinary Medicine, Ithaca, NY, USA Cardiopulmonary resuscitation (CPR) is often performed in dogs following cardiopulmonary arrest. In patients with pulseless electrical activity (PEA) or asystole, intravenous therapy with vasoactive medications, such as epinephrine or vasopressin is currently recommended. During CPR, catheter placement may be challenging, and due to concurrent chest compressions, placement in the saphenous vein (SV) may be technically easier. In critically ill dogs, with pre‐existing catheters in place, any readily accessible catheter may be chosen. The purpose of this study was to evaluate the time to reach the heart following intravenous administration of contrast material in the SV, the cephalic vein (CV) or jugular vein. Dogs that were euthanized for unrelated purposes and donated to the teaching hospital after death were enrolled. The dog's weight was recorded. Catheters were placed in the JV, CV and SV within 5 minutes of death in order to simulate CPR. Iodinated contrast material was injected (1 ml/5 kg) and flushed with saline while monitoring the cardiac silhouette using fluoroscopy. The time from starting the injection until the appearance of the contrast material in the right heart was recorded in seconds. Saline was flushed until the contrast material was seen in the heart or until a total of 20 ml has been injected. Injections of contrast material that took longer than 30 seconds to reach the heart were recorded as 30 seconds. The order of injection was randomized. Times between groups were compared using a Kruskal‐Wallis test with a p‐value of < 0.05 considered significant. Correlation between time and body weight was evaluated. Fourteen cadaver dogs were enrolled, ranging in size from 3.5 to 35 kg, with a median of 16 kg. Time to reach the heart was significantly longer (p < 0.0001) from the saphenous vein (median 30 seconds; range 6.3 to 30 seconds) then from the cephalic vein (median 3.2 seconds, range 1‐7.2 seconds) and from the jugular vein (median 2.3 seconds; range 1‐4.1 seconds). Nine of 14 saphenous vein injections did not reach the heart within 30 seconds, and only one injection reached the heart in less than 15 seconds. There was no correlation between body weight and time for a contrast material to reach the heart. As fluoroscopy was used to evaluate the time to reach the right heart, concurrent cardiac compression were not performed. This may have influenced the time to reach the heart. However, the results of this study support using either the cephalic vein or the jugular vein for intravenous injections of medications during CPR. C44 CARDIAC BIOMARKERS AND ACUTE PHASE PROTEINS IN FELINE CARDIAC DISEASE Inigo Sanz, Paul Wotton, Jorge Prieto‐Ramos, Valentina Palermo, David Eckersall, Tim D.H. Parkin, Anne T. French Glasgow University, Scotland, UK Congestive heart failure (CHF) is a dynamic neuro‐endocrine disease process characterized by complex pathophysiology, and inflammation is thought to play a role in its progression. Pro‐inflammatory mediators may exert an effect on cardiac structure and function, contributing to progression of cardiac disease. Mild inflammatory infiltrates are commonly seen in the myocardium of cats with cardiomyopathies. In contrast to the human and canine literature, no published reports examine the presence of systemic inflammation in cats with cardiac disease. This prospective study evaluated whether systemic inflammation is present in cats with cardiac disease and if differences exist between animals with or without CHF. Two cardiac biomarkers, cardiac Troponin I (hs‐cTnI) and N‐terminal prohormone of brain natriuretic peptide (NT‐proBNP), and 4 acute phase proteins (APP), serum amyloid A (SAA), C‐reactive protein (CRP), haptoglobin (Hp) and alpha glycoprotein (AGP) were analyzed in serum from 99 cats. 22 cats were diagnosed with cardiac disease, 11 with and 11 without CHF. In the 22 cardiac cases, 20 and 19 of the individuals respectively had hs‐cTnI and NT‐proBNP concentrations above the reference range. Twelve cats had increased AGP and 9 of those were in CHF. Median concentrations of hs‐cTnI, NT‐proBNP and AGP were significantly higher in animals in CHF whereas median concentrations of SAA, CRP and Hp were not. ROC curves demonstrated that AGP, NT‐proBNP and hs‐cTnI were all good indicators of CHF in cats (AUC=0.8). These results confirmed that cats with cardiac disease experience some degree of systemic inflammation, in particular cats in congestive heart failure. C45 DOES SIZE MATTER? A SIGNIFICANT RELATION BETWEEN HEAD WIDTH AND LEFT VENTRICULAR FREE WALL GROWTH RATES IN YOUNG CATS Ingrid van Hoek 1, David J. Connolly2, Lisa Freeman3, John E. Rush3, Lucie Leclerc4, Alexandre Feugier1 1Royal Canin SAS, Aimargues, France, 2Royal Veterinary College, University of London, London, UK, 3Cummings School of Veterinary Medicine, Tufts University, Medford, Massachusetts, USA, 4Oniris, Nantes Atlantic College of Veterinary Medicine, Food Science and Engineering, Nantes, France Cats with hypertrophic cardiomyopathy (HCM) are larger (Yang et al. 2008, Freeman et al. 2013), and body size is related to left ventricular hypertrophy in adult cats (Freeman et al. 2014). Faster early growth might increase the risk for HCM later in life. The objective was to describe morphometric and echocardiographic (echo) measures in cats during early growth. Twenty‐four female European short hair cats born in the same week and fed ad libitum since birth were examined at 6 and 12 months of age. Morphometric measures included body weight (BW), head length (HL), and head width (HW). Echo measures included diastolic (2D‐ and M‐mode) interventricular septum, left ventricular internal dimension, and left ventricular free wall (LVWd), and left atrium in short axis (2D‐mode). Growth rate was % relative growth. A correlation matrix of morphometric measures was determined with Pearson's correlation coefficient adjusted by False Discovery Rate method. Multivariate stepwise regression (Bonferroni correction) analyzed relationships between growth rate and morphometric and echo measures at 6 and 12 months. There was a significant correlation at 6 months for BW‐HW (P = 0.03), BW‐HL (P < 0.001), HL‐HW (P = 0.003), and at 12 months for BW‐HL (P = 0.01). A significant relationship was found at 6 months for BW‐LVWd_M‐mode (P < 0.0001, R2 = 0.51), and for growth rate of HW‐LVWd_M‐mode (P = 0.007, R2 = 0.26). This is the first study describing morphometric and echo measures in cats in the 1st year of life. Further research is needed to investigate a possible association between growth rate at young age and development of cardiac pathology later in life. N01 NEUROPROTECTIVE EFFECTS OF DANTROLENE AND MESENCHYMAL STEM CELLS ON ACUTE SPINAL CORD INJURY Bruno Torres 1, Bernardo Martins2, Carla Maria Silva2, Mario Sergio Lavor3, Milene Alvarenga2, Eliane Melo2 1Federal Rural University of Pernambuco, Garanhuns/Pernambuco, Brazil, 2Federal University of Minas Gerais, Belo Horizonte/Minas Gerais, Brazil, 3The State University of Santa Cruz, Ilheus/Bahia, Brazil This study aimed to evaluate the effects of dantrolene (DAN) and mesenchymal stem cells (MSCs) in acute spinal cord injury (SCI). Sixty three‐months‐old male Wistar rats were divided into five groups: MSCs; MSCs + DAN; DAN; positive control (PC) (trauma and placebo); and negative control (NC) (no trauma + placebo). All procedures were performed according to the principles adopted by the NIH Guide for the Care and Use of Laboratory Animals and by the Ethics Committee on Animal Use (CEUA‐UFMG ‐ protocol number 46/2012). Laminectomy was performed at T12 level in all animals, followed by a weight‐drop model of SCI, except for the NC group. One hour later, the MSCs + DAN and DAN groups received 10 mg/kg of DAN intraperitoneally and after 7 days, the MSCs and MSCs + DAN groups received 1x106 MSCs intravenously. At those times, other groups received the same volume of placebo. Basso, Beattie, Bresnahan (BBB) scale was performed for 30 days to access neurological status. Spinal cords were evaluated through H.E. histology, immunohistochemistry (anti‐NeuN) and qRT –PCR to assess quantitative gene expression of neurotrophic factors (BDNF and NT‐3), antiapoptotic (Bcl‐2 and Bcl‐xl) and pro‐apoptotic proteins (caspase‐3, caspase‐9 and Bax), four milimiters rostral and caudal to the epicenter of the lesion. Traumatized animals showed paraplegia. There was a significant neurological recovery in groups CTM and CTM + DAN from the 22th and 25th days, respectively (p < 0.05) compared to CP. There was severe spinal cord injury characterized by axonal degeneration and neuronal loss, which extended rostrocaudally from epicenter. All groups showed average number of NeuN‐positive neurons significantly higher when compared to CP (p < 0.05). The SCI in PC group resulted in reduced BDNF, NT‐3 and Bcl‐xl gene expression, and increased Bax and caspase‐3 expression compared to non‐traumatized rats (NC). After treatments, there was a significant increasing in BDNF expression in MSCs, MSCs + DAN and DAN groups compared to PC group (p < 0.05). The NT‐3 expression was higher in MSCs + DAN and DAN groups in cranial segments (p < 0.05), and DAN group in the caudal ones (p < 0.01). The Bax expression was significantly lower in the MSCs + DAN (p < 0.05) and DAN (p < 0.001). The caspase‐3 and 9 expression were significantly lower in DAN, MSCs and MSCs + DAN groups (p < 0.05). It was concluded that DAN, MSCs or both in association promote functional recovery and neuroprotection through neurotrophic and antiapoptotic effects. N02 CONGENITAL MYASTHENIC SYNDROME IN JACK RUSSELL TERRIERS IS CAUSED BY A FRAMESHIFT MUTATION IN CHRNE G. Diane Shelton 1, Caitlin Rinz2, Fiona James3, James Thoreson4, Anthony Palmer5, Vanda Lennon4, Leigh Anne Clark2 1University of California San Diego, La Jolla, CA, USA, 2Clemson University, Clemson, SC, USA, 3University of Guelph, Guelph, Ontario, Canada, 4Mayo Clinic, Rochester, MN, USA, 5Cambridge University, Cambridge, UK Congenital myasthenic syndromes (CMS) are a group of structural or functional disorders of neuromuscular transmission resulting from mutations in genes encoding proteins localized to the pre‐synaptic, synaptic or post‐synaptic regions of the neuromuscular junction. The predominant clinical sign is generalized fatigable weakness beginning in the first few weeks of life. In the 1970s a syndrome of early onset weakness and exercise intolerance was described in Jack Russell Terrier (JRT) puppies by Palmer and Lennon. A positive response to an edrophonium chloride challenge was noted, a decremental response of the compound muscle action potential (CMAP) to repetitive supramaximal nerve stimulation supported a disorder of neuromuscular transmission, and an end‐plate deficiency of nicotinic acetylcholine receptor (AChR) was confirmed. Recently, two intact male 7 week‐old JRTs were presented to the Ontario Veterinary College for evaluation of generalized weakness. Similar to the original JRTs, a positive response to edrophonium chloride, a decremental response of the CMAP to repetitive nerve stimulation, and end‐plate AChR deficiency were identified. Inheritance patterns of the chromosomal segments encompassing the 5 genes encoding AChR subunits were evaluated in the JRT family using microsatellite markers, and CHRNE and CHRNB1 were identified as candidates. An insertion in exon 7 of CHRNE that predicts a frameshift and premature stop codon was homozygous in both affected puppies and heterozygous in the parents. DNAs from the original 1970s JRT cases were tested and harbored the identical mutation. The mutation was not identified in 246 unaffected JRTs. Although the carrier rate is very low, the mutation has persisted in the JRT population for over 40 years. N03 PROGNOSTIC INDICATORS FOR THE FUNCTIONAL OUTCOME OF DEEP PAIN NEGATIVE DACHSHUNDS Kathryn M. Winger 1, Bridget Schulte1, Philip H. Kass2, Beverly K. Sturges3, Karen M. Vernau3 1Veterinary Medical Teaching Hospital, School of Veterinary Medicine, University of California‐Davis, Davis, California, USA, 2Department of Population Health and Reproduction, School of Veterinary Medicine, University of California‐Davis, Davis, California, USA, 3Department of Surgical and Radiological Sciences, School of Veterinary Medicine, University of California‐Davis, Davis, California, USA Loss of deep pain perception in dogs with type 1 disc disease is thought to represent functional spinal cord transection with a poor prognosis for recovery to ambulation. Previous studies have reported outcome ranges from 0% to 76%, however there are few known variables that determine which of these patients will ultimately recover. The medical records of paraplegic Dachshunds presented to the Veterinary Medical Teaching Hospital at the University of California, Davis were reviewed to assure deep pain was absent in both pelvic limbs and the tail. Only patients with a T3‐L3 myelopathy localization were included. All patients underwent advanced imaging and decompressive hemilaminectomy. Outcome measures included the return of deep pain perception, ambulation, and urinary continence. Prognostic factors evaluated included speed of onset of clinical signs, duration of paraplegia prior to surgery, duration of surgery, duration of anesthesia, length of decompressive hemilaminectomy, completion of a durotomy, body condition score, age, and sex. A total of 47 dogs met the inclusion criteria. Of these, 31 (66%) regained deep pain perception. A total of 20/31 (65%) regained the ability to ambulate while 22/31 (71%) regained voluntary urination. Of the patients that became ambulatory, 3/20 (15%) required continued manual bladder expression. In contrast, there were 5 non‐ambulatory patients that did regain voluntary urination. None of the prognostic factors evaluated have been shown to significantly correlate with functional recovery. While all facets of the patient's history, physical and neurological examinations, and diagnostic proceedings should be carefully reviewed when prognosticating functional outcome, no single factor can be definitely utilized to determine which deep pain negative patients will recover. Furthermore, all clients should be informed that return of deep pain perception does not guarantee the return of voluntary urination, which is likely a more important consideration for long term management. N04 RETROSPECTIVE EVALUATION OF HORSES DIAGNOSED WITH LYME NEUROBORRELIOSIS ON POSTMORTEM EXAMINATION: 13 CASES (2004‐2014) Laura Johnstone 1, Amy Johnson1, Julie Engiles1, Helen Aceto1, Joy Tomlinson2, Thomas Divers2, Virginia Buechner‐Maxwell3, Deepanker Tewari4, Rachel Gardner5, Rachael Levine6, Nicole Scherrer1 1University of Pennsylvania, New Bolton Center, Kennett Square, PA, USA, 2Cornell University, Ithaca, NY, USA, 3Virginia–Maryland Regional College of Veterinary Medicine, Blacksburg, VA, USA, 4Pennsylvania Veterinary Laboratory, Pennsylvania Department of Agriculture, Harrisburg, PA, USA, 5B.W. Furlong & Associates, Califon, NJ, USA, 6Henderson Veterinary Associates, Elizabethtown, PA, USA Serological analysis indicates that infection of horses with Borrelia burgdorferi is common. However, diagnosis of equine Lyme neuroborreliosis (LNB) is rare, with limited description in the literature. This retrospective study provides a detailed description of clinical and diagnostic data from horses with a histologic diagnosis of LNB. Cases were identified via review of medical records at the University of Pennsylvania and via an ACVIM listserv query. Inclusion criteria included possible exposure to B. burgdorferi and histologic findings consistent with previous reports of LNB, without evidence for other etiologic agents. Thirteen horses were identified, 8 of which had additional evidence of LNB including documentation of intrathecal antibody production (4/6), and detection of spirochetes in CNS tissue by B. burgdorferi PCR (3/7) or Warthin Starry stain (7/8). Clinical signs were variable including ataxia (9/13), muscle atrophy (8/13), cranial nerve deficits (8/13), fasciculations (6/13), neck stiffness (6/13), uveitis (3/13), hyperesthesia (3/13), dullness (2/13) and fever (2/13). Serologic analysis was negative for B. burgdorferi infection in 5/9 cases tested. CSF abnormalities were present in 6/10 cases tested, including xanthrochromia (4/10), elevated total protein (5/10; median: 116 mg/dL, range: 37 ‐ 219 mg/dL) and a lymphocytic (2/10) or neutrophilic (4/10) pleocytosis (median: 48 nucleated cells/μL, range: 0 ‐ 922 nucleated cells/μL). PCR of CSF was negative in 4/4 cases. Diagnosis of equine LNB is challenging due to variable clinical presentation and lack of sensitive and specific diagnostic tests. Negative serology and normal CSF analysis do not exclude the diagnosis of LNB. N05 BRAIN HERNIATION IN 113 CATS: A DESCRIPTIVE STUDY Maria Ines de Freitas 1, Francois Xavier Liebel2, Rita Goncalves3, Holger Volk4, Heike Rudorf5, Sergio Guilherme2, An Vanhaesebrouk6 1n/a, Staffordshire, UK, 2Davies Veterinary Specialists, Higham Gobion, UK, 3University of Liverpool, Neston, UK, 4Royal Veterinary College, Hatfield, UK, 5University of Cambridge, Cambridge, UK, 6University of Oxford, Oxford, UK Brain herniation is commonly reported in cats, however, little is known about their clinical presentation. This cross‐sectional analysis investigated the clinical presentation and MRI finding associated with brain herniation in cats. Clinical records from 113 cats diagnosed with brain herniation by MRI at 4 referral centers were retrospectively collected (2004‐2014). The following data were evaluated: (1) signalment, (2) onset of neurological dysfunction, (3) neurological examination findings, (4) type of brain herniation, (5) location of brain lesion(s), (6) other MRI findings (7) diagnosis and (8) outcome. Mean (SD) age at presentation was 9.8 (3.4) years of age (range: 0.4‐17.0). Domestic short hair cats were over‐represented (76%). Median duration of neurological signs was 12 days (range: 1‐365; IQR: 2‐41). The most common neurological abnormalities found in order of frequency were proprioceptive deficits (N = 86, 76%), obtunded mentation (N = 72, 64%) and nystagmus (N = 44, 39%). 100 (88%) cats were diagnosed with foramen magnum herniation, 34 (34%) of which had concurrent caudal subtentorial herniation. A total of 13 (12%) cats suffered subtentorial herniation alone. Focal rostro‐tentorial lesions were observed in 85 (75%) cats, a caudal fossa lesion in 10 (9%), while 18 (16%) were diagnosed with diffuse or multifocal disease. Most common MRI based diagnoses were neoplasia (N = 79, 66%) and inflammatory brain diseases (N = 13, 12%). 56 (66%) of the cats with focal rostro‐tentorial lesions and concurrent brain herniation showed signs of brainstem dysfunction. Additionally, 27 (48%) of this subgroup of cats presented with nystagmus. This study suggests that cats with focal rostro‐tentorial lesions and concurrent brain herniation may have a higher proportion of brainstem dysfunction than what was previously observed by Walmsley et. al (2006) in dogs (27%). The proportion of cats presenting with nystagmus also seems to be higher than what previously reported in the same study in dogs (16%). N06 EVALUATION OF TEMOZOLOMIDE AND GADOLINIUM CONJUGATED PLGA MICROCYLINDERS IN THE CANINE BRAIN Jill Hicks 1, Simon Platt1, Shannon Holmes1, Elizabeth Howerth1, Ed Kaplan2 1University of Georgia, Athens, GA, USA, 2Microspherix, Boca Raton, FL, USA Poly lactic‐co‐glycolic acid (PLGA) polymers can be used as a drug delivery system that bypasses the blood brain barrier via direct brain implantation. MRI visibility of PLGA is achieved with gadolinium conjugation, but safety of intraparenchymal administration of contrast has not been evaluated. Temozolomide is a chemotherapeutic considered standard‐of‐care for treatment of gliomas in humans. The objectives were to evaluate safety and MRI visibility of PLGA microcylinders conjugated with temozolomide and gadolinium in the canine brain. Four beagles received six microcylinders with 6.25%, 12.5%, 25% or 0% gadolinium. Two received twelve microcylinders with 6.25% or 12.5% gadolinium. Microcylinders other than 0% contained 0.5 mg temozolomide. MRI was performed before implantation, immediately after and 28d later. Cerebrospinal fluid analysis was performed on days 0, 3, 7, 14 and 28. Neurologic status was evaluated on days 0, 1, 7, 14, 21 and 28. On day 28 dogs were sacrificed for necropsy and histopathology of the brain. All dogs had normal neurologic exams at completion of the study. The dog with blank microcylinders had mild inflammation and all other dogs had moderate to severe inflammation within the brain on histopathology, which increased with higher gadolinium concentration. Mild inflammation was present within many of the cerebrospinal fluid samples. All microcylinders with gadolinium were identifiable. Implantation of PLGA microcylinders with gadolinium and temozolomide into the brain is tolerated in healthy beagles. Higher concentrations of gadolinium cause increased inflammation and are unnecessary. It is appropriate to evaluate this method of chemotherapy delivery for dogs with brain tumors. N07 BIOMECHANICAL AND RADIOLOGICAL COMPARISON BETWEEN TRANSARTICULAR SCREW FIXATION AND 2 MULTI‐IMPLANT CONSTRUCTS FOR VENTRAL ATLANTOAXIAL STABILIZATION IN THE DOG Guillaume Leblond 1, Luis Gaitero1, Noel Moens1, Alex Zur Linden1, Fiona James1, Gabrielle Monteith1, John Runciman2 1Ontario Veterinary College, University of Guelph, Guelph, Ontario, Canada, 2School of Engineering, University of Guelph, Guelph, Ontario, Canada Numerous stabilization techniques have been described to treat canine atlantoaxial instability, yet there is currently little data comparing these methods. The main objective of this study was to compare the biomechanical properties, anatomical reduction and safety of 3 ventral atlantoaxial stabilization techniques in canine cadavers. 21 Beagle dog cadavers were randomly assigned 1 of 3 stabilization techniques including transarticular screw fixation (TSF) and 2 multi‐implant constructs (MI5 or MI6) respectively composed of 5 or 6 screws embedded in polymethyl‐methacrylate cement. All screws were placed based on a separate study defining ideal implant placement in Beagles (unpublished data) using a new drill guide prototype designed to improve accuracy of implant placement. C1‐C2 spinal segments were then extracted and computed tomography images were obtained to assess for anatomical reduction and safety of implant placement. Ventral flexion bending moment at failure and stiffness of the constructs were subsequently determined and compared between groups. Ventral stabilization was achieved successfully in all cadavers. No major technical issues were encountered using the drill guide prototype. Significant vertebral canal violation was identified in 4/91 screws (4.4%). Intervertebral apposition was significantly better in the TSF group compared to the multi‐implant groups (p = 0.018). Bending moment at failure was significantly lower in TSF (3.66 ± 1.23N.m) compared to MI5 (8.51 ± 1.96N.m; p = 0.0004) and MI6 (7.73 ± 2.31N.m; p = 0.0021). Mean stiffness measured in the 0‐5° region were not significantly different between groups. TSF provided better intervertebral apposition but was significantly less resistant to ventral flexion compared to 2 multi‐implant fixation techniques in a Beagle cadaver model. N08 COMPUTED TOMOGRAPHY GUIDED STEROTAXIC NAVIGATION OF THE CANINE THORACOLUMBAR SPINE Nicholas Archambault 1, Joan Coates1, Patrick Pithua1, Stephen Frey2, Fred Wininger3 1University of Missouri, Columbia, Missouri, USA, 2Rogue Research Inc, Montreal, Quebec, Canada, 3Veterinary Specialty Services, Manchester, Missouri, USA Implant placements into the canine thoracolumbar spine carry a formidable risk of vertebral canal violation and disruption of juxta‐ vascular and neural structures. Surgical planning using cross sectional imaging and stereotaxy can guide implant placement and mitigate surgical complications. The purpose of this study was to investigate the safety and clinical application of a computed tomography (CT) guided stereotactic technique for use during implant placement in the canine thoracolumbar spine. CT studies using 0.63 mm slice thickness were performed from T12 to L3 vertebral segments in 6 medium sized cadavers. A bilateral dorsal approach of the thoracolumbar spine was used to simulate exposure of the same vertebrae. Stereotactic navigation was achieved with commercially available neuronavigation software. A novel subject tracker and 3‐dimentional CT reconstructions were used for the registration process. A mechanized air drill allowed for a bicortical vertebral body trajectory via CT guided navigation. Stereotactic guided trajectories were stored. A Steinmann pin was placed into the drill hole to trace the trajectory with a post‐operative CT. Vertebral bodies were disarticulated and removed to visually examine for canal violation. Stereotactic guided trajectory and postoperative CT trajectory angle and depth values were compared statistically a using Wilcox rank sum analysis at a significance value of 0.05. A total of 59 trajectories were drilled, with no vertebral canal violations (partial or full). Fifty three of 59 (90%) trajectories involved the vertebral midline or contralateral vertebral body. Mean difference between stereotactic guided trajectory and postoperative CT trajectory angle was 3.8 ± 2.2 degrees (range: 0.3 to 8.3 degrees). Mean difference between stereotactic guided trajectory and postoperative CT trajectory depth was 1.5 ± 1.1 mm (range: 0 to 4 mm). There was no significant difference between stereotactic guided and postoperative CT trajectory angle and depth measurements. Computed tomography guided vertebral stereotaxy provides an accurate means of implant placement into the canine thoracolumbar vertebral column, reducing the risk of damage to vital structures and spinal cord disruption. Additionally, CT guided stereotactic navigation afforded excellent bone purchase. N09 PHARMACOKINETICS OF SINGLE DOSE RECTAL ZONISAMIDE IN HEALTHY DOGS Jennifer Michaels 1, William Thomas1, Dawn Boothe2, Lindsay Williams3, Amy Hodshon1 1University of Tennessee, College of Veterinary Medicine, Knoxville, TN, USA, 2Auburn University, College of Veterinary Medicine, Auburn, AL, USA, 3Carolina Veterinary Specialists, Winston‐Salem, NC, USA The purpose of the study was to evaluate the pharmacokinetics of zonisamide administered per rectum via commercially available encapsulated powder mixed with either sterile water or polyethylene glycol and to determine whether a dose of 20 mg/kg or 30 mg/kg would result in target serum concentrations (10‐40 mcg/mL). Eight healthy mixed‐breed dogs were randomly assigned to 4 groups of 2 dogs each in a crossover‐design study. Zonisamide was administered rectally in a suspension with sterile water or polyethylene glycol at a dose of 20 mg/kg or 30 mg/kg, and blood samples were collected at predetermined time points. After a 7‐day washout period, each group received an alternate treatment until all groups had received all treatments. Serum concentrations of zonisamide were analyzed via high‐performance liquid chromatography. Serum concentrations of zonisamide were within the targeted range at > 1 time point after rectal administration in only 2 dogs. These concentrations were reached after 3 hours (n = 1 dog) or 6 hours (n = 1 dog). In the remaining 6 dogs, serum concentrations of zonisamide were below the target range at all time points. On the basis of these results, rectal administration of 20 mg/kg or 30 mg/kg of zonisamide via a sterile water or polyethylene glycol suspension used in the present study cannot be recommended. N10 FENESTRATION OF THE LATERAL VENTRICLE IN THE MANAGEMENT OF OBSTRUCTIVE HYDROCEPHALUS IN YOUNG SMALL ANIMALS Andy Shores 2, Simon Kornberg1, Michaela Beasley1, Erin Brinkman‐Ferguson1, Jennifer Gambino1, Maria Perez1, C S Lee1, Jennifer Rich1 1Mississippi State University CVM; Department of Clinical Sciences, Mississippi State, MS, USA, 2Veterinary Specialty Center, Starkville, MS, USA Conventional valve shunting for treatment of hydrocephalus has been associated with complications, as high as 22% in one study. Reported complications include dislodgement, occlusion, infections, ventricle collapse, and subdural hematoma. The necessity for shunt revision surgery is reported by several authors. In humans, complication frequency is related to young age and was 32% in a study of 14,000 patients. Because of the level of complications associated with ventriculoperitoneal shunts and an assumed higher rate in very young dogs and cats, the authors sought a different method of treating obstructive hydrocephalus in small animals. Human infants may be preferentially treated with endoscopic assisted 3rd ventricle fenestration into the subarachnoid space through the basilar cistern. The size of the young canine or feline ventricular system, however, is prohibitively small. An alternative procedure (lateral ventricle fenestration) was designed by one member of the team. The procedure had successful short‐term outcomes in 5 patients, but 3 were lost to follow‐up. In a DSH kitten (10 weeks old) and a Labrador puppy (14 weeks old), following an obstructive hydrocephalus diagnosis using advanced imaging, clinical signs of resolved or improved over a period of 6 months. Follow‐up imaging demonstrated patency of the fenestrations and a reduced ventricle sizes. This presentation demonstrates the indications and the technique for this procedure and includes pre‐operative and post‐operative imaging. The authors conclude the short‐term results of this procedure are beneficial in the treatment of obstructive hydrocephalus in puppies and kittens, is easy to perform, and has not resulted in complications. N11 CLINICAL CHARACTERISTICS AND OUTCOME OF CASES DIAGNOSED WITH ELECTROGRAPHIC SEIZURE IN 73 DOGS AND 13 CATS Liana K. Granum 1, William W. Bush1, Christine Weaver1, D. Colette Williams2, Mark M. Stecker3 1Bush Veterinary Neurology Service, Rockville, MD, USA, 2William R. Pritchard Veterinary Medical Teaching Hospital, Davis, CA, USA, 3Winthrop University Hospital, Mineola, NY, USA Electrographic seizure (ES) is defined as ictal discharges consisting of a rhythmic pattern with definitive evolution in frequency, amplitude and/or morphology persisting for at least 10 seconds. In electrographic status epilepticus (ESE) the ictal discharges are noted for greater than 30 minutes. In humans, ES/ESE often manifests without overt signs of convulsion and is reported in 8‐20% of critically ill patient populations. A delayed diagnosis and treatment may lead to increased mortality. This study seeks to better understand the risk factors, characteristics, and outcome of ES/ESE in veterinary patients having had an EEG for suspected seizure. Complete medical records were identified in 73 dogs and 13 cats and data analyzed for prevalence of ES/ESE, EEG patterns, associated risk factors, and outcome. ES/ESE was detected in 15/86 (17%). Felines were at significant risk for ESE. Younger animals and those with other EEG abnormalities were at significant risk for ES. 13/15 (87%) of patients had non‐convulsive seizure with subtle clinical findings like twitching, changes in temperature or respiratory rate, and coma. Although not statistically significant it is important to note that 40% or the ES/ESE cases died in hospital compared to 21% of the non ES/ESE group. ES/ESE is relatively common in patients with suspected seizure and often has subtle clinical manifestations. No clinical characteristic can differentiate ES/ESE from control cases, highlighting the necessity for EEG to make this distinction. Prospective, multi‐center studies are needed to define the true incidence in different populations, better define risk factors, outcome and then best treatment options. N12 UTILITY OF A CLINICAL SCORING SYSTEM AND OCULAR PHARMACOLOGICAL TESTS FOR ANTE‐MORTEM DIAGNOSIS OF FELINE DYSAUTONOMIA Thomas Cave 1, Tim Scase3, Clare Knottenbelt2 1Cave Veterinary Specialists, Wellington, UK, 2Glasgow University Veterinary School, Glasgow, UK, 3Bridge Pathology Ltd, Bristol, UK Feline dysautonomia (FD) is a disease of domestic cats characterised by extensive degeneration of the autonomic nervous system. It was first reported in 1982 in the UK but has now also been reported in the USA, UAE, New Zealand, and several European countries. The aetiology of this condition remains unknown but it has clinical and pathological similarities to primary dysautonomias affecting dogs, hares, rabbits, and horses. FD has been reported to cause a variety of clinical signs including reduced tear secretion, dry oral mucosa, dry nasal planum, mydriasis, prolapsed nictitating membranes, bradycardia, regurgitation and oesophageal dysmotility, constipation, proprioceptive deficits, dysuria, and anal areflexia. Clinical signs typically develop over a matter of days and less than a third of affected cats have survived. Diagnosis is by demonstration of chromatolytic change within neurons in autonomic ganglia but a clinical scoring system (CSS) and ocular pharmacological tests (OPT) have been proposed for ante‐mortem diagnosis [Sharp NJH, Gookin JL (1995) BSAVA Manual of small animal neurology 2nd Ed; 179‐188]. OPT have included a reduction in pupil size within 15 minutes of topical application of 0.1% pilocarpine or retraction of prolapsed nictitating membrane within 15 minutes of topical application of 1:10,000 Epinephrine. The UK FD study group has investigated suspected cases of FD within the UK reported to us by veterinarians. We have used this data to assess the proposed CSS and the ocular pharmacological tests. Since December 2000, 126 suspected cases of FD have been reported. The outcome of 9/126 cases was unknown. 34/126 cases survived for a median follow up period of 39 days (range 5‐168). In 23/34 cases clinical signs improved. The remaining 83/126 cases had died a median of 7 days following presentation (range 0‐155). In 74 of these cases the cat was euthanased with a suspected diagnosis of FD based on the CSS. 51/78 cases were classified by the CSS as positive for FD, 23/78 as probable, and 4/78 as inconclusive. A necropsy was performed in 41/83 cases that died. In 5/41 cases no samples were submitted for histopathology. In 3/41 cases submitted tissue did not contain autonomic ganglia. In the remaining 33 cases FD was histopathologically confirmed in 26 cases and excluded in 7 cases. The association between clinical findings and a FD diagnosis in these 33 cases were assessed by binary logistic regression. No aspect of signalment (age, breed, sex), individual clinical signs, radiographic findings, or the proposed CSS, was significantly associated with a histopathology diagnosis of FD. The median CSS score for FD cases was 11 of a total of 16 and for non‐FD cases it was 10. The clinical signs which were most strongly associated with FD were bradycardia of less than 120 bpm (odds ratio 10.2 [95%CI 0.72, 145]) and dysuria or bladder atony (OR 4.8 [0.46, 50]). OPTs were performed in only a single non‐FD case; both tests were positive. In FD cases the tests were negative in 5/13 (pilocarpine) and 1/5 (epinephrine). These findings suggest that the proposed CSS and OPTs are not reliable means for ante‐mortem diagnosis of FD. It is also possible that FD is not a single condition but a syndrome with several causes with varying histopathology findings. Additional tests such as resting heart rate variability and heart rate response to intravenous atropine may prove beneficial in the diagnosis of FD. N13 UTILITY OF VON FREY ANESTHESIOMETRY TO ASSESS RECOVERY OF SENSORY FUNCTION AFTER THORACOLUMBAR INTERVERTEBRAL DISC EXTRUSION IN DOGS Rachel Song 1, Michele Basso1, Ronaldo da Costa1, Leslie Fisher1, Xiaokui Mo1, Sarah Moore1 1The Ohio State University, Columbus, OH, USA Von Frey anesthesiometry (VFA) provides objective sensory threshold (ST) measurements and can be utilized to monitor recovery from spinal cord injury (SCI). The purpose of this study was to determine the utility of VFA in SCI‐affected dogs in quantitating recovery of sensory function over time when compared to normal dogs. ST with an electric VFA were determined in all four limbs of 20 normal dogs and 29 dogs with acute thoracolumbar SCI caused by spontaneous intervertebral disc extrusion. ST was assessed at three separate time points in normal dogs and on day 3, 10 and 30 following decompressive surgery in dogs with SCI. ST values were compared between groups and correlated with locomotor recovery in SCI‐affected dogs using the Olby spinal cord injury scale. ST was significantly higher (consistent with hypoalgesia) in the pelvic limbs of SCI‐affected dogs at days 3, 10 and 30 post‐operatively when compared to normal dogs (p < 0.05). No significant difference in thoracic limb ST values was observed between groups. A progressive decrease in pelvic limb ST values occurred in SCI‐affected dogs over time, consistent with improvement toward normal sensation or development of allondynia. This finding correlated inversely with locomotor scores (p < 0.05, r range ‐0.44 to ‐0.63). A decline in thoracic limb ST was also observed across testing sessions on both normal and SCI‐affected dogs. This may be the result of patient acclimation, operator training effect, or in SCI‐affected dogs an effect of analgesic medications. This study supports the use of VFA to assess differences in ST between normal and SCI‐affected dogs. However, future studies must focus on techniques to minimize or compensate for clinical, environmental and behavioral factors that may impact measurement of ST values in the clinical setting. N14 ENDOSCOPIC LAMINECTOMY TECHNIQUE IN DOGS WITH HANSEN TYPE I DISC DISEASE Natalya Ulanova 1,2, Andrei Tarassov1,2 1Veterinary clinic Best, Novosibirsk, Russia, 2Southeast Valley Veterinary Hospital, Sandy, UT, USA  Endoscopic laminectomy technique is widely used in humane medicine, but not commonly used in veterinary medicine. We performed 15 surgeries between January 2012 to December 2014 at our hospital using endoscopic approach for treatment of Hansen type I intervertebral disc disease. All dogs were presented with clinical signs of acute paraparesis, duration from several hours to several days. 7 dogs were Dachshunds, 3 French Bulldogs, 2 Pekingese, 1 English Cocker Spaniel, 1 Russian Spaniel and 1 mixed breed dog. One Dachshund had absent deep pain for 12 hours prior to presentation to our hospital. Age of the dogs was between 2 and 9 years. After the neurologic examination 10 dogs had neurolocalization of T3‐L3 spinal segments and 4 dogs had neurolocalization of L4‐S1. All patients had diagnosis confirmed by MRI, using 0.3T MRI. Surgical technique: We used Karl Storz endoscopic equipment and Karl Storz neurosurgical system EasyGo to perform the surgery. Needle was placed near the spinal process to mark correct intervertebral space and placement was confirmed with a lateral view X‐ray. Patient was placed in ventral recumbency and surgical field was aseptically prepared. After confirmation of the disc space, we made 19 or 23 mm skin incision, located spinal process of the vertebrae and removed it with Luer Rongeur. Puncture needle was installed at 45 degree to the base of removed spinal process and then sequential dilators are used for bloodless access to the bone. After surgical field is exposed, we installed a throacar with 360 degree rotating head. The throacar and the rotating head are used allow endoscope movement down to the spinal canal for the surgical field visualization. From this point on, the surgery is guided by continuous endoscopic vision. Remnant muscle tissues are removed with forceps and bleeding is stopped with bipolar electrocautery. Shaver is used for drilling the bone, with irrigator and suction used to remove bone particles. Kerrison's rounger is used to widen the entrance into the spinal canal. Extruded disc material is grasped with a forceps and removed from the spinal canal. Spinal cord is evaluated via endoscopic visualization. Two adjacent disc spaces were fenestrated before closure of the surgical wound. All dogs became ambulatory after surgery, 9 dogs return to normal gait between day 2 and day 9 post operatively. At 3 months follow up 12 patients did not have any neurological deficits. Described technique provides great benefits for the surgeon as well as for the patient. In comparison to standard hemilaminectomy, 16 fold magnification allows superior visualization of the surgical field and spinal cord and nerve roots. Improved differentiation of the anatomic structures results in gentler manipulation of the nervous structures and reduces risks of bleeding. Two adjacent disc spaces can be fenestrated via the same incision. Dilation system and smaller incision reduce muscle trauma and postsurgical pain. Owners are very pleased to see a smaller incision. N15 AN INTERNET SURVEY OF THE APPROACH TO NEUROSURGICAL EMERGENCIES BY ACVECC DIPLOMATES Elizabeth Parsley 1, Elizabeth Rozanski1, Katherine Weiss1 1Tufts University, North Grafton, MA, USA Neurological disease, including paresis and paralysis are commonly evaluated on an emergency basis in small animal medicine. While prompt imaging and decompression is warranted, there is not a consensus on when dogs that present with urgent signs should be addressed. The purpose of this study was to evaluate the off‐hour diagnostic and surgical approach to dogs with evidence of spinal cord compression. An internet‐based survey was distributed to the ACVECC listserve. The survey consisted of 17 questions, including 9 demographic questions, 7 scenarios and one question regarding myelomalacia. In the scenarios, the respondents were asked to respond with their current approach at their hospital. While the survey was anonymous, respondents were requested to limit the responses to one per hospital. 120 responses were received. The majority (73%) of respondents worked in private specialty practice, with 27% working in academia. 60% of the respondents had 2‐4 ACVECC diplomates in their practices, although 25% were the only diplomate, and 4% had more than 6. Out of hours, 86% were able to perform myelograms, 85% Computed tomography (CT) scans, and 48% Magnetic resonance imaging (MRI). Both during business hours and off‐hours, neurologist sand neurology residents, and surgeons and surgery residents operate the majority of the cases on a shared basis. 33% of practices have no neurologist on staff; in the remaining practices with neurologists, 21% of the time a board‐certified neurologist (BCN) evaluates case out of hours at least 50% of the time, 79% of the time a BCN does not. Of practices with veterinary surgeons, 25% of the time, the case is evaluated out of hours. Out of‐hours imaging is read by the radiologist or radiology resident in 54% of cases, and by a teleradiology service in 34%. 43% of practices operated at least 10 dogs per month with spinal cord compression. The potential for myelomalacia was discussed in for all dogs by 38% of respondents and only dogs with no deep pain sensation by 25% of respondents. In the scenario response section, given a dachshund presenting at 6:45 pm with no motor but deep pain sensation, 91% of respondents would proceed with imaging and surgery that night. In the same dog arriving at 1:30 am, 57% of respondents would wait until am. In a dachshund arriving at 6:45 pm with no deep pain acutely, 95% would proceeding immediately with imaging and surgery, while at 11:45 pm, 75% would proceed to imaging and surgery, while 20% would wait until am, and 5% would recommend euthanasia. In a dachshund arriving without deep pain at 3:15 pm with 40 hours since last known to be ambulatory, 86% would recommend immediate imaging and surgery. In a tetraparetic Rottweiler, with increased PCO2, presenting on a Saturday evening, 82% of respondents would proceed with imaging and surgery that night, with 17% waiting until Sunday am, and 1% waiting until Monday. In a Labrador, presenting with lateralizing T3‐L3 signs after chasing a ball, at noon on a holiday, 48% of respondents would pursue MRI immediately, while 22% would wait until the following day. Most respondents commented that their approach was directed by the availability and wishes of facility neurologists and surgeons. There was no apparent differences between university and private practices. Urgent neurosurgical emergencies are common and current policies appear to be more based by availability and preference, than clearly guided by evidence. N16 NEUROPROTECTIVE EFFECTS OF RILUZOLE AND DANTROLE ASSOCIATION ON ACUTE SPINAL CORD INJURY Bernardo Martins 2, Bruno Torres1, Eliane Melo2 1Federal Rural University of Pernambuco, Garanhuns, Pernambuco, Brazil, 2Federal Rural University of Minas Gerais, Belo Horizonte, Minas Gerais, Brazil Riluzole and dantrolene has been shown to be neuroprotective by reducing neuronal apoptosis after brain and spinal cord injury in several animal models of neurological disorders. In this study, we investigated the effects of the association of riluzole and dantrolene on experimental spinal cord injury (SCI). Twenty‐nine Wistar rats were laminectomized at T12 and divided in five groups: GI (n = 6) underwent laminectomy alone and was treated with placebo. GII (n = 6) underwent laminectomy followed by SCI and was treated with placebo. GIII (n = 5) underwent laminectomy followed by SCI and was treated with riluzole and placebo 15 min and 1 h after laminectomy, respectively. GIV (n = 6) underwent laminectomy followed by SCI and was treated with placebo and dantrolene 15 min and 1 h after laminectomy, respectively. GV (n = 6) underwent laminectomy followed by SCI and was treated with riluzole and dantrolene 15 min and 1 h after laminectomy, respectively. A compressive trauma was performed to induce SCI. After euthanasia, the spinal cord was evaluated using light microscopy, TUNEL staining and immunochemistry with anti‐NeuN. All procedures were performed according to the principles adopted by the NIH Guide for the Care and Use of Laboratory Animals and by the Ethics Committee on Animal Use (CEUA‐UFMG – protocol number 186/2010). Animals treated with the association of riluzole and dantrolene showed a larger number of NeuN‐positive neurons adjacent to the epicenter of injury (P ≤ 0,05). Furthermore, the TUNEL staining was similar between animals treated with riluzole and dantrole and those that did not receive spinal cord trauma (P > 0,05). These results showed that riluzole and dantrolene protects spinal cord tissue after traumatic SCI by decreasing apoptotic cell death, especially when used in association. N17 COMPARISON OF CLINICAL OUTCOME WITH ZONISAMIDE, LEVETIRACETAM OR PHENOBARBITAL MONOTHERAPY IN DOGS WITH IDIOPATHIC EPILEPSY Rennie Waldron 1, Karen Munana1, Julie Nettifee‐Osborne1 1North Carolina State University College of Veterinary Medicine, Raleigh, NC, USA Zonisamide (ZNS) and levetiracetam (LEV) are utilized with increasing frequency as first‐line anticonvulsant medications in dogs with epilepsy, but little information is available regarding their use. The purpose of this study was to evaluate clinical outcome in idiopathic epileptic dogs administered ZON or LEV monotherapy, and compare this to dogs on phenobarbital (PB) monotherapy. Medical records from 2003‐2013 were retrospectively searched for dogs diagnosed with idiopathic epilepsy administered either ZON, LEV or PB monotherapy. Information on duration of therapy, adverse effects, and outcome was collected from medical records, referring veterinarians, and owner questionnaire. A total of 121 cases were included; 16 dogs on LEV, 40 dogs on ZON, and 65 dogs on PB. Dogs in the PB group were younger at the time of the first seizure compared to the LEV and ZON groups (p = 0.05). Mean duration of therapy was similar between groups (437, 583, and 488 days for LEV, ZON, and PB, respectively). For dogs alive at the time of writing, the PB group was more likely to still be on monotherapy than the LEV or ZON groups (p = 0.04). Seizure frequency was the most common reason for monotherapy failure in all groups. The incidence of adverse effects was higher in the PB group (78%) compared to the LEV (17%) or ZON groups (36%) (p = 0.0001), although the incidence of sedation was similar. These results suggest that when compared to PB, ZNS and LEV monotherapy are associated with fewer adverse effects in dogs, although add‐on therapy might be required sooner. N18 EFFICACY OF MIDAZOLAM CONTINUOUS RATE INFUSION FOR EMERGENT SEIZURE THERAPY Kathryn Bray 1, Christopher L. Mariani1, Peter J. Early1, Melissa Lewis1, Karen R. Munana1, Natasha J. Olby1, Rennie J. Waldron1 1North Carolina State University, Raleigh, NC, USA The purpose of this retrospective study was to describe the use and efficacy of a midazolam continuous rate infusion (CRI) in dogs with cluster seizures (CS) or status epilepticus (SE). The medical records of dogs admitted for emergent seizures that received a midazolam CRI were reviewed. The signalment, seizure history, results of the neurological examination, diagnosis, and initial therapy were recorded. For midazolam CRI the initial dose, timeframe of dose increases and breakthrough seizures, highest dose given, total time received, total time hospitalized, and survival to discharge were documented. A total of 103 dogs were included in the study. There were 85 dogs with CS and 18 dogs with SE; of these, 64 dogs had a previous history of seizures. The dogs were divided into genetic epilepsy (39), structural epilepsy (43), unknown epilepsy (15), and reactive seizures (6). The median dose of midazolam CRI that resulted in seizure resolution was 0.3 mg/kg/hr. The median time to seizure resolution was 9 hours; in 52 dogs seizure resolution was achieved upon starting the CRI. The mean time of midazolam infusion was 28.9 hours, with a mean length of hospitalization of 3 days. Overall, 79 dogs achieved seizure resolution with midazolam CRI. This represented 84.6% of dogs with genetic epilepsy, 74.4% with structural epilepsy, 66.7% with unknown epilepsy, and 83.3% with reactive seizures. 75 dogs (72.8%) survived to discharge; of the non‐responders, 46% had structural epilepsy. Midazolam CRI appears to be a safe and effective therapy in dogs with CS or SE. N19 MRI CHARACTERISTICS OF ATLANTO‐AXIAL SUBLUXATION IN 42 DOGS Kathryn Bray 1, Simon R. Platt2, Marc Kent2, Allison Haley2, Jill Narak3, Amy W. Hodshon4, Natasha J. Olby1 Shannon P. Holmes2 1North Carolina State University, Raleigh, NC, USA, 2University of Georgia, Athens, GA, USA, 3Auburn University, Auburn, AL, USA, 4University of Tennessee, Knoxville, TN, USA The purpose of this retrospective study was to describe the magnetic resonance imaging (MRI) characteristics of canine atlanto‐axial (AA) subluxation and to identify associations with neurologic deficits. A multicenter review of the medical records and MR images of dogs with a diagnosis of AA subluxation was conducted. Signalment, neurologic grade, time to onset of signs, and MRI characteristics were recorded. The evaluated MRI characteristics included (1) spinal cord compression, (2) degree of joint subluxation, (3) integrity of odontoid ligamentous structures, (4) presence of a dens, (5) presence of cord signal intensity, and (6) presence of syringohydromyelia, hydrocephalus, and Chiari‐like malformation. A control population of age and breed matched dogs with normal AA junctions were also evaluated. Reviewers were blinded to affected status. MR images of 42 dogs affected with AA subluxation were reviewed and compared to 26 control dogs. The majority of affected dogs were Yorkshire Terriers (47.5%). The median age was 27 months. The median weight was 2.7 kg. The control population had a median age of 96 months, and a median weight of 4.7 kg; the most common breeds were Maltese (18.5%) and Pomeranian (18.5%). 28% of dogs were affected less than 1 week, 30% were affected between one week and one month, and 42% were affected for greater than 1 month. Neurologic grade ranged from 1 (spinal pain only) to 5 (tetraplegic without nociception), but 50% of affected dogs were ambulatory tetraparetic (grade 2). Associations of MRI characteristics with affected status of dog (AA subluxation or control) and neurologic grade were assessed by logistic regression. Associations of MRI characteristics with each other was performed using Spearman and Pearson correlations, student t‐tests, analysis of variance and Kruskal‐Wallis tests as appropriate. No differences were noted between the affected and control dogs with respect to the presence of syringohydromyelia, hydrocephalus and Chiari‐like malformation. Significant differences between control and affected dogs were noted with respect to presence of a dens, apical ligament, and transverse ligament. The presence of cord signal intensity was associated with affected status, as was increased AA joint cavity size and percentage cross‐sectional cord compression at the level of the dens and mid‐body C2. No associations were found between cord compression, appearance of the dens or cord signal intensity and neurologic grade of affected dogs. There was an association between the presence of cord signal intensity and increased joint cavity size. There was no association between dens appearance and cord signal intensity. Dens abnormalities were associated with greater subluxation distance and odontoid ligament absence was associated with increased joint cavity size. Dogs affected by AA subluxation do not have a higher incidence of Chiari‐like malformation, syringohydromyelia, or hydrocephalus than control dogs, and their neurologic grade is not associated with MRI findings. However, lack of a dens and/or odontoid ligaments are associated with larger subluxations. N20 DETERMINATION OF IDEAL IMPLANT PLACEMENT IN CANINE ATLANTOAXIAL VERTEBRAE – DESCRIPTION OF A NEW METHOD USING OSIRIX Guillaume Leblond 1, Luis Gaitero1, Noel Moens1, Alex Zur Linden1, Fiona James1, Gabrielle Monteith1, John Runciman2 1Ontario Veterinary College, University of Guelph, Guelph, Ontario, Canada, 2School of Engineering, University of Guelph, Guelph, Ontario, Canada The main objectives of this study were 1‐ to propose an objective method for the determination of ideal implant placement in the atlantoaxial joint, 2‐ to develop a mathematical model allowing calculation of implants 3D position and 3‐ to assess the validity of the mathematical model. Current literature on atlantoaxial anatomy and ventral atlantoaxial stabilization techniques was reviewed to identify safe bone corridors. Each bone corridor was geometrically simplified and ideal implant placements defined as axes centered within the simplified corridor. CT images of the craniocervical junction were obtained in 27 dogs and each previously defined ideal implant was positioned using OsiriX. For each implant an insertion point, an exit point and 4 safety margin points were localized. Mathematical equations calculating 2 projected angles (ProjA) and 4 safety angles (SafA) from those point coordinates were determined. A sample of 12 simulated implants was randomly selected and each angle was mathematically calculated and compared to direct measurements obtained with OsiriX measurement tools (2 observers x 2 repeats). 9 bone corridors and 13 ideal implant placements could be objectively defined. OsiriX could be used successfully to simulate the ideal implant positions in all cases. There was excellent agreement between the calculated and measured values for both ProjA (ρc=0.9992) and SafA (ρc=0.9997). Absolute differences between calculated and measured values were respectively [ProjA=0.44 ± 0.53°; SafA=0.27 ± 0.25°] and [ProjA=0.26 ± 0.21°; SafA=0.18 ± 0.18°], for each observer. A new method for determination of ideal implant placement is provided. Semi‐automated calculation of ideal implant 3D position could be developed using this method. N21 CLINICAL EFFICACY AND SAFETY OF ZONISAMIDE MONOTHERAPY IN DOGS WITH NEWLY DIAGNOSED IDIOPATHIC EPILEPSY Akinori Nomura 1, Miyoko Saito2, Daisuke Hasegawa3, Naoyuki Watanabe4, Keiko Uchida5, Seiichi Okuno6, Masahiro Nakai1, Kensuke Orito2 1DS Pharma Animal Health Co., Ltd., Osaka, Japan, 2Azabu University, kanagawa, Japan, 3Nippon Veterinary and Life Science University, Tokyo, Japan, 4Watanabe Animal Hospital, Shizuoka, Japan, 5AC‐plaza Kariya Animal Hospital, Tokyo, Japan, 6Animal Clinic Kobayashi, Saitama, Japan Zonisamide is an antiepileptic medication and is used for canine and feline epilepsy. Information on the clinical efficacy as monotherapy is, however, limited. In this study, anticonvulsant efficacy and safety of zonisamide (Consave®) were evaluated in a multicenter open‐label uncontrolled study in dogs with newly diagnosed idiopathic epilepsy. Fifty‐seven dogs were enrolled. All were antiepileptic drug‐naive and had > 2 seizures per 12 weeks. 2.7 to 14.4 mg/kg zonisamide was administered orally twice daily. Data in a 12‐week maintenance period were compared with those in a 4 to 12‐week historical baseline period. Four dogs were excluded from the efficacy evaluation because one was finally diagnosed with symptomatic epilepsy and the others dropped out in a titration period. Forty of 53 dogs (76%) were judged as responders with over 50% reduction in the seizure frequency. In > 90% of responders, the average dosage of zonisamide was 4.8 (2.7 to 8.6) mg/kg and the average plasma concentration was 18.9 (8.0 to 48.0) μg/mL. Twenty‐nine dogs became seizure free. In 22 of 57 dogs, anorexia, weight loss, reduced activity, vomiting, diarrhea, constipation, erythema or itching was observed. All these adverse events were not severe and were disappeared during or immediately after the maintenance period. This study demonstrated that zonisamide monotherapy is effective and well‐tolerated in dogs with newly‐diagnosed idiopathic epilepsy. N22 CANINE MENINGIOMA: COMPARISON OF PALLIATIVE THERAPY, SURGERY AND STEREOTACTIC RADIOSURGERY Mario Dolera 1, Luca Malfassi1, Silvia Marcarini1, Simone Pavesi1, Nancy Carrara1, Sara Finesso1, Giovanni Mazza1, Massimo Sala1 1La Cittadina Fondazione Studi e Ricerche Veterinarie, Romanengo, Italy Meningiomas represent about half of primary intracranial tumours in dogs. There are limited comparative studies regarding the various treatment modalities. The aim of this study was to compare palliative therapy, surgery and stereotactic radiosurgery in canine meningioma. The data was collected retrospectively from 198 dogs referred to one institution over a 15‐year period with histopatologically confirmed or MRI consistent with meningioma. Dogs were grouped by anatomical site (supratentorial‐E, infratentorial‐T, spinal‐S) and by therapeutic option (palliation‐P, surgery‐S, radiosurgery‐R). The surgery goal was a total resection of tumour. LINAC based VMAT radiosurgery was performed in 1‐5 fractions. Serial clinical and MRI examinations were conducted. Signalment, clinical signs, neuroanatomical tumour location, relapse specifics, adverse events, best response and overall survival (OS) time were evaluated. The OS estimates were calculated using the Kaplan‐Meier method and the differences between compared groups using log‐rank analysis. Multivariate analysis was performed using Cox regression. 91 dogs (51 E, 33 T, 7 S) had been palliated, 69 dogs (33 E, 31 T, 5 S) had been treated with stereotactic radio‐surgery, 38 dogs (32 E, 1 T, 5 S) with surgery. OS in PE was 190 days, in PT 38 days, in PS 89 days, in RE 781 days, in RT 654 days, in RS 813 days, in SE 567 days, in ST 3 days, in SS 210 days. The predictive variables of OS are localisation and therapy option. The dogs suffering from meningioma undergoing stereotactic radiosurgery had superior results comparing with those treated with surgery or palliation. N23 SUSPECTED HYDRATED NUCLEUS PULPOSUS EXTRUSIONS ASSOCIATED WITH THE DORSAL LONGITUDINAL LIGAMENT IN DOGS Mario Dolera 1, Luca Malfassi1, Giovanni Mazza1, Sara Finesso1, Massimo Sala1, Silvia Marcarini1, Simone Pavesi1, Nancy Carrara1 1La Cittadina Fondazione Studi e Ricerche Veterinarie, Romanengo CR, Italy The aim of this study was to describe Magnetic Resonance Imaging (MRI), surgical and cytological findings in patients with hydrated nucleus pulposus (HNPE) cervical disc extrusions associated with the dorsal longitudinal ligament. A prospective study with the following inclusion criteria was conducted: high‐field MRI examination, videotaped microsurgical decompression, and cytological examination of the removed material. The following MRI features were assessed: grading of spinal cord compression, intervertebral spaces involved, signal intensity and distribution of the material, T2‐W spinal cord signal, thickness and signal intensity of the discs. Surgical findings considered were: macroscopic appearance of the material, its localization and relationship with the DLL. Thirty‐six patients were included. The anatomical site was C4‐C5 in 88%, C3‐C4 in 12% of cases. On MRI, lesions had signal intensity similar to the CSF in both T1‐W and T2‐W sequences. The distribution was central in all cases. The thickness of the corresponding discs was slightly reduced in 75% of cases. Signal intensity of the nucleus pulposus was normal in 82% of cases, and slightly reduced in T2‐W sequences in 18%. Spinal cord compression ranged from 13.5% to 57.2%. At surgery the material was found to be intraligamentous and of liquid consistence with a gelatinous component in 42% of cases, plain liquid in 33%, liquid with frustules in 25%. Cytological examination allowed to exclude the presence of inflammation, bacteria, mycosis, neoplasia or foreign material. In HPNE diagnosed dogs, extruded disc material of intraligamentous localization was observed. Further studies are needed to clarify the pathophysiology. N24 THE USE OF MAGNETIC RESONANCE SPECTROSCOPY IN DIFFERENTIAL DIAGNOSIS OF CANINE BRAIN DISEASES Luca Malfassi 1, Mario Dolera1, Giovanni Mazza1, Silvia Marcarini1, Massimo Sala1, Nancy Carrara1, Sara Finesso1 1La Cittadina Fondazione Studi e Ricerche Veterinarie, Romanengo CR, Italy The aim of this study was to investigate Magnetic Resonance Spectroscopy (MRS) spectra of canine brain diseases. From 2008 to 2014 578 patients underwent brain MRI and MRS. Patients with intracranial lesions were grouped into inflammatory (IG), neoplastic (NG) and hepatic encephalopathy (HE) affected groups. MRS was performed twice, once using a short TR35 and once using a long TR144. A 1x1x1 cm VOI was used. The studied metabolites were Glutamine‐glutamate (Glx), N‐acetylaspartate and Nacetyl aspartyl glutamate (NAA), Choline derivatives (Cho), Myo‐Inositol (My). The NAA/Cr, Cho/Cr and My/Cr ratios were analyzed as the presence of lipids and lactate. Out of 578 patients: 112 IG, 144 NG, 22 with HE. MRS spectra from IG showed increased Cho (mean Cho/Cr: 2.23), reduced NAA (mean NAA/Cr: 1.15) and the presence of the lipid peak was found in 46/112. Mean values and relative standard deviations were: Glx/Cr 0.26 std. 0.2; Cho/Cr 2.23 std. 0.9; My/Cr 0.54 std. 0.1; NAA/Cr 1.35 std. 0.59. MRS spectra from NG were characterized for high values of Cho and low or absent NAA. The presence of lipids and lactate was found in 31/144. Mean values were: Glx/Cr 0.22 std. 0.32; Cho/Cr 3.27 std. 0.69; My/Cr 0.54 std. 0.11; NAA/Cr 1.14 std. 1.03. The HE displayed decreased values of My, Cho and NAA and increased Glx. Mean values were: Glx/Cr 0.40 std. 0.13; Cho/Cr 1.14 std. 0.34; My/Cr 0.15 std. 0.08; NAA/Cr 1.02 std. 0.35. In vivo single‐voxel proton MRS provides a sensible and non‐invasive characterization of canine brain lesions. N25 DEVELOPMENT OF F‐RESPONSE METHOD USING MULTIPLE‐POINT STIMULATION FOR MOTOR UNIT NUMBER ESTIMATION TECHNIQUE IN NORMAL DOGS Go Togawa 1, Miyoko Saito1, Yukihiro Fujita1, Akikazu Ishihara1, Ryota Watanabe1 1Azabu University, Sagamihara, Kanagawa, Japan Motor unit number estimation (MUNE) is an electrophysiological technique for quantifying the number of motor units, and can be a valuable biomarker in neurodegenerative diseases such as canine degenerative myelopathy. MUNE is determined by dividing the maximum compound muscle action potential by average single motor unit potential (SMUP). In dogs, there has been only one study reporting MUNE, in which an incremental method was used (L Vasquez, 2010). This method requires very small successive increments of stimuli to collect SMUP; however, not all stimulators have such a function. SMUP is also collected from the F‐response, and MUNE using the F‐response method does not require small stimulus increments. The F‐response method is also free from alternation, which is a major problem with the incremental method. This study evaluated MUNE using the F‐response method (D Stashuk, 1994) but the nerve was stimulated at multiple points (MPS‐F method) along the deep peroneal nerve innervating the extensor digitorum brevis muscle. Two trial sets were performed to examine the test‐retest reliability in six healthy dogs. MUNE was obtained from all dogs. Multiple‐point was better than single‐site stimulation regarding the ability to obtain more SMUP numbers. MUNE numbers in the first trial were 77, 68 (left, right) and those in the second were 65, 65. There were no significant differences in MUNE values between them. Intraclass correlation coefficients between the first and second trials were 0.73, 0.81 (left, right). This study indicates that the MPS‐F method can be useful for the MUNE technique in dogs. O01 PHASE I LEAD‐IN AND SUBSEQUENT RANDOMIZED TRIAL ASSESSING SAFETY AND MODULATION OF REGULATORY T CELL NUMBERS FOLLOWING A MAXIMALLY TOLERATED DOSE DOXORUBICIN AND METRONOMIC DOSE CYCLOPHOSPHAMIDE COMBINATION CHEMOTHERAPY PROTOCOL IN TUMOR‐BEARING DOGS RM Rasmussen 1, ID Kurzman1, B Biller2, DM Vail2 1School of Veterinary Medicine, University of Wisconsin‐Madison, Madison, WI, USA, 2Flint Animal Cancer Center, Colorado State University, Fort Collins, CO, USA Introduction: Two paradigms of chemotherapy dosing schemes exist; maximally tolerated dose (MTD) and metronomic dosing. This study establishes the safety of combining metronomic cyclophosphamide (CTX) with MTD doxorubicin (DOX) and the effect of the combination on regulatory T cells (Treg). Methods: Thirteen tumor‐bearing dogs were enrolled in a 3+3 dose escalation phase I study of DOX (30 mg/m2, q3wk) and dose‐escalating CTX cohorts up to 15 mg/m2, PO, daily. Subsequently, 16 dogs were randomized to receive MTD‐DOX alone or in combination with metronomic CTX. Peripheral blood mononuclear cells (PBMC) were collected before and at 7 and 21 days after beginning therapy for analysis of Treg numbers. Results: All phase I dosing cohorts were well tolerated and the DOX (30 mg/m2 q3wk)/CTX (15 mg/m2/d) combination was deemed safe. The subsequent 16 randomized dogs tolerated the combination without significant toxicity. No significant difference was observed in relative and absolute Treg numbers between dogs receiving DOX vs dogs receiving DOX/CTX. However, DOX alone and in combination with CTX resulted in significant decreases in absolute lymphocyte counts (p=0.0007) and a trend towards decreasing absolute Treg numbers. Conclusion: Metronomic CTX at published dosages can be safely given concurrent with MTD‐DOX in tumor‐bearing dogs. It is likely that any difference in Treg numbers resulting from the addition of metronomic CTX to MTD‐DOX chemotherapy is not observable due to lymphocyte depletion caused by DOX alone in this small cohort. (VCS Award Winner) O02 INHIBITION OF HEDGEHOG SIGNALING INHIBITS PROLIFERATION IN CANINE TRANSITIONAL CELL CARCINOMA Tanya Gustafson 1, Barbara E. Kitchell1, Barbara J. Rose1, Barbara Biller1 1Colorado State University, Fort Collins, CO, USA Background: Transitional cell carcinoma (TCC) is the most commonly diagnosed tumor of the canine urinary system. Overall survival with traditional chemotherapy in this disease has not improved for many years, and so additional research into new therapeutic targets is needed. Hedgehog signaling, which regulates normal embryonic development, represents one such target. When activated in adult cells, Hedgehog signaling promotes oncogenesis and has been found to play a central role in human bladder cancer. Methods: In this study, TCC cell lines highly expressing Hedgehog pathway mediators were treated with the inhibitors cyclopamine and GANT61. Down‐regulation of the pathway was confirmed by real time RT‐PCR for Patched 1 (PTCH1), GLI1 and GLI2. Cell growth was assessed at 48 and 96 hours after inhibitor treatment. Cell viability was assessed through the Alamar Blue assay. Apoptosis was assessed through Annexin V staining. Results: Hedgehog pathway mediators, including IHH, GLI1, GLI2 and PTCH1, were found to be expressed in 5 canine transitional cell carcinoma cell lines. Indian Hedgehog was expressed in 5‐25% of tumor cells in 5 canine bladder tumor tissues, but not in normal canine bladder tissue. Inhibition of Hedgehog signaling with cyclopamine and GANT61, confirmed by decreased expression of downstream target genes, led to significantly decreased cell proliferation but had a lesser effect on apoptosis. Conclusions: This study suggests that Hedgehog signaling does have a role in the pathogenesis of canine TCC. Further studies using silencing RNA techniques to inhibit signaling alone and in combination with chemotherapy may provide additional evidence of the importance of this pathway in canine TCC and provide the basis for future clinical trials. (VCS Award Winner): O03 ADJUVANT α‐TOCOPHEROL & IMPROVED ANTICANCER EFFICACY IN AN IMMUNE COMPETENT PRE‐CLINICAL MODEL Michelle Beehler 1, Shannon Pease1, Alysha Hartman1, Christiane Loehr1, Debbie Mustacich1 1Linus Pauling Institute, College Of Veterinary Medicine, Oregon State University, Corvallis, OR, USA Platinum compounds, including cisplatin (CDDP), are highly active anticancer agents for use against solid tumors. Like most chemotherapeutics, platinum compounds are toxic to non‐tumor tissues and development of unwanted side effects, including neuropathy and nephrotoxicity, often limits treatment dose and duration. Mechanisms of these side effects, particularly neurotoxicity, are undefined. However, accumulation of platinum in spinal cord tissues is associated with development of neuropathy. Remarkably, the clinical and histologic features of CDDP‐induced neurotoxicity are similar to those seen in vitamin E (αT) deficiency. Using a preclinical model of epithelial ovarian cancer we tested the hypothesis that adjuvant αT would improve CDDP antitumor efficacy while reducing CDDP‐induced oxidative stress and non‐tumor toxicity. Tumor‐bearing immune competent female rats received one of 4 treatment regimens: (1) saline (2X/week), (2) αT (2X/week), (3) CDDP (2X/week), or (4) CDDP + αT (2X/week). All regimens were administered for a period of 4.5 weeks. At necropsy, tumor burden was reduced to undetectable levels in 10/16 CDDP + αT treated rats compared to 2/16 rats treated with CDDP alone. CDDP decreased indices of tumor proliferation (measured by Ki‐67 immunohistochemistry) by 28% (p < 0.05), while CDDP + αT decreased tumor proliferation by >75% (p < 0.01), compared to saline. Importantly, adjuvant αT decreased accumulation of platinum in spinal cord tissue by >30% (p < 0.05) without altering tumor platinum levels, compared to rats treated with CDDP alone. Interestingly, the presence of tumors in saline treated rats increased plasma F2‐isoprostane and F2‐isoprostane metabolite levels 40%, compared to non‐tumor controls; αT (with or without CDDP) returned these levels to those of non‐tumor controls. Finally, adjuvant αT prevented CDDP‐induced plasma and tissue αT depletion. These data are the first to document the ability of adjuvant αT to (1) improve chemotherapeutic antitumor efficacy, (2) decrease tumor proliferation, and (3) improve antioxidant/oxidative stress status in an immune competent tumor‐bearing animal. Further elucidation of the ability of adjuvant αT to improve antitumor efficacy and quality of life outcomes is essential for development of improved treatment regimens for both human and veterinary patients. O04 METRONOMIC ANTIANGIOGENIC THERAPY AFTER AMPUTATION AND SURAMIN/DOXORUBICIN CHEMOTHERAPY DOES NOT IMPROVE THE DISEASE FREE INTERVAL OR SURVIVAL IN RETIRED RACING GREYHOUNDS WITH OSTEOSARCOMA Francisco Clemente‐Vicario 1, Duncan Russell2, Guillermo Couto3 1The Ohio State University, Columbus, OH, USA, 2Oregon State University, Corballis, OR, USA, 3Couto Veterinary Consultant, Hilliard, OH, USA During the last 20 years, the survival times in dogs with osteosarcoma (OSA) treated with surgery and postoperative adjuvant chemotherapy has not changed. Therefore, new treatment approaches are needed. The antiangiogenic chemotherapy hypothesis proposes that tumors cannot growth more tan a few millimeters without recruiting blood vessels that will supply oxygen and nutrients to the neoplastic cells. One of the major players in the new blood vessel growth is vascular endothelial growth factor (VEGF). Dogs with aggressive tumors, such as OSA, have higher circulating activity of VEGF compared with dogs with benign lesions and it has been proposed that VEGF plays a role in the development of metastasis. Former racing Greyhounds presented to The Ohio State University Veterinary Medical Center with appendicular OSA and no evidence of metastases on thoracic radiographs, were treated by amputation of the affected limb and chemotherapy, or amputation and chemotherapy followed by metronomic chemotherapy. Prior to initiating chemotherapy, all dogs underwent a physical examination, complete blood count (CBC), serum chemistry profile, urinalysis, and 3‐view thoracic radiographs. All dogs were treated with a protocol of suramin/doxorubicin once every 2 weeks for 5 treatments. Prior to each treatment, a physical examination and a CBC were repeated; thoracic radiographs were repeated before the fifth treatment, and every three months thereafter. Starting in June 2010, dog owners were given the option to use a metronomic chemotherapy protocol after completing the suramin/doxorubicin treatment. The metronomic chemotherapy protocol consisted of artemisinin (Holley Pharmaceuticals, Tustin, CA), 100 mg PO q24 h; toceranib phosphate (Palladia, Zoetis, Florham Park, NJ), 2.5 mg/kg PO on Mondays, Wednesdays, and Fridays; and cyclophosphamide, 10 mg/m2 PO on Tuesdays, Thursdays, and Saturdays. Dogs were on this protocol until development of metastases or death. A CBC, serum chemistry profile, urinalysis and thoracic radiographs were done at each three months visit. Survival time (ST) and disease free interval (DFI) were generated using the Kaplan‐Meier method and compared using log rank analysis. A P value of less than 0.05 was considered significant. There were 71 Greyhounds treated; 19 dogs received metronomic chemotherapy after completing standard chemotherapy, and 52 received only suramin/doxorubicin protocol. Three dogs still alive and the one that died due to an unrelated cause were censored for the DFI and ST analyses. There was no significant difference in DFI or ST when dogs treated with chemotherapy alone were compared with those treated with chemotherapy plus metronomic chemotherapy. DFI was 202 days for the Chemotherapy group and 154 days for the Chemotherapy and Metronomic group (p = 0.54). Survival time was 279 days fro the Chemotherapy and 237 days for the Chemotherapy and Metronomic group (p = 0.88). O05 INVESTIGATING NQO1 AS A RADIATION INDUCIBLE ENZYME IN OSTEOTROPIC NEOPLASMS AND TARGETED THERAPY WITH THE NOVEL DRUG DEOXYNYBOQUINONE Alycen Lundberg 1, Holly Pondenis1, Mark Byrum1, Katie Wycislo1, Elizabeth Parkinson1, Jayme Looper2, Richard L. Fredrickson1, Paul J. Hergenrother1, Timothy M. Fan1 1University of Illinois at Urbana‐Champaign, Urbana, IL, USA, 2Chicago Veterinary Cancer Center, Chicago, IL, USA Osteotropic neoplasms including feline oral squamous cell carcinoma (OSCC), and canine prostatic carcinoma (PC), oral malignant melanoma (OMM) and osteosarcoma (OS) are managed with ionizing radiation therapy, yet treatment responses are not durable. NQO1 is a detoxifying enzyme overexpressed in human tumors, but remains uncharacterized in veterinary oncology. Deoxynyboquinone (DNQ) is a potent substrate for NQO1, and results in reactive oxygen species generation with consequent cell death. We hypothesize that NQO1 will be expressed by osteotropic solid tumors and serve as druggable target for improving the cytotoxic activities of radiation therapy. Western blot and immunohistochemistry identified NQO1 in cell lines and an enzymatic activity assay determined the presence of a functional protein. One cell line per tumor histology was selected to demonstrate a temporal and dose‐dependent regulation of NQO1 by radiation. The cytotoxic effects of irradiation and DNQ were investigated by treating selected cell lines and measuring surviving fraction of cells in conjunction with fluorescent microscopy to identify double‐stranded DNA breaks via the biomarker gammaH2AX. NQO1 was expressed by all malignant cell lines and enzymatic activity confirmed. In selected cell lines cells it is shown that NQO1 is upregulated with radiation therapy. Irradiation and DNQ therapies alone, and in combination, are cytotoxic. These results indicate that NQO1 is expressed by malignant solid tumors and expression is upregulated following irradiation. DNQ is a potent substrate for NQO1 and induces cytotoxicity alone and in combination with radiation, supporting investigation of DNQ for targeted treatment of osteotropic solid tumors in a clinical setting. O06 A PROSPECTIVE, OPEN‐LABEL STUDY EVALUATING TREATMENT OF CANINE B‐CELL LYMPHOMA WITH L‐ASPARAGINASE, DOXORUBICIN AND A CANINE ANTI‐CD20 MONOCLONAL ANTIBODY Carlos Rodriguez Jr 1, Teri Guerrero1, Lydia Cook2, Genevieve Hansen2 1University of California, Davis, Davis, CA, USA, 2Aratana Therapeutics, Inc., Kansas City, KS, USA The use of doxorubicin chemotherapy protocols (with or without L‐asparaginase) in canine patients has been shown to be efficacious in the treatment of B‐cell lymphoma, particularly when multiple doses of doxorubicin are administered. A caninized monoclonal antibody (MAb) targeting CD20 has the potential to improve the treatment outcome of B‐cell lymphoma patients treated with a minimal chemotherapy protocol. This study evaluated the safety, progression‐free survivals (PFS), and overall survival (OS) of combining an abbreviated protocol of L‐asparaginase and doxorubicin with MAb to treat B‐cell lymphoma. Dogs (n = 20) with newly diagnosed B‐cell lymphoma and measurable peripheral lymph node enlargement were enrolled in this prospective, open‐label study. In Phase 1 subjects were treated with three doses of doxorubicin (30 mg/m2) once every three weeks and seven doses of MAb (5 mg/kg) over a period of fourteen weeks. At relapse, nine of the eleven dogs who were eligible to participate in Phase 2 were treated with an additional dose of doxorubicin (30 mg/m2) and four doses of MAb (5 mg/kg). Two of the dogs treated in Phase 1 never relapsed, one of which is still alive (OS: 790 days). At the start of both Phase 1 and 2, dogs were able to receive 1 dose of L‐asparaginase. Nineteen of the twenty dogs enrolled received treatment with MAb (125 total treatments; 1 dog withdrew prior to MAb administration). Three of the dogs administered MAb experienced MAb infusion‐related reactions (1 reaction/dog). One of the twenty dogs administered doxorubicin (60 total treatments) experienced doxorubicin toxicity after the third dose of doxorubicin. The intent‐to‐treat population data demonstrated an initial median PFS of 103 days, a median PFS of 49 days following treatment in Phase 2 and a median OS of 242 days. Dogs in the per protocol population (n = 14) who received three or more doxorubicin treatments had an initial median PFS of 148 days, a median PFS of 41 days in Phase 2 and a median OS of 344.5 days. These data suggest the addition of CD20 MAb to a protocol including L‐asparaginase and doxorubicin was safe under these treatment conditions and may provide benefit to dogs with B cell lymphoma. O07 CHARACTERIZATION OF IMMUNOGLOBULIN GENE USE AND MUTATION STATUS IN CANINE B CELL CHRONIC LYMPHOCYTIC LEUKEMIA Emily Rout 1, Robert Burnett1, Courtney Abbott1, Stacey George1, Anne Avery1 1Colorado State University, Fort Collins, CO, USA Canine B cell chronic lymphocytic leukemia (CLL) is common in dogs and shares many features with human CLL. In human CLL, immunoglobulin (Ig) gene usage and mutation status are important markers for disease behavior, but these factors have not been assessed in canine B‐CLL. We sequenced the immunoglobulin heavy chain variable region (VH) genes from neoplastic peripheral blood B cells in 45 dogs with B‐CLL. Thirteen VH genes were used. Excluding the Boxer breed, the most commonly used genes were VH1‐44, VH1‐62, and VH1‐35. In non‐neoplastic B cells, VH1‐44 and VH1‐62 are the most commonly used genes. 59% of all genes used in the B‐CLL patients had greater than 2% mutations compared to germline sequence and were classified as mutated, while 41% were unmutated. Of the cases using VH1‐44, one third used an unmutated gene. All the cases using VH1‐62 and VH1‐35 used mutated genes. VH1‐41 is overrepresented in Boxers with B‐CLL, but is not preferentially used in Boxers with other B‐cell neoplasms. Of 8 Boxer dogs with B‐CLL that were sequenced, 6 used an unmutated VH1‐41 gene. These findings suggest that antigen selection may play a role in the pathogenesis of B‐CLL in Boxers. These data lay the foundation to correlate VH gene usage and mutation status with outcome in dogs, further establish canine patients as good models of human CLL, and identify groups, like the Boxers, that may be useful in studying the pathogenesis of CLL. O08 MOLECULAR CHANGES ASSOCIATED WITH THE DEVELOPMENT OF RESISTANCE TO IMATINIB IN AN IMATINIB‐SENSITIVE CANINE NEOPLASTIC MAST CELL LINE CARRYING A KIT C.1523A>T MUTATION Masato Kobayashi 1, Shiori Kuroki1, Yuji Uehara1, Yukari Kozutsumi1, Yu Tanaka1, Yukari Moriya1, Kyoichi Tamura1, Tsukimi Washizu1, Makoto Bonkobara1 1Nippon Veterinary and Life Science University, Tokyo, Japan Imatinib has potent anti‐tumor activity against canine mast cell tumor (MCT) carrying c‐kit mutation. Although the tumors initially respond well to imatinib, they eventually develop resistance. Here, molecular mechanisms that confer imatinib resistance to neoplastic mast cells were investigated using an imatinib‐sensitive canine neoplastic mast cell line VI‐MC carrying a KIT c.1523A>T activating mutation. Two imatinib‐resistant sublines were established by culturing VI‐MC cells in increasing concentrations of imatinib (1 □M‐resistant, rVI‐MC1; 10 □M‐resistant, rVI‐MC10). Both sublines had a second KIT mutation c.2443G>C. Recombinant KIT with the second mutation was insensitive to 1 □M but sensitive to 10 □M imatinib. The effect of imatinib on the phosphorylation of KIT and its downstream signalling proteins was then examined using these sublines. KIT and ERK were constitutively phosphorylated in both sublines and their phosphorylation was suppressed by 10 □M imatinib in rVI‐MC1 cells. However, KIT but not ERK phosphorylation was supressed in rVI‐MC10 cells. The phosphorylation of ERK in rVI‐MC10 cells was also not diminished by the Src family kinase (SFK) inhibitor dasatinib. This second mutation in KIT may play an important role for imatinib resistance in neoplastic mast cells. Furthermore, KIT/SFK‐independent activation of ERK would be involved in imatinib resistance when the neoplastic cells are exposed to higher concentrations of imatinib. O09 FIXED DOSE RATE GEMCITABINE IN THE CAT Crystal Garnett 1, Carlos O. Rodriguez Jr1 1University of California, Davis, CA, USA Gemcitabine is an antimetabolite chemotherapy agent with clinical activity against a number of human malignancies. Gemcitabine is activated to its cytotoxic triphosphate form (dFdCTP) and in a concentration‐dependent manner, competes for incorporation into the DNA and induces apoptosis. Whereas intracellular accumulation of dFdCTP is schedule dependent and relies on achieving and maintaining optimal extracellular plasma concentrations (10‐20 □m), excess gemcitabine plasma concentrations inhibit the critical activation process. There is no current standard dosing scheme for the use of gemcitabine in feline patients with cancer. The purpose of this study is to identify the fixed‐dose rate (FDR) of gemcitabine that results in the optimal plasma concentration (OPC) in the cat. Cancer‐bearing, client owned cats with various malignancies were enrolled in this study. Gemcitabine was administered as FDR intravenously (IV) for 1‐6 hours and plasma collected before administration, during, and at the end of infusion. Plasma proteins were precipitated, pelleted, and plasma gemcitabine was quantitated by HPLC. FDR of 1 mg/m2/min for up to 6 hours failed to achieve OPC. FDR of 2.5 mg/m2/min achieved OPC between 4 and 6 hours. FDR of 5 mg/m2/min achieved OPC between 2 and 5 hours of infusion. FDR of 10 mg/m2/min did not achieve OPC at 1 hour, however exceeded OPC within 2‐3 hours. These data suggest that the ideal FDR in the cat lies between 2.5‐5 mg/m2/min due to the maintenance of OPC for 2‐4 hours. A follow up study evaluating gemcitabine in the cat using FDR of 2.5‐5 mg/m2/min is recommended and currently underway. O10 PHARMACOKINETICS AND BIOAVAILABILITY OF METFORMIN IN DOGS Charlotte Johnston 1, Valerie MacDonald1, Jane Alcorn2, Casey Gaunt1 1Western College of Veterinary Medicine, University of Saskatchewan, Saskatoon, SK, Canada, 2College of Pharmacy and Nutrition, University of Saskatchewan, Saskatoon, SK, Canada Metformin is used for management of non‐insulin dependent diabetes mellitus in humans. Clinical evidence suggests patients taking metformin have a reduced lifetime incidence of cancer. Metformin's mechanism of action is not understood, but may target AMP activated protein kinase (AMPK), which induces glucose uptake from blood into muscles. The upstream regulator of AMPK is LKB1, a well‐recognized tumor suppressor. The study's purpose was to determine the pharmacokinetics and oral bioavailability of metformin to facilitate clinical trials evaluating metformin's effect in dogs with lymphoma. Seven healthy adult mixed breed dogs were enrolled in a randomized crossover design study consisting of a paired IV/oral crossover, with a 1‐week washout period between treatments. Dogs were given 25 mg/kg metformin IV or 20 mg/kg metformin orally. Before administration and at time 0.33, 0.57, 1, 1.33, 1.67, 2, 2.5, 3, 4, 5, 6, 7, 8, 10, 12, 15, 18, 24, 30, 36, 48 and 72 hours after administration blood samples were collected via indwelling catheter for drug analysis. Metformin serum concentrations were determined using a validated LC‐MS/MS method. Pharmacokinetic parameters were estimated from each individual dog serum concentration‐time curves by non‐compartmental methods using GraphPad Prism software. Concentration‐time profiles demonstrated three‐compartment kinetics. Mean±SD pharmacokinetic parameter estimates were: ClS (mL/min/kg) = 24.1 ± 7.8; Vd,area (L/kg) = 44.8 ± 23.5; k (h−1) = 0.035 ± 0.007; t1/2 (h) = 20.4 ± 4.1; oral F = 0.32 ± 0.06; Cmax (ng/mL) = 835.8 ± 308.3; and tmax (h) = 2.5 ± 0.4. The general disposition pattern of metformin in dogs is similar to that reported in cats and humans. O11 GENESCAN ANALYSIS FOR THE DIFFERENTIATION OF CLONAL ORIGIN OF CELLS IN DOGS THAT DEVELOPED TWO DIFFERENT TYPES OF LYMPHOID MALIGNANCIES Saaya Hiyoshi 1, Yuko Goto‐Koshino1, Masashi Takahashi1, Kazuyuki Uchida2, Masaya Tsuboi2, Yasuhito Fujino1, Koichi Ohno1, Hajime Tsujimoto1 1Department of Veterinary Internal Medicine, Graduate School of Agricultural and Life Sciences, The University of Tokyo, Bunkyo‐Ku, Tokyo, Japan, 2Department of Veterinary Pathology, Graduate School of Agricultural and Life Sciences, The University of Tokyo, Bunkyo‐ku, Tokyo, Japan Development of secondary high‐grade lymphoma from indolent lymphoid malignancies has been reported in humans, commonly referred to Richter's syndrome. Similar conditions have been also reported in dogs. Recently, we have constructed a new GeneScan analytical system using multicolor‐labeled primers and capillary electrophoresis to detect the rearrangement of immunoglobulin heavy chain (IgH) and T‐cell receptor gamma chain (TCRγ) genes in dogs. Herein, we used this GeneScan analysis to differentiate the clonal origin of cells in dogs that developed two different types of lymphoid malignancies. Five dogs with an initial diagnosis of marginal zone lymphoma (MZL; n = 2), B‐cell chronic lymphocytic leukemia (B‐CLL; n = 1), or T‐zone lymphoma (TZL; n = 2) developed secondary high‐grade B‐cell lymphoma (diffuse large B‐cell lymphoma, DLBCL) more than 1 year after the initial diagnosis. GeneScan analysis revealed a single peak of the IgH gene with the same length in both the primary (MZL and B‐CLL) and secondary tumors (DLBCL) in 3 dogs, indicating that the secondary DLBCL evolved from the pre‐existing primary indolent neoplasm clone. A single TCRγ peak was identified in the 2 TZL samples. However, subsequent DLBCL samples from the same 2 dogs revealed a single peak of the IgH gene instead of the TCRγ□ peak, thereby suggesting a different origin between the 2 lymphoid neoplasms developed in the same subjects. This study indicated that GeneScan analysis would be a convenient and accurate method to determine the clonal origin of lymphoid neoplasms in dogs. O12 ANALYSIS OF LOMUSTINE CONTENT IN FDA APPROVED AND COMPOUNDED LOMUSTINE CAPSULES Butch KuKanich 1, Matt Warner1, Kevin Hahn2 1Kansas State University, College of Veterinary Medicine, Dept. Anatomy & Physiology, Manhattan, KS, USA, 2NextSource Biotechnology, LLC, Miami, FL, USA The purpose of this study was to compare lomustine content in FDA approved and compounded capsules. The a priori acceptance range was 90‐110% of the stated capsule content. Capsules were obtained from an FDA approved manufacturer (NextSource Biotechnology) and 3 national (USA) compounding pharmacies at 10 mg and 40 mg capsules or the nearest available content. Replicates of 5 capsules per source and strength were assessed by validated high pressure liquid chromatography with ultraviolet detection method. Contents of capsules were emptied individually into methanol to yield a stated concentration of 1 mg/mL, vortexed for 30 seconds and sonicated for 10 minutes. The samples were then vortexed, centrifuged and further diluted in 50% methanol to a final stated concentration of 50 μg/mL. The lomustine concentration was determined by comparing to a standard curve in 50% methanol. The FDA approved capsules had 100% of the capsules for both strengths within 90‐110% of stated content. Compounder A had 60% and 0% of 11 mg and 42 mg capsules, respectively within 90‐110%. Compounder B had 20% and 0% of 10 mg and 40 mg capsules, respectively within 90‐110%. Compounder C had 60% and 0% of 7 mg and 48 mg capsules, respectively within 90‐110%. All product failures were <90% of stated content. In conclusion, compounded lomustine capsules failed to have the stated amount of drug, but the FDA approved formulation contained the stated amount of drug. The clinical implications of using compounded lomustine may include treatment failures and increased risk of tumor resistance. O13 EXPRESSION OF MULTIPLE DRUG RESISTANCE MARKERS ABCB1 AND LRP IN CATS WITH ALIMENTARY LYMPHOMA Valter de Medeiros Winkel, Archivaldo Reche Jr., Bruno Cogliati, Silvia Regina Ricci Lucas School Of Veterinary Medicine and Animal Science of the University of São Paulo, São Paulo‐ SP, Brazil Alimentary lymphoma is a common neoplasia in cats and systemic chemotherapy is indicated for the treatment. Although most of cats can achieve a complete remission with chemotherapy, relapse can be associated to the multidrug resistance, that occurs with the use of cytotoxic agents or a previous intrinsic cellular resistance, provided by transmembrane proteins that decreases intracellular drug bioavailability. Furthermore, the lung resistance‐related protein (LRP) known as the human major vault protein was found to be overexpressed in many chemoresistant primary tumors of different origins, including lymphomas, that did not express ABCB1 (P‐gp) The aim of this study was to investigate ABCB1 and LRP expression in cats with alimentary lymphoma. Immunohistochemistry was performed in 16 samples of cats with alimentary lymphoma without previous treatment, using the monoclonal antibodies: mouse anti‐P‐glycoprotein (C494; Enzo) and mouse anti‐LRP (1014; Santa Cruz Biotech). ABCB1 and LRP immunoreactivity were analysed by degree of labeling intensity, between 0 (no expression) to 3 (strong expression). Lymphoma was classified as B cell in 6 cats (one small and 5 large cells) and as T cells in 10 cats (small cells). ABCB1 expression as well as LRP was higher in cats with B cell lymphoma than in T cell lymphoma. The survival time was longer in cats that did not express or had slightly expression of ABCB1, even in B cell lymphoma. The same was not observed for LRP. More studies are necessary to verifiy if ABCB1 expression could be associated with complete or partial response and survival time. O14 METALLOTHIONEIN AND KI‐67 IN CATS WITH ALIMENTARY LYMPHOMA Valter de Medeiros Winkel, Bruno Cogliati, Archivaldo Reche Jr., Ana Luisa Nairismagi Alves, Silvia Regina Ricci Lucas Scholl of Veterinary Medicine and Animal Science of the University of São Paulo, São Paulo‐ SP, Brazil Alimentary lymphoma comprises one of the most common neoplasia in cats. Ki‐67 antigen is expressed in proliferating cells during all phases of the cell cycle, but is not found in G0. Metallothionein (MT) is a low molecular weight protein that increases significantly during oxidative stress to protect the cells against citotoxicity and DNA damage and these properties contribute to the anti‐apoptotic function. MT as well as Ki‐67 represents a poor prognostic factor in DLBCL. The goal of this study was investigate Ki‐67 and MT expression in cats with alimentary lymphoma. Immunohistochemistry in paraffin samples was performed in 16 cats with alimentary lymphoma without previous treatment, using monoclonal antibodies: mouse anti‐PAX‐5 (PAX5 Clone 24; Invitrogen); polyclonal rabbit anti‐CD3 (DAKO); mouse anti‐Ki67 (clone MIB‐1; DAKO); and mouse anti‐metallothionein (clone E9; DAKO). The Ki‐67 immunoreactivity was determined by count 1,000 cells in 40X fields, while metallothionein was analyzed by degree of labeling intensity between 0 (no expression) to 3 (strong expression). Six cats had B‐cell (one small cell and 5 large cells) and 10 had T‐cell (small cell) lymphoma. The mean of Ki‐67 expression was 60.5% in cats with B‐cell and 10.8% in T‐cell lymphoma (p < 0.0001), and the median of survival time was 3 and 27 months, respectively. Only one cat expressed MT (grade 2) associated with low levels of Ki‐67 (5%) and had a long survival time. In conclusion Ki‐67 had relation with immunophenotype and survival time and MT was rarely express in alimentary lymphoma in cats. O15 RESPONSE OF IL‐2 AND IL‐6 AFTER UNILATERAL TOTAL MASTECTOMY ALONE OR IN ASSOCIATION WITH OVARIOSALPINGOHISTERECTOMY IN BITCHES WITH MAMMARY TUMORS Sabrina Rodigheri 1, Taise Fuchs2, Evandro Zacché1, Andrigo De Nardi1 1Universidade Estadual Paulista UNESP, Jaboticabal, Brazil, 2Universidade Positivo Unicenp, Curitiba, Brazil Surgical resection of the primary tumor is an effective treatment for mammary neoplasms in dogs. Nevertheless, the surgical stress response may provide conditions for persistence of residual minimal disease and the development of pre‐existing micrometastases. Several mechanisms underlying these processes have been proposed, including postoperative suppression of cell‐mediated immunity, the first‐line defense mechanism against cancer. Existing data in human patients suggest that an increase in the level of interleukin 6 (IL‐6) and interleukin 8 (IL‐8), in combination with a decrease of T helper cell‐induced cytokine production, interleukin 2 (IL‐1) and interferon gama (IFN‐ γ), could account for the profound suppression of natural killer cytotoxic activity in the perioperative period. Given the lack of related studies evaluating the immune response associated with mastectomy in dogs, the aim of this study was to evaluate the serum concentration of IL‐2 and IL‐6 in the perioperative period of unilateral total mastectomy (UTM) alone or in association with ovariosalpingohisterectomy (OSH) in bitches with mammary tumors. For this purpose, were included in the study 8 bitches with mammary tumors submitted to UTM and 8 to UTM associated to OSH. Venous blood samples were taken preoperatively (M0), at the end of surgery (M1), 3 hours (M2), 6 hours (M3), 24 hours (M4), 72 hours (M5) and 10 days postoperatively, for serum IL‐2 and IL‐6 levels. Data were logarithmically transformed and then submitted to Friedman test for differences between moments within the group and Mann Whitney test for moments between groups. There was no significant difference between moments or groups in any of the studied parameters. Results are summarized in Table 1. Table 1. Mean ± SD of serum IL‐2 and IL‐6 of dogs submitted to UTM (n = 8) and UTM + OSH (n = 8). Parameter Group M0 M1 M2 M3 M4 M5 M6 IL‐2log UTM 2.00 ± 1.39 1.91 ± 1.38 1.88 ± 1.37 1.79 ± 1.48 1.77 ± 1.46 1.74 ± 1.46 1.82 ± 1.48 UTM+OSH 1.64 ± 1.45 1.59 ± 1.41 1.56 ± 1.45 1.53 ± 1.44 1.44 ± 1.46 1.40 ± 1.50 1.48 ± 1.42 IL‐6log UTM 2.01 ± 1.18 2.03 ± 1.14 2.28 ± 0.91 2.32 ± 0.88 2.19 ± 1.04 1.98 ± 1.20 1.95 ± 1.32 UTM+OSH 2.03 ± 1.26 1.98 ± 1.26 2.17 ± 1.06 2.13 ± 1.05 2.12 ± 1.10 1.96 ± 1.24 1.91 ± 1.27 Under the conditions of this study and based on the observed results it can be concluded that UTM alone or in association with OSH does not induce significant changes in serum levels of IL‐2 and IL‐6 in the postoperative period. O16 STRESS RESPONSE TO SURGICAL TRAUMA OF UNILATERAL TOTAL MASTECTOMY ALONE OR IN ASSOCIATION WITH OVARIOSALPINGOHISTERECTOMY IN BITCHES WITH MAMMARY TUMORS Sabrina Rodigheri 1, Taise Fuchs2, Evandro Zacché1, Andrigo De Nardi1 1Universidade Estadual Paulista UNESP, Jaboticabal, Brazil, 2Universidade Positivo Unicenp, Curitiba, Brazil Increased secretion of pituitary hormones, activation of the sympathetic nervous system and the release of cytokines and acute phase proteins are the main metabolic changes associated with surgical trauma. The intensity of the organic response depends on the magnitude, duration and nature of the stimuli. Minor surgery commonly induces smaller responses compared to medium and large procedures. Excessive release of hormones and humoral mediators after surgical trauma may be responsible for multiple adverse events in the postoperative period, included high protein catabolism, metabolic acidosis, blood hypercoagulability, interstitial edema, immune depression and delayed healing. Given the lack of related studies evaluating the metabolic response associated with mastectomy in dogs, the aim of this study was to evaluate the serum concentration of cortisol, glucose and albumin in the perioperative period of unilateral total mastectomy (UTM) alone or in association with ovariosalpingohisterectomy (OSH) in bitches with mammary tumors. For this purpose, were included in the study 8 bitches with mammary tumors submitted to UTM and 8 to UTM associated to OSH. Venous blood samples were taken preoperatively (M0), at the end of surgery (M1), 3 hours (M2), 6 hours (M3), 24 hours (M4), 72 hours (M5) and 10 days postoperatively (M6), for serum cortisol, glucose and albumin levels. Data were logarithmically transformed and then submitted to two‐way ANOVA and Tukey's post hoc test for differences between treatment and across time. There was no difference between groups across time in any of the studied parameters. Results are summarized in Table 1. Considering the anesthetic and analgesic protocol used in this study and based on the observed results it can be concluded that UTM alone or in association with OSH does not induce significant changes in serum levels of cortisol, glucose and albumin in the postoperative period. In animals submitted to UTM was found only a slight increase in serum glucose levels at the end of surgery. Table 1. Mean ± sd of serum cortisol, glucose and albumin of dogs submitted to UTM (n = 8) and UTM + OSH (n = 8). Parameter Group M0 M1 M2 M3 M4 M5 M6 Cortisollog UTM 3.95 ± 0.31 4.11 ± 0.36 4.10 ± 0.29 4.02 ± 0.23 3.90 ± 0.23 3.78 ± 0.23 4.01 ± 0.57 UTM+OSH 3.92 ± 0.25 3.94 ± 0.24 3.99 ± 0.16 3.92 ± 0.24 4.02 ± 0.17 3.99 ± 0.24 3.99 ± 0.25 Glucoselog UTM 1.95 ± 0.05a 2.06 ± 0.12b 2.05 ± 0.08ab 2.04 ± 0.04ab 2.01 ± 0.09ab 2.00 ± 0.05ab 2.00 ± 0.07ab UTM+OSH 2.00 ± 0.06 2.10 ± 0.09 2.11 ± 0.14 2.06 ± 0.03 2.05 ± 0.05 2.03 ± 0.05 2.02 ± 0.05 Albuminlog UTM 0.42 ± 0.11 0.38 ± 0.07 0.37 ± 0.14 0.41 ± 0.08 0.40 ± 0.07 0.40 ± 0.08 0.44 ± 0.08 UTM+OSH 0.47 ± 0.12 0.38 ± 0.16 0.43 ± 0.08 0.45 ± 0.14 0.42 ± 0.15 0.44 ± 0.09 0.46 ± 0.08 Means followed by the same letter in the line are not significantly different O17 PHARMACOKINETICS OF ORALLY ADMINISTERED LOW‐DOSE RAPAMYCIN IN HEALTHY DOGS Jeanne Larson, Sara Allstadt Frazier, Tim Fan, Chand Khanna, Paul Lunghofer, Ryan Hansen, Dan Gustafson, Alfred Legendre, Gina Galyon, Amy LeBlanc, Tomas Martin‐Jimenez University of Tennessee, Knoxville, TN, USA Introduction: Rapamycin is an antifungal antibiotic agent that has immunosuppressant and anti‐cancer properties. By inhibiting mTOR, rapamycin acts as an anti‐proliferative and anti‐angiogenic agent. The purpose of this pilot study was to determine the blood concentrations of rapamycin achieved following a single dose and multiple oral doses in the dog. Methods: Five healthy purpose‐bred hound dogs were enrolled. In Experiment 1, dogs received a single 0.1 mg/kg dose orally. Blood was collected at 0, 0.5, 1, 2, 4, 6, 12, 24, 48, and 72 hours. In Experiment 2, dogs received 0.1 mg/kg daily for 5 days. Blood was collected at 0, 3, 6, 24, 27, 30, 48, 51, 54, 72, 75, 78, 96, 96.5, 97, 98, 100, 102, 108, 120, 144, and 168 hours. Samples were analyzed using a validated LC/MS/MS assay and evaluated via non‐compartmental analysis. Results: Variable blood concentrations between single and consecutive dosing were noted. The mean pharmacokinetic values for Experiment 1 were: half life (t1/2) = 38.7 ± 12.7 hr, area under the curve (AUC) = 140 ± 23.9 ng*hr/ml, and maximum concentration (Cmax) = 8.39 ± 1.73 ng/ml. The mean pharmacokinetic values on day 5 for Experiment 2 were: t1/2 = 99.3 + /‐ 89.5 hr, AUC = 126 ± 27.1 ng*hr/ml, and Cmax = 5.49 ± 1.99 ng/ml. Conclusions: Rapamycin administered orally at 0.1 mg/kg achieves measurable blood concentrations. Goals of ongoing studies in tumor‐bearing dogs are to determine the optimal dose and frequency to achieve clinically relevant concentrations and to determine toxicity. (VCS Award Winner) O18 MINOXIDIL INHIBITS THE METASTATIC PHENOTYPE OF OSTEOSARCOMA Christine Mullin 1, Allyson Koyen1, Arnulfo Mendoza1, Michael Lizardo1, Ling Ren1, Charles Halsey2, Peter Ho3, Chand Khanna1 1Pediatric Oncology Branch, Tumor and Metastasis Biology Section, Center for Cancer Research at the National Cancer Institute, NIH, Washington, DC, USA, 2Molecular Pathology Unit, Laboratory of Cancer Biology and Genetics, NIH, Washington, DC, USA, 3Metastagen, Inc., Wilimington, DE, USA Introduction: A recent study documented the activity of the antihypertensive medication minoxidil in a murine model of sarcoma. Specifically, minoxidil inhibited in vitro cell migration at 0.5 mM and reduced pulmonary metastasis in mice at 1‐3 mg/kg. A proposed MOA involves the down‐regulation of the collagen‐modifying enzyme PLOD2, which occurred at 0.5 mM minoxidil. The activity of minoxidil and role of PLOD2 as a druggable target in osteosarcoma have yet to be investigated. PK/PD data are needed to clarify the anti‐metastatic mechanism of minoxidil. Methods: The growth inhibitory effect of minoxidil was assessed in osteosarcoma by SRB assay. An ex vivo pulmonary metastasis assay (PuMA) was used to examine minoxidil's activity against metastatic progression. Plasma minoxidil levels were measured in mice receiving the previously reported active doses. PLOD2 expression and modulation by minoxidil were assessed via IHC and WB. Results: The IC50 of minoxidil was ~1 mM across all cell lines. PuMA demonstrated metastatic inhibition at minoxidil concentrations ≥1 mM. Plasma minoxidil concentrations achieved after 1‐3 mg/kg dosing were in the low micromolar range. PLOD2 was expressed in osteosarcoma in all three species, but only inconsistent PLOD2 knockdown at minoxidil exposures ≥0.5 mM was noted. Conclusions: Minoxidil has activity against pulmonary metastasis ex vivo, but at significantly higher doses than are biologically achievable. Investigation of the in vivo effects of minoxidil on osteosarcoma metastasis in mice is planned; intratumoral drug levels will be measured to determine whether minoxidil concentration in tissues can explain the discrepancy between low plasma levels achieved and the high exposures required for anti‐metastatic activity and PLOD2 modulation. (VCS Award Winner) EN01 THE FELINE AIP GENE: THE KEY TO HYPERSOMATOTROPISM TUMORIGENESIS? C. Scudder, S.J.M. Niessen, B. Catchpole, R.C.F. Fowkes, D.B.C. Church, Y. Forcada Royal Veterinary College, North Mymms, Hertfordshire, UK Feline acromegaly is an increasingly recognised endocrinopathy, caused by chronic excessive serum growth hormone concentration produced by a functional somatotrophinoma in the pars distalis of the anterior pituitary gland. The underlying pathogenesis for the development of most human somatotrophinomas is poorly understood. The majority of human somatotrophinomas are sporadic, however up to 20% of familial isolated pituitary adenomas are caused by germline mutations of the aryl‐hydrocarbon‐receptor interacting protein (AIP). Feline acromegaly has phenotypic and biochemical similarities to human families with AIP mutations, such as male predominance, somatotroph macroadenoma and resistance to octreotide therapy. The objective of this study was to identify the feline AIP gene, identify SNPs within the feline AIP gene and compare any SNPs with reported human AIP SNPs. Stored pituitary tissue from an acromegalic cat was used to create and amplify feline AIP cDNA using feline specific AIP primers. Stored EDTA blood from 10 acromegalic cats (diagnosis of insulin resistant diabetes mellitus, serum IGF‐1 > 1000 ng/ml and pituitary mass >4 mm identified using contrast‐enhanced pituitary computed tomography or at necropsy) and 10 control cats (no history of diabetes mellitus and greater than 15 years of age) were selected, DNA extracted and genotyped using PCR, agarose gel electrophoresis and Sanger sequencing. The feline AIP gene was identified, which encoded for a 330 amino acid protein. This protein had 98% homology to the human AIP protein. A BLAST search revealed this gene contained 6 exons, and exon specific primers were created. A single non‐conservative SNP was identified in exon 1 (AIP:c.9G>T), encoding for an amino acid change from aspartic acid to glutamic acid in 2/10 acromegalic patients and 0/10 control samples. Two additional conservative SNPs were also identified (AIP:c.826T>C and AIP:c.481T>C). Exon 1 encodes for a region of the AIP protein considered essential for AIP‐AIP receptor interaction. Although 70 different human mutations have been identified to date, a human AIP:c9G>T mutation has not yet been identified. The AIP N‐terminal is required for the stability of the AIP protein ‐ AhR‐receptor complex, and essential for the regulation of translocation into the nucleus, where it binds to aryl hydrocarbon receptor nuclear translocator leading to activation of genes thought to act as tumour suppressors. Loss of normal AIP activity is thought to promote somatotrophinoma development. It is possible that the AIP:c.9G>T mutation predisposed to somatrotrophinoma tumourigensis in the two affected patients, and a study containing larger numbers of acromegalic and control cases would improve our understanding of the AIP:c.9G>T SNP. (ESVE Award Winner) EN02 EVALUATION OF SERUM THYROID STIMULATING HORMONE CONCENTRATION AS A DIAGNOSTIC TEST FOR HYPERTHYROIDISM IN CATS Mark Peterson 1, Jade Guterl1, Rhett Nichols2, Mark Rishniw3 1Animal Endocrine Clinic, New York, New York, USA, 2Antech Diagnostics, Lake Success, New York, USA, 3College of Veterinary Medicine, Cornell University, Ithaca, New York, USA In human patients, measurement of serum TSH is commonly used as a first‐line discriminatory test of thyroid function, even before evaluation of serum T4, T3, or free T4 (FT4). Recent reports indicate that the canine TSH assay (Immulite Canine TSH, Siemens) can be used to measure feline TSH, and that this may be used to help diagnose or exclude hyperthyroidism in cats suspected of having the disease. Because high circulating T4 suppresses TSH secretion, serum TSH levels should be low in hyperthyroid cats, possibly even before serum T4 or free T4 are clearly elevated. Some have suggested that finding a measurable cTSH in a hyperthyroid cat is so unlikely that this alone can be used to rule out this disease. The purpose of this study was to investigate the usefulness of cTSH measurements in the diagnosis of mild to severe hyperthyroidism and in differentiating hyperthyroid cats (n = 786) from healthy cats (n = 120) and euthyroid cats suspected of hyperthyroidism but found not to have the disease (n = 28). For inclusion, all cats had routine laboratory testing, a complete serum thyroid panel (T4, T3, FT4 by dialysis, TSH), and thyroid scintigraphy. Subsequent calculation of thyroid‐to‐salivary [T/S] and thyroid‐to‐background [T/B] ratios were used to confirm or exclude hyperthyroidism, as previously described (Thyroid scintigraphy findings in 2,096 cats with hyperthyroidism. Vet Radiol Ultrasound, in press). Median serum concentrations of T4, T3, and FT4 in the 786 hyperthyroid cats (7.9 μg/dl, 151 ng/dl, 100 pmol/L, respectively) were significantly higher (P < 0.001, Kruskal–Wallis test) than median concentrations in the 28 euthyroid cats with suspected thyroid disease (2.2 μg/dl, 54 ng/dl, 38 pmol/L) or the 120 clinically normal cats (2.2 μg/dl, 50 ng/dl, 33 pmol/L). Of the hyperthyroid cats, 742 (94%), 513 (65%), and 757 (96%) had high serum concentrations of of T4, T3, and FT4, respectively; of the 28 euthyroid cats, 2 (7%,), 1 (3.5%), and 5 (18%) had falsely high values of T4, T3, or FT4, respectively. The calculated sensitivity of T4, T3, and FT4 as a diagnostic test for hyperthyroidism was 94%, 65%, and 96%, respectively, whereas the specificity of each test was 93%, 96%, and 82%, respectively. Median serum TSH concentration in the 786 hyperthyroid cats (0.03 ng/ml) was significantly (P < 0.001) lower than median values in the normal cats (0.05 ng/ml) and the euthyroid cats with suspected thyroid disease (0.06 ng/ml). No significant difference existed between the median concentrations of T4, T3, free T4, or TSH in clinically normal cats and the values in the suspect cats. Only 14 (1.8%) hyperthyroid cats had measurable TSH values, whereas 98.2% had TSH concentrations at or below the level of quantification (<0.03). In contrast, TSH concentrations were detectable in 71 (59%) of the 120 normal cats and in 24 (86%) of the 28 euthyroid cats. Overall, the sensitivity of serum TSH as a diagnostic test for hyperthyroidism was 98.2%, but the test specificity was only 49.3%. Of the 14 hyperthyroid cats with detectable serum TSH values, 10 had mild disease (T4 < 6 μg/dl) and 4 had moderate hyperthyroidism (T4 ≥ 6‐12 μg/dl); none of the cats with severe thyroid disease had measurable TSH values. Prevalence of detectable serum TSH values in cats with mild hyperthyroidism (4.4%) was significantly (P = 0.016) higher than the prevalence in cats with moderate disease (1.2%). In conclusion, serum TSH concentrations are suppressed in over 98% of hyperthyroid cats, but values are indeed measurable in a few cats with mild to moderate hyperthyroidism. Although most clinically normal cats and euthyroid cats suspected of having thyroid disease will have detectable TSH concentrations, about 40% of the euthyroid cats in this study also had serum concentrations that were below the level of quantification. Overall, serum TSH determination represents a highly sensitive but poorly specific test for diagnosis of feline hyperthyroidism when used in combination with T4 and FT4 determinations. EN03 EFFICACY OF LOW‐DOSE (2 MILLICURIE) VERSUS STANDARD‐DOSE (4 MILLICURIE) RADIOIODINE (131I) TREATMENT FOR CATS WITH MILD‐TO‐MODERATE HYPERTHYROIDISM John Lucy 1, Mark Peterson2, John Randolph1, Peter Scrivani1, Meg Thompson1, Danielle Davignon1, Janet Scarlett1 1Cornell University College of Veterinary Medicine, Ithaca, NY, USA, 2Animal Endocrine Clinic, New York, NY, USA Radioiodine (131I) is an effective treatment for feline hyperthyroidism, but optimal dosing to restore euthyroidism without inducing hypothyroidism remains unclear. Post‐treatment hypothyroidism can lead to azotemia and reduced survival. The objective of this prospective, longitudinal study was to compare the frequency of persistent hyperthyroidism and iatrogenic hypothyroidism in cats with mild‐to‐moderate hyperthyroidism (thyroxine [T4] > 4 but < 13 μg/dL) treated with either a low‐dose (2 milliCuries [mCi]; n = 131) or a standard‐dose (4 mCi; n = 37) of 131I. Serum concentrations of T4, thyroid stimulating hormone [TSH], and creatinine were assessed at 0, 3, and 6 months after treatment. Cats were categorized at each time interval as having persistent hyperthyroidism, euthyroidism (normal T4 [0.9‐4.0 μg/dL] and TSH [≤ 0.3 ng/mL]), overt hypothyroidism (low T4, high TSH), or subclinical hypothyroidism (normal T4 but high TSH) (Table). Categorical and continuous data were analysed using Chi‐Square and Wilcoxon Rank Sum, respectively, with P < 0.05 considered significant. At baseline, there were no differences in sex, age, or serum concentrations of T4, TSH, or creatinine between treatment groups. n Hyperthyroidism Euthyroidism Overt Hypothyroidism Subclinical Hypothyroidism 2 mCi 3 months 131 6 (5%) 95 (72%) 1 (1%) 29 (22%) 6 months 110 5 (5%) 80 (72%) 1 (1%) 24 (22%) 4 mCi 3 months 27 0 (0%) 9 (33%) 4 (15%) 14 (52%) 6 months 19 0 (0%) 4 (21%) 3 (16%) 12 (63%) Cats treated with 2 mCi 131I were significantly (P ≤ 0.01) less likely to develop overt hypothyroidism at 3 and 6 months compared to cats treated with 4 mCi. There was no significant difference in persistent hyperthyroidism between groups. Cats dosed with 4 mCi 131I had significantly (P ≤ 0.007) higher median creatinine concentrations at 3 (1.9 mg/dL) and 6 months (2.0 mg/dL) than did cats receiving 2 mCi (1.6 mg/dL and 1.7 mg/dL, respectively). In conclusion, most cats with mild‐to‐moderate hyperthyroidism were effectively treated with 2 mCi 131I with reduced risk of overt hypothyroidism compared to cats dosed with 4 mCi. EN04 ASSESSMENT OF RENAL FUNCTION IN HYPERTHYROID CATS MANAGED WITH HILL'S® PRESCRIPTION DIET® Y/D® FELINE Heather Vaske 1, Laura Armbrust1, Steven Zicker2, Dennis Jewell2, Greg Grauer1 1Kansas State University, Manhattan, KS, USA, 2Hill's Pet Nutrition, Topeka, KS, USA Previous studies demonstrated that glomerular filtration rate (GFR) declines with hyperthyroid treatment independent of the treatment modality. Hill's® Prescription Diet® y/d® Feline is a controlled iodine food used to manage feline hyperthyroidism, however, its effects on renal function have not been evaluated. The purpose of this study was to assess the effects of y/d® Feline on renal function in hyperthyroid cats. Hyperthyroidism was diagnosed in 15 client owned cats based on total thyroid hormone concentration (TT4) and compatible clinical signs. Baseline excretory renal function was assessed by plasma clearance of technetium diethylene‐triamine‐pentaacetate (GFR), serum creatinine (sCr), and serum symmetrical dimethyl arginine (SDMA) concentrations. Ultrasound was used to assess muscle mass (epaxial muscle diameter [EMD]). After an initial transition period to the food, it was then fed exclusively for 6 months and baseline parameters were re‐assessed. A t‐test was used to evaluate changes over time; a P‐value <0.05 was considered significant. Feeding y/d Feline for 6 months resulted in a significant decline in mean TT4 (185.5 vs. 60.5 nmol/l) and sCr (1.11 vs. 0.93 mg/dl). No change in GFR (2.31 vs. 2.26 ml/min/kg), SDMA (12.47 vs. 12.26 μg/dl), or EMD (left 1.56 vs. 1.62 cm and right 1.55 vs. 1.56 cm) was observed. SDMA better correlated with GFR than sCr (r2 0.50 vs. 0.30). Although not all cats became/remained euthyroid during the study period, management with y/d® Feline reduced TT4 without decreasing GFR or EMD. Compared with sCr, SDMA may be a better marker of renal excretory function in hyperthyroid cats. EN05 INCREASES IN SERUM BETA HYDROXYBUTYRATE IN CATS WITH CHRONIC KIDNEY DISEASE, HYPERTHYROIDISM, AND HEPATIC LIPIDOSIS Lisa Gorman, Jane Armstrong, Leslie Sharkey University of Minnesota, Saint Paul, MN, USA Ketones, including beta hydroxybutyrate (BHB), are produced in conditions of chronic negative energy balance and/or decreased glucose utilization. Little is known, however, about whether serum BHB concentrations increase in catabolic states other than diabetes mellitus in cats. The primary aim of this study was to determine if serum BHB concentrations are increased in cats with three potentially catabolic disease states: chronic kidney disease (CKD), hyperthyroidism (HT), and hepatic lipidosis (HL). Secondary aims were to determine if serum BHB concentrations are correlated with other clinical variables, and to determine if serum BHB concentrations are correlated with urine ketone measurement by dipstick, which primarily measures acetoacetate. Serum BHB concentrations were measured in samples drawn at admission in 28 client‐owned cats with CKD, 34 cats with HT, and 15 cats with HL, and were compared to a reference range established based on serum BHB concentrations in healthy cats. Variables of clinical relevance in each disease condition (including serum concentrations of creatinine, total thyroxine (T4), and serum activities of liver enzymes) were examined for relationships with serum BHB using Kendall's correlation. Results of urine ketone measurement by dipstick, when available, were compared to BHB measurement. Incer Serum BHB concentrations were above the reference range (>0.11 mmol/l) in 6/28 cats (21.4%) with CKD, 7/34 cats (20.6%) with HT, and 11/15 cats (73.3%) with HL. For the CKD and HT groups, this increase in serum BHB was mild in affected cats (0.12‐0.26 mmol/l for CKD, 0.12‐0.51 mmol/l for HT). For the HL group, this increase had a wider range (0.12‐2.78 mmol/l) and was more marked, as 3/15 HL cats had a serum BHB concentration >2 mmol/l. For all groups, the percentage of cats with serum BHB concentrations above the reference limit was significantly higher than 2.5%, which would be the percentage expected for a normal population of cats (p < 0.001 for all groups). Of the clinical variables evaluated, only total T4 in the hyperthyroid group correlated positively with BHB (P = 0.03). Of 44 cats with a urine dipstick performed within 24 hours of serum BHB sampling, none were positive for ketones; however, 12/44 cats had a serum BHB concentration above the upper limit of the reference range, and 2/44 cats had a serum BHB > 2 mmol/l. Increases in serum BHB concentrations occur commonly in cats with CKD, HT, and HL, and may provide a useful index of catabolism. Furthermore, serum BHB appeared to be more sensitive than urine ketone measurement in detecting ketonemia in this population of cats. Catabolic states other than diabetes mellitus must be considered when evaluating serum BHB concentrations in sick cats. EN06 INVESTIGATION OF DYSPROTEINEMIA IN HYPERTHYROID CATS AS A BIOMARKER OF CONCURRENT CHRONIC KIDNEY DISEASE (CKD) Tim Williams 1, Harriet Syme2, Jonathan Elliott2, Joy Archer1 1University of Cambridge, Cambridge, UK, 2Royal Veterinary College, London, UK Several studies have attempted to identify reliable biomarkers of CKD in hyperthyroid cats, with little success to date. Low plasma total globulin concentration was independently associated with the development of azotemia following treatment of hyperthyroidism in one study, which suggests that dysproteinemia is a marker of concurrent CKD. Characterisation of dysproteinemia is possible by serum protein electrophoresis (SPE). Further investigation of the changes that occur in the globulin fraction in hyperthyroid cats, and the possible association of these changes with the presence of concurrent CKD, is warranted, since this might provide a novel biomarker for CKD. Newly diagnosed non‐azotemic hyperthyroid cats (plasma total thyroxine (TT4)>55 nmol/l) and a group of healthy (control), non‐azotemic, euthyroid (TT4 < 40 nmol/l), non proteinuric (urine protein: creatinine ratio <0.4), geriatric (>9 years old) cats were recruited from five UK first opinion practices. Hyperthyroid cats were monitored for the development of renal azotemia (defined as plasma creatinine concentration >2 mg/dl with urine specific gravity <1.035) during a four month period following restoration of euthyroidism (TT4 10‐40 nmol/l). Agarose gel SPE was performed to calculate the serum concentrations of albumin, α1, α2, β and γ globulins. Non‐parametric statistics were used to compare baseline serum concentrations of α1, α2, β, γ and total globulin between hyperthyroid and control cats, between hyperthyroid cats which developed azotemia in the follow up period (pre‐azotemic) and those which remained non‐azotemic (at baseline), and in hyperthyroid cats before and after restoration of euthyroidism. Data are presented as median [25th, 75th percentile] and statistical significance was defined as P < 0.05. Hyperthyroid cats (n = 56) had higher serum α2 globulin concentrations (12.5 [10.9, 13.1] g/l vs. 9.8 [3.0, 11.4] g/l; P < 0.001) and lower serum γ globulin concentrations (11.4 [9.1, 13.3] g/l vs. 14.0 [12.4, 16.8] g/l; P = 0.001) than control cats (n = 26). Following treatment of hyperthyroidism, serum total globulin concentration increased (38.6 [35.4, 42.8] g/l vs. 42.3 [39.0, 45.7] g/l; P < 0.001), serum α2 globulin concentration decreased (12.5 [10.9, 13.9] g/l vs. 11.5 [10.1, 12.6] g/l; P < 0.001) and serum γ globulin concentration increased (11.4 [9.0, 13.3] g/l vs. 14.0 [12.4, 16.8] g/l; P < 0.001). Serum total protein concentration was higher in pre‐azotemic hyperthyroid cats compared to those which remained non‐azotemic (70.8 [67.3, 75.9] g/l, n = 21, vs. 68.1 [65.5, 72.2] g/l, n = 35; P = 0.049), however no significant differences in the serum concentrations of α1, α2, β, γ or total globulin were evident between the pre‐azotemic and non azotemic groups. Hyperthyroidism is associated with selective dysproteinemias of the α2 and γ globulin fractions, which mostly resolve following treatment, however these changes were not associated with the presence of concurrent CKD. The pathogenesis of these dysproteinemias requires further investigation to establish if they reflect altered production, metabolism or loss of proteins. EN07 EFFICACY OF PROZINCR INSULIN IN NAïVE AND INSULIN‐ESTABLISHED CATS USING CONTINUOUS INTERSTITIAL GLUCOSE MONITORING Cynthia Ward, Annette Louviere University of Georgia CVM, Athens, GA, USA Diabetes mellitus in cats is a relatively easily diagnosed endocrine disease that can be frustrating to treat and manage. ProZincR is a human recombinant PZI insulin that is particularly effective in cats and is also FDA‐approved for veterinary use. Previous studies have shown that ProZincR is effective in clinical control of diabetes mellitus in cats; however, the blood lowering characteristics of ProZincR have not been studied. Our objective in this study was to evaluate the blood glucose response to ProZincR insulin in newly‐diagnosed and stably‐treated diabetic cats using continuous interstitial glucose monitoring. Our hypothesis is that ProZincR is effective in controlling hyperglycemia in cats with diabetes mellitus. Ten cats recently diagnosed with diabetes mellitus that had not received any insulin therapy were recruited into the study (Group 1). Ten diabetic cats being treated with ProZincR for at least 2 months that were clinically stable according to their owners were also recruited (Group 2). Inclusion criteria consisted of a diagnosis of diabetes mellitus, availability of complete medical records, and current rabies vaccination. Exclusion criteria included presence of ketosis, hyperosmolarity, concurrent disease, non‐neutered status, steroid injection within the past 30 days, and current steroid treatment. A complete history, physical examination, CBC, chemistry profile, urinalysis, urine culture, T4, and fructosamine were obtained on each cat. Sensors for the Minimed GuardianR system were equilibrated and calibrated on each cat. Group 1 cats were started on ProZincR at 0.5 U/kg bid with a meal‐fed low carbohydrate diet. Group 2 cats were given ProZincR and fed according to their existing regime. Glucose curves were obtained over 72 hours with readings every 5 minutes. Gender included 8 MN, 2 FS and 7 MN, 3 FS for groups 1 and 2 respectively. The average age of group 1 cats was 9.8 yr (5‐14 yrs). The average age of group 2 cats was 10.5 yrs (9‐15 yrs). Group 1 cats were all DSH, Group 2 cats included 8 DSH, 1 DLH, and 1 Maine Coon. Group 1 cats had been diagnosed 7‐14 days previously except for 1 cat that had been diagnosed 3 months previously. Group 2 cats had been treated with ProZincR for 3 mos‐4 years. Four cats in group 1 were being fed low carbohydrate diets; 8 cats in group 2 were fed low carbohydrate diets with 9 of them being fed free choice. Glucose curve data from group 1 cats showed an average onset of action of ProZincR of 30 min (peakless‐60 min), duration of action of 18.5 hrs (11.5‐23 hrs), and time to peak activity of 9 hrs (peakless‐14 hrs). The glucose nadir ranged from 80‐200 mg/dl. Group 2 cats were receiving an average insulin dose of 0.57 U/kg (0.14‐0.99 U/Kg). 9 cats were receiving ProZincR bid, and 1 cat sid‐bid. In group 2 cats, the average onset of action of ProZincR was 37 min (30‐60 min), duration of action was11‐ > 12 hrs, and time to peak activity was 5.8 hrs (peakless‐7 hrs). The blood glucose nadir was 50‐180 mg/dl. Owner‐supplied follow‐up data on group 1 cats revealed that 7 cats were doing well on ProZincR, given twice daily, and 2 cats were being treated with once daily ProZincR. One cat became euglycemic for 3 weeks before requiring insulin again. Clinical signs, including plantigrade stance in 2 cats, improved or completely resolved. For the group 2 cats, 3 cats increased ProZincR dose and 2 cats decreased ProZincR dose as a result of glucose monitoring. ProZincR effectively lowers blood glucose and controls clinical signs in cats with diabetes mellitus for up to 4 years. The onset of action and duration of actions of ProZincR indicate bid therapy for optimal control. However, the duration of action or ProZincR in some cats is greater than 12 hours and it may be used as a once per day therapy if owners cannot comply with twice daily insulin administration. EN08 METABOLIC PROFILING OF DIABETIC CATS IN REMISSION Susan Gottlieb 2, Jacquie Rand1, Stephen Anderson3, John Morton4, Ziad Ramadan5, Daniel Dias6, Ute Roessner6 1School of Veterinary Science, The University of Queensland, Gatton, QLD, Australia, 2The Cat Clinic, Brisbane, QLD, Australia, 3School of Biomedical Science, The University of Queensland, Brisbane, QLD, Australia, 4Jemora Pty Ltd, Geelong, VIC, Australia, 5Nestle Purina Pet Care, St Louis, MO, USA, 6Centre for Molecular, Environmental, Genetic & Analytic MEGA Epidemiology, Melbourne School of Population and Global Health, The University of Melbourne, Melbourne, VIC, Australia Diabetic remission, defined as the ability to maintain euglycemia without insulin administration, occurs in over 80% of newly diagnosed diabetic cats, provided tight glycemic control is instituted early after diagnosis. However, approximately 25‐30% of cats in remission relapse, and require further insulin therapy to control blood glucose. The majority of diabetic cats in remission have impaired glucose tolerance, and 19% have impaired fasting blood glucose, indicating that these cats do not have normal glucose metabolism. Except for documentation of impaired glucose tolerance, the metabolic disturbances in diabetic cats in remission are poorly understood. The aim of this study was to determine metabolic profiles of cats in remission using gas chromatography mass spectrometry (GC‐MS), and to compare them to healthy control cats. The study was a prospective study in a feline‐only clinic. For inclusion, diabetic cats had to have achieved remission through insulin therapy, insulin withheld for a minimum of two weeks after remission, and had no clinical signs of diabetes present prior to testing. Twenty diabetic cats in remission and 22 non‐diabetic cats (frequency‐matched based on age and body condition score) participated in the study. A blood sample was collected after the cats were fasted for 18‐24 hours and sent for metabolomic analysis. Fasting blood glucose was measured for all cats, and a simplified glucose tolerance test (GTT) (glucose 1 g/kg) performed in all control cats and 19 of the 20 remission cats. Mean fasting blood glucose concentration and glucose concentration at 2 hours post glucose administration in a GTT were significantly different between remission and control groups. Twenty metabolites were identified that were significantly (p < 0.05) different between remission and control groups. The majority of amino acids (alanine, asparagines, glutamine, homoserine, leucine, methionine, phenylalanine, proline, serine, tyrosine and valine) that were significantly changed were decreased by 25 to 66%, and were negatively correlated with glucose parameters (casual blood glucose, fasting blood glucose, 2 h reading during GTT, and return to baseline during GTT). The exception was glycine, which had a more than two‐fold increase, and was positively correlated with glucose parameters. Three fatty acids were identified as being significantly different (after adjustment for body condition score). Hexadecenoic (palmitoleic) acid and octadecanoic (stearic) acid were decreased with moderate negative correlations to glucose parameters, and 1‐mono hexadecanoyl glycerol (1‐palmitoylglycerol) was increased, but had little correlation with glucose parameters. While the majority of cats in remission were able to maintain normal fasting blood glucose concentration, their fasting glucose concentrations were 30% higher than control cats when analysed by mass spectrometry. The sugar alcohol xylitol was decreased in remission cats and correlated negatively with all glucose parameters. The organic acid α‐ketoglutaric acid and the inorganic acid phosphoric acid were increased and decreased respectively, but were not significantly correlated with glucose parameters. Urea was also increased approximately 70% in remission cats compared to control cats, even after removal of 6 cats with IRIS stage ≥ 2 renal disease, and was positively correlated with both 2 hour glucose and time to return to baseline. In summary, the most prominent metabolic differences in diabetic cats in remission compared to control cats were significantly decreased glucogenic amino acids, which were negatively correlated with increased fasting glucose, suggesting increased consumption of glucogenic amino acids for gluconeogenesis. The increase in urea may reflect that urea is a by‐product of the glucose‐alanine cycle, whereby muscle provides substrate for hepatic gluconeogenesis. Pyruvate in muscle is converted to alanine and transported to the liver, where it is converted back to pyruvate (releasing urea) and used for gluconeogenesis or the TCA cycle. In conclusion, the results of this study show that in addition to fasting blood glucose concentration and glucose tolerance status, diabetic cats in remission have a range of metabolic disturbances compared to control cats, some of which may reflect increased gluconeogenesis. EN09 SAFETY AND EFFICACY ASSESSMENT OF A GLP‐1: GLARGINE COMBINATION FOR FELINE DIABETES Margaret Scuderi, Elisabeth Snead, Suraj Unniappan, Cheryl Waldner, Susan Mehain, Mariana Ribeiro, Donna Michosiw Western College of Veterinary Medicine, Saskatoon, Saskatchewan, Canada Diabetes mellitus (DM) is a common endocrine disorder in cats. Most cats have a form of DM equivalent to type‐2 diabetes (T2D) in humans. One new therapeutic strategy for T2D in humans involves use of synthetic incretin hormones, including Exenatide, a GLP‐1 hormone agonist. These can be used alone or as an adjunct treatment with insulin to improve overall glycemic control and promote weight loss. Use of Exenatide has recently been shown to be safe and effective at lowering blood sugar concentrations and promoting weight loss in healthy cats. Given the significant benefits seen in people with T2D, and the similar nature of the disease in cats, a clinical trial to assess Exenatide in diabetic cats was warranted. The primary objective of this study was to evaluate the efficacy and safety of a combination of a short‐acting GLP‐1 mimetic (Exenatide) with glargine insulin (Protocol A) compared to glargine with a placebo (Protocol B) for the treatment of DM in overweight cats. The effect of treatment protocol A versus B on serum concentrations of specific hormones that play a role in glycemic regulation (insulin and glucagon) or in altering insulin sensitivity and resistance (leptin and adiponectin) was also determined. A double blind, placebo‐controlled, crossover design study was preformed. Eight diabetic cats with a body condition score (BCS) ≥ 4/9 without any evidence of significant concurrent disease based on routine screening tests were recruited. All cats had been treated with glargine insulin and maintained a stable body weight for a minimum of one month prior to enrolment. Glycemic parameters monitored included serum fructosamine, fasting and postprandial blood glucose concentrations, mean blood glucose concentration for a blood glucose curve (BCG), and the total daily dose (TDD) of insulin required. Fasting and postprandial blood samples were collected for measurement of insulin and glucagon and fasting blood samples for leptin and adiponectin on the day 0, 7, 28, 42 of both arms of the study. All assays used in this study had previously been validated for use in cats. Treatment with Protocol A was well tolerated; only two cats (n = 8) displayed adverse signs that required a decrease in dose of Exenatide. These clinical signs included anorexia (n = 2), and hypoglycemia (n = 1). During a12‐hour BGC five cats experienced asymptomatic hypoglycemia (BG < 3.5), one on Protocol A, three on Protocol B, and one during both protocols (n = 8). Two cats (25%) went into diabetic remission during Protocol A (n = 8). The average change in the TDD of insulin from enrolment was 0.8 units (range‐ 0‐ 3 units); 0.9 units (range‐ 0‐ 3units) during Protocol A and 0.69 units (range‐ 0‐ 1units) during Protocol B. Three cats had a decrease in their insulin dose on Protocol A whereas only one cat had a decrease on Protocol B (n = 8). The fasting blood glucose was not significantly lower in cats receiving Protocol A versus Protocol B (P = 0.8). The average change in body weight of the cats was 0.6 kg. Seven cats (87.5%) lost weight on Protocol A (median= 0.66 kg, range‐ 0.1‐ 1.6 kg) and five (62.5%) cats gained weight on Protocol B (median‐ 0.48 kg, range‐ 0.14‐ 1.14 kg). There was no significant difference in any of the hormone concentrations evaluated for cats on Protocol A versus B. Only two cats experienced clinical signs that required decreasing the Exenatide dose during Protocol A and neither owners thought signs were severe enough to remove their cat from the study. While 87.5% (n = 8) of cats experienced weight loss during treatment Protocol A, this finding was not significant. Treatment with Protocol A did not change the concentration of hormones involved in glycemic regulation or insulin sensitivity. This study revealed that a treatment protocol including glargine insulin and a fixed dose of GLP‐1 mimetic is considered safe for diabetic cats. Since weight gain is a common problem of overweight diabetic cats treated with insulin, the weight loss experienced with Protocol A could be a significant benefit and warrants further evaluation of this treatment protocol. EN10 DIAGNOSIS OF CANINE HYPERADRENOCORTICISM IS ASSOCIATED WITH GENDER, AGE, BREED, AND COMORBID CONDITIONS Bianca Lourenço 1, Jessica Hoffman3, Kate Creevy1, Daniel Promislow2 1Department of Small Animal Medicine and Surgery, College of Veterinary Medicine, University of Georgia, Athens, Georgia, USA, 2Department of Pathology, University of Washington, Seattle, Washington, USA, 3Department of Genetics, University of Georgia, Athens, Georgia, USA This report describes a large canine patient population with a diagnosis of hyperadrenocorticism (HAC) at the time of death, and concurrent diagnoses and mortality associated with HAC in veterinary teaching hospitals. This retrospective study evaluated 1519 dogs with HAC included in a twenty‐year subset of the Veterinary Medical Database (1984‐2004). Signalment, HAC presence or absence, etiology of HAC (if described), frequency of comorbidities previously associated with HAC, and cause of death were assessed in dogs with and without HAC. The frequency of HAC steadily increased with increasing age at the time of death. HAC was more frequent in females, with no significant effect of sterilization status. HAC or related comorbidities were the cause of death of a minority (25.9%) of affected dogs. PDH (73.4%) was more common than FAT (25.8%) but FAT was over‐represented as a cause of death (80.9% of HAC deaths). HAC frequency was over‐represented in both expected (Miniature Poodle, Dachshund) and unexpected (Irish Setter, Bassett Hound) breeds. HAC dogs were at increased risk for concurrent diabetes mellitus, urinary tract infection, urolithiasis, hypertension, gall bladder mucocele, and thromboembolic disease compared with dogs without HAC. There was no significant difference in frequency of skin infections, or ligament or tendon ruptures between dogs with or without HAC. These results confirm that HAC is significantly associated with comorbid conditions, and that HAC is not a significant cause of mortality in dogs. Documented patterns now provide multiple targets for prospective clinical research and intervention. EN11 A NOVEL PROTOCOL FOR GLUCAGON STIMULATION TEST IN CATS USING RECOMBINANT DNA GLUCAGON Chen Gilor, Rebecca Glock, Shir Gilor The Ohio State University, Columbus, OH, USA The glucagon stimulation tests (GST) is an important tool in diabetes research. GST were described previously in cats, using a purified porcine glucagon formulation that is no longer available. The aim of this study was to re‐establish a protocol for GST using commercially available recombinant‐DNA glucagon (rdGlucagon). Initially, a dose of 500 mcg/cat, IV (over 30 seconds) was used in five healthy cats (as described by Kirk et. al AJVR 1993) however, lip‐smacking, abdominal contractions and occasional vomiting were observed 1‐2 minutes post injection. Consequently, slower IV push, intra‐muscular injections and lower doses were attempted. With 10 and 20mcg/kg IM no side effects were observed, however, the 10mcg/kg dose resulted in inadequate stimulation of blood glucose concentrations (BG). The 20 mcg/kg dose significantly increased BG from baseline (mean of ‐15 and 0 minutes pre‐glucagon, BG [mean ± sd] = 79.5 ± 6.6 mg/dL) to peak at 15 minutes (165.4 ± 24.2 mg/dL). BG gradually decreased at 25, 35 and 45 minutes until returning to baseline at 60 minutes. Plasma insulin concentrations also peaked at 15 minutes (baseline = 186.7 ± 116.6 ng/L, peak >800 ng/L [upper limit of the Mercodia feline insulin Assay]), were still significantly increased at 25 minutes (546.6 ± 213.9 ng/L) and returned to baseline at 35 minutes. In conclusion, an IM injection of 20 mcg/kg recombinant‐DNA glucagon in healthy cats resulted in a peak BG and insulin 15 minutes post injection without causing any side effects. This is a safe and effective alternative to the previously described glucagon stimulation test in cats. EN12 DAY‐TO‐DAY VARIABILITY OF SERIAL BLOOD GLUCOSE CURVES IN DIABETIC DOGS TREATED WITH TWICE DAILY NPH OR INSULIN DETEMIR Viviani De Marco1, Thais Grunho2, Chen Gilor 3 1University of Santo Amaro/Naya Especialidades, São Paulo, SP, Brazil, 2University of Santo Amaro, São Paulo, SP, Brazil, 3The Ohio State University, Columbus, OH, USA Day‐to‐day variability of serial blood glucose curves (BGC) impedes balancing control of clinical signs and avoidance of hypoglycemia in diabetics. Day‐to‐day variability differs between insulin formulations but was previously reported only for porcine lente in dogs. Here we report on the day‐to‐day variability of BGC in well controlled (based on stable clinical signs including body weight) dogs treated with human‐recombinant NPH (n = 5) or insulin detemir (n = 3). In each dog, four successive 12‐h BGC's were performed at home by the owners, during no more than a 28 days period. Insulin dose and meals (including quantities, type, time of administration etc.) as well as blood sampling times remained unchanged throughout this time period. Recorded BG parameters included: BG obtained just prior to insulin injections (BGfast0, BGfast12), maximum (BGmax), minimum (Nadir), mean (BGmean) and Delta [BGfast0‐Nadir]. Day‐to‐day variability was calculated as previously described (Fleeman & Rand, JAVMA 2003) except that for each dog, for each parameter, the 2 most extreme values (out of the 4 BGC's) were used to calculate maximal differences (Deltamax). For each parameter, the mean ± sd and CV (CV=SD/mean) of Deltamax were calculated for all dogs in each insulin group and presented below. Deltamax (mg/dL) Detemir (N = 3) NPH (N = 5) Mean SD CV Mean SD CV BGfast0 115.0 44.2 38.4 175.0 71.1 40.6 BGfast12 95.0 49.8 52.4 214.8 88.9 41.4 BGmax 82.3 40.0 48.5 100.0 82.6 82.6 Nadir 28.3 13.1 46.3 85.6 56.8 66.3 BGmean 45.4 14.4 31.7 87.5 31.7 36.2 Delta [BGfast0‐Nadir] 109.7 36.4 33.2 157.0 43.4 27.6 In conclusion, for each parameter, mean maximal differences were smaller between BGC's in the detemir group compared to the NPH group, suggesting overall lesser day‐to‐day variation in insulin action and greater confidence when interpreting the results of BGC's. EN13 EFFECT OF GLUCOSURIA ON URINE SPECIFIC GRAVITY (USG) IN DOGS Ellen Behrend 1, Samuel Mueller1, Dennis Chew2, Hollie Lee1 1Auburn University, Auburn, AL, USA, 2The Ohio State University, Columbus, OH, USA Diabetic patients that are historically polyuric and polydipsic can have concentrated urine. Addition of 1 g/dL glucose is expected to change the specific gravity of water by 0.003‐0.005. A commonly held belief, therefore, is that concentrated USG in a diabetic patient is due to the presence of glucose, and renal concentrating ability cannot be accurately assessed if glucosuria is present. However, the effect of glucose on USG has not been evaluated. Thus, the purpose of our study was to assess the change in USG in relation to the degree of glucosuria. Consecutive canine urine samples submitted for urinalysis that were yellow and had no glucose on dipstick evaluation were used. Sample supernatants were pooled based on USG to obtain a final volume of ≥ 3 mL. Pools were divided into ranges by USG: USG of 1.005‐1.009, 1.010‐1.014, 1.015‐1.019, 1.020‐1.024, 1.025‐1.029, 1.030‐1.034, 1.035‐1.039, and 1.040‐1.045 and named range A, B, C, D, E, F, G, H and I, respectively, with five pools in each range. A solution of 0.1 g/mL glucose was added to an aliquot of each pool to obtain a final concentration of 2,400 mg/dL. Serial dilutions were performed with the remainder of each pool to obtain glucose concentrations of 1,200, 600, 300, 150 and 50 mg/dL. The USG was measured in the undiluted samples and all spiked samples with a handheld refractometer (Reichert TS 400). For statistical comparisons, the change in USG from the undiluted sample was calculated and divided by the final glucose concentration (in g/dL) in the sample. A repeated measures ANOVA on ranks was used to compare the change in USG per g/dL glucose between the USG ranges. Post hoc comparisons were made with the Tukey test. Median values are reported. Significance was set at the p < 0.05 level. With addition of 50 mg/dL or 150 mg/dL glucose, the change in USG per g/dL glucose was not significantly different between ranges. With addition of 300 mg/dL glucose, there was an overall difference between ranges (P = 0.005) but individual pairwise differences were not identified. With addition of 600 mg/dL glucose, the change in USG per g/dL glucose was significantly different between ranges A (change = 0.003) and H (change = ‐0.001). With addition of 1,200 mg/dL glucose, the change in USG per g/dL glucose was significantly different between ranges A (change = 0.003), G (change = 0), H (change = ‐0.001) and I (change = 0). With addition of 2,400 mg/dL glucose, the change in USG per g/dL glucose was significantly different between ranges A (change = 0.003), G (change = 0), H (change = ‐0.001) and I (change = ‐0.001). Therefore, the change in USG per g/dL glucose is significantly affected by the USG of the urine without glucose. As the initial USG in non‐glucosuric samples increases, the change created by the addition of glucose diminishes significantly. In addition, the changes in USG created by the presence of glucose are small and clinically insignificant. Thus, even in the presence of glucosuria, an accurate estimation of renal concentrating ability can be made. EN14 A NOVEL CORTISOL BASED METHOD FOR MONITORING TRILOSTANE THERAPY IN DOGS WITH HYPERADRENOCORTICISM Laura Cosgrove, Timothy Parkin, Ian Ramsey University of Glasgow, Glasgow, UK The aim of this study was to compare a new cortisol based method to 1 hour post ACTH cortisol measurements in the monitoring of dogs receiving trilostane. Dogs diagnosed with hyperadrenocorticism that were being treated with trilostane were recruited from first opinion and referral practices. Serum cortisol was measured at 3 time points in each dog; pre‐trilostane (pre‐pill), 3 hours post‐trilostane (post‐pill) and 1 hour post‐ACTH (post ACTH). Clinical control was assessed using the responses of owners to a questionnaire. Each of nine questions was given a score, with a maximum of 26 points available. The total scores were then divided into one of four clinical groups: over‐controlled (0 points), well‐controlled (1 to 6 points), moderately‐controlled (7‐12 points) and poorly‐controlled (≥13 points). A novel algorithm that combined the pre‐pill and post‐pill cortisol measurements was devised (pre‐post). Using the results of the post ACTH stimulation cortisol or the new pre‐post method, dogs were also placed into four categories of endocrine control. Weighted Kappa was calculated to assess the level of agreement between the actual clinical control, as categorised using the owner's questionnaire score and the categorisation of endocrine control according to either the post ACTH cortisol results or the pre‐post cortisol result. Pearsons correlation co‐efficient was used to compare the three absolute cortisol results to each other and to the total score from the owners’ questionnaire. Univariate and multivariate regression analysis was performed using dose frequency, total daily dose, total duration of therapy, duration of therapy at current dose, signalment, visit number and weight as variables to ascertain their effect on the three cortisol results. In total 116 tests were analysed. 12 tests did not have a post ACTH cortisol and 19 tests had a higher pre‐pill than the post‐pill cortisol. Clinical control, according to the results of owner's questionnaire, was correctly categorised using the post ACTH cortisol in 31.73% of tests, compared to the pre‐post method which was correct in 50.52% tests. Both methods correctly recognised the one over‐controlled dog that had been identified by the owner's score. The weighted kappa for the endocrine control using post‐ACTH cortisol compared to the owner questionnaire score categories was 0.17. In contrast the weighted kappa for pre‐post trilostane categories compared to the owner questionnaire score categories was 0.39, indicating better agreement. There were significant correlations between the three cortisol concentrations with the best correlation being seen between then pre‐pill and post‐pill (r = 0.760, p < 0.000) results. There were also significant but weak correlations between the pre (r = 0.399, p < 0.000), the post (r = 0.350, p < 0.000) and the post ACTH (r = 0.246, p < 0.012) cortisol and the total score from the owner questionnaire. On multivariate analysis male neutered dogs were found to have significantly increased pre‐pill cortisol (p < 0.0027) and post‐pill cortisol (p < 0.044) when compared to female entire dogs and all dogs respectively. When dogs were receiving the current dose for more than 3 months, the cortisol was significantly lower at both the post‐pill (P < 0.037) and post ACTH (P < 0.000) cortisol time‐points. None of the other variables were associated with an effect on the three cortisol results. The novel algorithm of pre‐post trilostane cortisol better reflected the clinical control, as evidenced by the owners’ questionnaire, of hyperadrenocorticism than the post ACTH stimulation cortisol. The pre‐post trilostane cortisol should be further investigated as a monitoring tool for dogs receiving trilostane. EN15 EFFECT OF NON‐THYROIDAL ILLNESS ON SERUM CONCENTRATIONS OF T4, FREE T4 AND THYROID STIMULATING HORMONE IN CATS Danielle Davignon 1, John Lucy1, John Randolph1, Jan Scarlett1, Mark Peterson2 1Cornell University College of Veterinary Medicine, Ithaca, NY, USA, 2Animal Endocrine Clinic, New York, NY, USA In cats, non‐thyroidal illness (NTI) may result in decreased total thyroxine (T4) concentration, similar to the “euthyroid sick syndrome” in dogs. The pathogenesis of this phenomenon is not completely understood, but may involve suppression of thyroid stimulating hormone (TSH) and reduced thyroid hormone protein binding. Although high TSH and normal T4 concentrations have been reported in humans with chronic kidney, liver, or heart disease, low serum concentrations of T4 and TSH commonly develop with severe NTI in humans. Serum TSH concentrations have not been evaluated in cats with NTI, except for a small number of cats with mild chronic kidney disease in which TSH levels were normal. The objective of this study was to measure TSH in a larger series of cats suffering from a variety of mild‐to‐severe NTI to determine if the spectrum of TSH concentrations reported in humans also develops in cats. Serum T4, free T4 (FT4) by equilibrium dialysis, and TSH concentrations were measured in 120 clinically healthy cats and 46 cats presenting for evaluation of NTI. Median concentrations of T4, FT4, and TSH were compared using the Mann‐Whitney U test with P ≤ 0.05 considered significant. In sick cats, illness was graded as mild (outpatient exam; n = 31), moderate (short hospitalization for treatable condition; n = 6), or severe (intensive hospitalization associated with substantial morbidity/mortality; n = 9). Diseases in sick cats included cardiac (n = 11), endocrine (n = 4), hepatic (n = 2), infectious (n = 2), neoplastic (n = 11), renal (n = 5), respiratory (n = 1), and other (n = 10) disorders. Median serum concentrations of T4 (1.5 μg/dL vs. 2.0 μg/dL) and FT4 (26 pmol/L vs. 32 pmol/L) were significantly (P < 0.05) lower in the 46 sick cats compared to the healthy cats, respectively. However, no significant difference in median TSH concentration was detected between ill and healthy cats (Figure 1). Compared to the 31 sick cats with mild disease, the 9 cats with severe illness had significantly (P < 0.05) lower median serum concentrations of T4, FT4, and TSH (Figure 2). In one cat with mild neoplastic disease, high TSH concentration with normal T4 and FT4 was identified (Figures 1 & 2). In conclusion, our data demonstrate that median serum concentrations of T4 and FT4 are lower in sick compared to healthy cats, and these parameters may be inversely correlated with disease severity. Overall, TSH concentrations in sick cats were not significantly different from healthy cats. However, TSH levels in cats may decrease in severe illness, and may occasionally be high, similar to findings in humans with NTI. EN16 EVALUATION OF THE USE OF BASELINE CORTISOL TO MONITOR TWICE‐DAILY TRILOSTANE THERAPY IN DOGS WITH SPONTANEOUS HYPERADRENOCORTICISM Andrew Woolcock, Andrew Bugbee, Ann Rychlicki, Kate Creevy University of Georgia, Athens, GA, USA The objective was to evaluate the use of baseline cortisol to predict the results of ACTH‐ stimulated cortisol concentration in a population of dogs with pituitary‐dependent hyperadrencorticism (PDH) being treated with twice‐daily trilostane (Vetoryl®). Medical records for dogs with PDH treated with twice‐daily trilostane therapy between 2008 and 2012 were reviewed retrospectively. Results of ACTH stimulation tests were classified as indicating inadequate, acceptable, or excessive suppression of cortisol production. Details regarding each clinical visit were recorded including comorbid conditions, clinical signs, and clinician decision regarding trilostane dose (increase, no change, or decrease). Baseline cortisol concentrations were evaluated for correlations between ACTH‐stimulated cortisol concentrations and the clinician's decisions, with determination of predictive values. Twenty‐two dogs with a total of 109 ACTH stimulation tests were evaluated. Use of a baseline cortisol concentration cutoff of 3.2 μg/dL or greater resulted in a negative predictive value of 100% that the ACTH‐stimulated cortisol concentration would indicate excessive suppression. However, 21.8% of the ACTH stimulation tests with baseline concentrations above this cut‐off had post‐stimulation cortisol concentrations that were less than baseline. In addition, there was substantial overlap of the baseline cortisol concentrations when compared to the clinician's decision regarding trilostane dose. There exists a baseline cortisol concentration in this population of patients that can predict an ACTH‐stimulated cortisol greater than 2.0 ug/dL. However, this cut‐off does not guarantee an acceptable adrenal reserve capacity or control of clinical signs, therefore routine use of baseline cortisol as the sole monitoring diagnostic in trilostane‐treated PDH patients cannot be recommended. EN17 BLOOD PRESSURE IN HYPERTHYROID CATS IN COMPARISON TO AGE MATCHED CONTROLS USING HIGH DEFINITION OSCILLOMETRY Sherisse Sakals, Anthony Carr Western College of Veterinary Medicine, University of Saskatchewan, Saskatoon, Saskatchewan, Canada Hypertension is a common finding in older cats. A variety of underlying reasons for hypertension have been suggested including hyperthyroidism, where it is estimated around 20% are hypertensive at diagnosis. The literature to date has predominantly consisted of studies using Doppler technology where diastolic blood pressures are not determined. In addition these studies did not use an age matched control group to assess the data. Blood pressure was measured using an HDO (S+B MedVet, Babenhausen, Germany) oscillometric device. Tracings were visually evaluated using the provided software to assess the blood pressure curves for quality (adequate amplitudes, lack of significant artifact). Cats were measured with minimal or no restraint with the cuff placed on the tail. A series of readings were taken till at least 3 verified readings were obtained. A total of 42 hyperthryoid and 18 age matched non‐hyperthyroid controls were available for analysis. Systolic blood pressure (hyperthyroid mean 158 ± 19 mmgHg; control 154 ± 16 mmHg), diastolic blood pressure (hyperthyroid mean 88 ± 13; control 84 ± 12), mean blood pressure (hyperthyroid mean 113 ± 13; control 109 ± 12) and heart rate (hyperthyroid mean 203 ± 39; control 202 ± 24) were not different based on paired T‐test. When cat's blood pressure values were categorized using the ACVIM Consensus Statement values for risk of target organ damage (systolic; mild, 150‐159, moderate 160‐179, severe > 180: diastolic; mild 95‐99, moderate 100‐119, severe >120) there was a significant difference between the groups only for systolic blood pressures using chi‐square (P = 0.03). According to this classification in the hyperthyroid group there were 8 severe, 8 moderate and 12 mild cases whereas in the control group it was one case of severe, 9 of moderate and 1 of mild. In the hyperthyroid group for diastolic blood pressure 9 moderate and 3 mild cases were present whereas in the control group it was 2 cases of moderate and 1 case of mild. The data shows that as a group older cats and hyperthryoid cats have similar blood pressures with a similar percentage of cats being considered at risk of target organ damage (61% of control cats and 67% of hyperthyroid cats based on systolic blood pressure) however in the hyperthyroid group there is a higher percentage of cats with severe risk of target organ damage. Diastolic risk of target organ damage was more common in the hyperthyroid group (29% vs. 17%), though generally only moderate or mild. EN18 EPIDEMIOLOGIC STUDY OF ADRENAL TUMORS IN DOGS: 49 CASES (2012–2014) Viviani De Marco 4, Ubukata Rodrigo2, Vanessa Matsumoto1, Fabricio Machado4, Marcia Kahvegian2, Andre Romaldini3, Thais Fernandes2, Rogerio Soila2, Nilson Kage4, Priscila Viau2 1University of Santo Amaro, São Paulo/SP, Brazil, 2PROVET, São Paulo/SP, Brazil, 3Pet Care, São Paulo/SP, Brazil, 4NAYA Especialidades, São Paulo/SP, Brazil The aim of this study was to evaluate epidemiological characteristics of adrenal tumors (AT) in dogs. A retrospective study was conducted based on clinical records considering age, breed, sex, clinical signs, laboratorial findings, hormonal tests, medical or surgical treatment, histological findings and survival from dogs diagnosed with AT between January 2012 and November 2014 at the Naya Especialidades Veterinary Clinics. Among 49 dogs diagnosed with AT, 37 (76%) were female and the mean age was 11.3 ± 2.5 years. Forty‐four (89,79%) dogs were purebred, mainly Shih Tzu (11), Lhasa apso (6) and Dachshund (6). The mean body weight was 12.37 ± 11.6 kg (Median 7 kg; range, 3.4 to 60 kg). Body weight was ≤ 10 kg in 69.4% of dogs. Polyuria, polydipsia (59.2%), polyphagia (59.2%) and abdominal distension (36.7%) were the most frequent clinical signs. Increased ALP (86.7%) and ALT (40%), hyperlipidemia (45.6%), low urinary density (57.5%), thrombocytosis (30.2%) and hypertension (51.4%) were reported. A diagnosis of adrenal‐dependent hyperadrenocorticism on the basis of nonsuppressed serum cortisol concentration by use of an LDDS test or an inappropriately high serum cortisol concentration after ACTH stimulation was observed in 48,5% and 38,8%, respectively. Forty‐eight per cent of dogs had right AT, 36.7% left and 22.4% bilateral. Adrenalectomy was performed on 20 dogs (40,8%). The histological findings revealed adenoma in 65%, carcinoma in 20% and pheocromocitoma in 15% of dogs. Twenty dogs were treated only with trilostane and 8 had no therapy. Of the 49 dogs, 45 (92%) are still alive in a 2 years follow up. Primary AT incidence may be increasing, especially in small dogs (< 10 kg). Cortex adrenal tumors are more common, but some animals may be asymptomatic and have negative LDDS and ACTH stimulation test. Therapy with adrenalectomy or trilostane should be encouraged. EN19 ASSOCIATION OF CANINE ACUTE PANCREATITIS WITH HYPERTRIGLYCERIDEMIA AND ADRENOMEGALY: A RETROSPECTIVE STUDY (2012–2014): 91 CASES Eric Vieira Januário 1, Márcia Marques Jericó2, Melissa Sanches Giacon1, José Henrique Hildebrand Grisi‐Filho3, Carla Alice Berl1, Marcelo Quinzani1 1Pet Care Veterinary Hospital, São Paulo, São Paulo, Brazil, 2Anhembi Morumbi University, São Paulo, São Paulo, Brazil, 3Veterinary Medicine College and Zootecny of São Paulo University, São Paulo, São Paulo, Brazil Pancreatitis is the most common exocrine pancreatic disease in dogs and has a multifactorial etiology. Endocrine abnormalities and hyperlipidemia are considered risk factors. The aim of this study was to investigate the association of acute pancreatitis with diabetes mellitus (DM), confirmed hyperadrenocorticism (HAC), possibility of HAC (adrenomegaly on ultrasound without previous diagnosis of HAC), hypothyroidism (HYPOT), hypertriglyceridemia (HYPERTRI) and hypercholesterolemia (HYPER CHOL). A retrospective study was performed in dogs admitted to a private veterinary hospital in São Paulo, Brazil. Dogs with symptoms and sonographic diagnosis of pancreatitis in the years 2012 and 2013 (group 1, n = 50) and control group dogs without sonographic pancreatic alterations, but with chronic kidney disease from 2013 to 2014 (group 2, n = 41) were studied. The past history was evaluated for the diagnosis of diabetes (blood glucose > 200 mg/dL, positive glycosuria), HAC (cortisol after low dose dexamethasone > 1.4 ug/dL or post ACTH > 21 ug/dL) and HYPOT (thyroid‐stimulating hormone > 0.5 ng/mL and dialysis free thyroxine < 0.82 ng/dL). Fasting triglyceride and cholesterol measurements during hospitalization identified the presence of HYPERTRI (triglycerides > 150 mg/dL) and HYPERCHOL (cholesterol > 300 mg/dL). Twenty‐one (42%) dogs in group 1 had HYPERTRI, 18 (36%) had HYPERCHOL, 11 (22%) had adrenomegaly, nine (18%) were diagnosed with DM, five (10%) were diagnosed with HAC, two (4%) were diagnosed with HYPOT. In group 2, 12 (29.26%) dogs had HYPERCHOL, five (12.19%) had HYPERTRI, two (4.87%) had adrenomegaly, two (4.87%) were diagnosed with HAC, two (4.87%) were diagnosed with HYPOT, one (2.43%) was diagnosed with DM. DM, adrenomegaly and HYPERTRI showed significant association (p ≤ 0.05) with pancreatitis. The inclusion of the variable DM in the logistic regression model showed no significance, mainly due to colinearity with the variable HYPERTRI. Dogs with HYPERTRI and adrenomegaly had increased odds ratio (OR=5.3 and 7.2 respectively; 95% confidence intervals) for pancreatitis. Results in this study suggested that HYPERTRI and possibility of HAC (represented by adrenomegaly) are associated with pancreatitis in dogs. EN20 Association between Endocrinopathies and Hyperlipidemia In 104 Dogs With Acute Pancreatitis: A Retrospective Study Hakhyun Kim, Hye‐Sun Kim, Minjii Lee, Ji‐Houn Kang, Gonhyung Kim, Dongwoo Chang, Mhan‐Pyo Yang College of Veterinary Medicine, Chungbuk National University, Cheongju, Chungbuk, Korea An association has been suggested between developing acute pancreatitis (AP) and concurrent endocrinopathies including diabetes mellitus (DM), hyperadrenocorticism (HAC) and hypothyroidism, although no conclusive data are available to confirm it. Moreover, there has been much debate whether hormone imbalances can directly affect the development of AP and/or whether hyperlipidemia induced by endocrinopathies might play a role. The objectives of the present study were to examine the prevalence of AP in dogs with endocrinopathies and to examine the association between those endocrinopathies and hyperlipidemia in dogs with AP. Medical records of 104 dogs newly‐diagnosed with AP from 2012 to 2014 were examined for the presence or absence of HAC, DM, and hypothyroidism. Two age‐ and breed‐matched randomized controls for each AP dog were examined for the same endocrinopathies. A matched case‐control analysis was performed, and the association between hyperlipidemia and endocrinopathies in the AP group was subsequently evaluated using binary logistic regression analysis. In dogs with HAC, the odds of AP were 4.55 times that of dogs without HAC, and in dogs with DM, the odds were 12.44 times that of dogs without DM. In addition, there was a significant association between the presence of DM and hyperlipidemia in the AP group. This study showed that hyperlipidemia caused by DM was associated with AP in dogs, but an association was not observed between HAC and hyperlipidemia in dogs with AP. However, further study will be necessary to clarify the prevalence of AP in dogs with DM and HAC. EN21 SERUM ADIPOKINE CONCENTRATIONS IN DOGS WITH MYXOMATOUS MITRAL VALVE DISEASE Hye‐Sun Kim, Ji‐Houn Kang, Eui‐Bae Jeung, Mhan‐Pyo Yang College of Veterinary Medicine, Chungbuk National University, Cheongju, Chungbuk, Korea Dysregulation of adipokines has been described in humans with cardiovascular disease, but information is lacking on concentrations of circulating adipokines in dogs with myxomatous mitral valve disease (MMVD). The objective of this case‐controlled observational study was to determine whether serum adipokine concentrations differed between healthy dogs and dogs with MMVD. Initially, a preliminary study was performed on 21 client‐owned dogs with MMVD and 30 healthy dogs with the same body condition score, and several adipokines were measured. Based on the preliminary results, serum leptin and adiponectin concentrations were re‐evaluated in 46 dogs newly‐diagnosed with MMVD which were divided into 3 groups according to the International Small Animal Cardiac Health Council (ISACHC) classification. In the preliminary study, the serum leptin and adiponectin concentrations differed significantly between the dogs with MMVD and the healthy dogs, regardless of whether or not congestive heart failure signs were absent, but there were no significant differences in the concentrations of interleukin (IL)‐1β, IL‐6, IL‐10, IL‐18, tumor necrosis factor‐α, resistin and visfatin. In the subsequent study, the dogs classified as ISACHC class III significantly showed the highest serum concentrations of leptin and adiponectin. These results indicated that dogs suffering from progressed MMVD showed increases in the serum concentrations of leptin and adiponectin. Our results suggest that the dysregulation of adiponectin and leptin might at least play a role in the pathogenesis of MMVD in dogs. However, further studies will be necessary before an association between adipokines and the progression of MMVD can be clearly understood. EN22 EVALUATION OF THE ISOLATION AND FUNCTIONALITY OF PANCREATIC ISLETS IN CATS Ines Rito Brandao1, Tania Garavello1, Eric Zini2, Claudia E. Reusch2, Thomas A. Lutz1, Melania Osto1 1Instute of Veterinary Physiology, University of Zurich, Zurich, Switzerland, 2Clinic for Small Animal Internal Medicine, University of Zurich, Zurich, Switzerland Diabetes mellitus is a common endocrinopathy in cats, characterized by impaired beta‐cell function and loss. Most of the current knowledge on the molecular mechanisms that negatively affect beta cell function and viability has been obtained in vitro in species other than cats. Although feline diabetes is similar to human type 2 diabetes, species‐dependent differences in the pathophysiology of the diseases may exist. Therefore, by establishing a method for the isolation and culture of pure pancreatic islets in cats, we aim at providing an in vitro tool for specific diabetes research in cats. Cats that died or were euthanized because of end‐stage disorders other than pancreatic disease were included in the study. Initial tests evaluated the best pancreatic infusion technique. Pancreata from nine cats were infused within one hour from death with 60 ml Collagenase Type IV (2 mg/ml, Worthington) through the major duodenal papilla. The perfused pancreas was then digested in a shaking water bath at 37 °C. The degree of dissociation of islets from acinar tissue was checked every 5 min by examination with dithizone which specifically stains Zn‐complexes in beta cells under a phase‐contrast microscope. Collagenase digestion was stopped when nearly 50% of islets were free of acinar tissue. Further purification of islets was achieved by using a filtration method and discontinuous density gradient solutions (1.077 and 1.100 g/cm3; Biochrom). The quality and purity of the final islet product was assessed by DTZ staining. Islet viability was assessed by visual assessment of the membrane integrity of islet cells after double fluorescent staining with fluorescein diacetate and propidium iodide (FDA/PI). In vitro functional activity of purified islets was evaluated by measuring the amount of insulin release (Mercodia Feline Insulin) upon glucose stimulation. Perfusion of the pancreas in situ through the major duodenal papilla allowed the complete perfusion of the pancreas compared to previously described ex situ techniques, and it markedly improved islet yield and purity. Digestion time ranged from 60′ to 90′. Islet yield, purity and viability were directly dependent on the proportion of perfused pancreas, with pancreas perfused in situ via the major papilla having higher quantity and quality of islets compared to all other infusion methods. After digestion, feline pancreatic islets remained viable for 5 days. The amount of insulin secreted upon glucose stimulation was comparable with that achieved in mice studies. The current study successfully optimized the isolation and purification of feline pancreatic islets. Further, the functional potency of islet upon glucose stimulation was assessed for the first time in this species. According to these findings, the yield of qualitatively and functional islets may provide a promising source of islets for diabetes research in cats EN23 EFFECT OF ADMINISTRATION WITH MIRTAZAPINE ON THE HYPOTHALAMIC‐PITUITARY‐ADRENOCORTICAL AXIS IN HEALTHY BEAGLE DOGS Sukhwan Lee, Ji‐Houn Kang, Hye‐Sun Kim, Bo Hye Shin, Byeong‐Teck Kang, Gonhyung Kim, Dongwoo Chang, Ki‐Jeong Na, Mhan‐Pyo Yang College of Veterinary Medicine, Chungbuk National University, Cheongju, Chungbuk, Korea Mirtazapine, a noradrenergic and specific serotonergic antidepressant, has been shown to suppress the hypothalamic‐pituitary adrenocortical (HPA) axis in humans, but little information is available on the effect of mirtazapine on the HPA axis in dogs. The objective of the present study was to examine whether, in healthy beagle dogs, the endogeneous adrenocorticotropic hormone (ACTH) and cortisol concentrations was affected by oral administration with mirtazapine. In 12 laboratory beagle dogs, oral administration with mirtazapine (1 mg/kg) was maintained once daily for 14 days. The corticotropin releasing hormone (CRH) stimulation test and ACTH stimulation test were performed before administration, and for 7 days and 14 days after the administration, respectively. Circulating concentrations of cortisol, ACTH, serotonin and leptin were measured. The administration with mirtazapine significantly reduced the post‐CRH ACTH concentrations but not the cortisol concentrations. Post‐ACTH cortisol concentrations were significantly reduced by the administration with mirtazapine, but no significant differences in the basal ACTH and cortisol concentrations were revealed. Serum serotonin concentrations were significantly increased by the administration of mirtazapin, and leptin concentrations were reduced significantly during mirtazapine administration periods; the lowest levels were observed at day 7. These results suggested that mirtazapine had a short‐term suppressive effect on the HPA axis in healthy dogs. In addition, serum serotonin and leptin levels might be influenced by mirtazapine administration. GI01 EFFECT OF GASTROINTESTINAL PARASITES AND VIRAL AGENTS ON FECAL S100A12 CONCENTRATIONS IN PUPPIES Romy Heilmann 1, Aurelien Grellet2, Niels Grützner3, Jan Suchodolski1, Jörg Steiner1 1Texas A&M University, College Station, Texas, USA, 2Royal Canin, Aimargues, France, 3Vetsuisse University Bern, Bern, BE, Switzerland Previous data suggest that S100A12 may be clinically useful as a biomarker of inflammation in both humans and dogs. In dogs with chronic gastrointestinal (GI) disease, fecal S100A12 concentrations were increased and correlated with the severity of clinical disease, endoscopic changes observed in the intestine, and the severity of histologic lesions in the colon. Higher fecal S100A12 concentrations also correlated with a negative outcome in dogs with inflammatory bowel disease. However, the effect of GI pathogens on fecal S100A12 concentrations in dogs has not yet been reported, and this information will be important for any further evaluation of the clinical utility of fecal S100A12 as a marker for GI inflammation. Thus, the aim of this study was to evaluate the effect of the presence and numbers of parasites as well as viral agents that are known to cause GI signs in dogs on fecal S100A12 concentrations in a large cohort of puppies. Spot fecal samples were collected from 307 puppies (median age [range]: 7 [4−13] weeks; 29 different breeds) in French breeding kennels, and fecal scores (semiquantitative system; scores 1−13) were assigned. Fecal samples were tested for Isospora spp. (I. canis and I. ohioensis), Toxocara canis, Giardia sp., as well as canine coronavirus (CCV) and parvovirus (CPV) by quantitative real‐time PCR (qPCR). S100A12 concentrations were measured in all fecal samples using an established and validated in‐house RIA. Statistical analysis was performed using non‐parametric 2‐group or multiple‐group comparisons and non‐parametric correlation analysis. Fecal S100A12 concentrations ranged from <24−14,363 ng/g. Significantly higher S100A12 concentrations were seen in dogs infected with Isospora spp. (P = 0.038) but not those infected with T. canis (P = 0.356), Giardia (P = 0.545), CCV (P = 0.752), or CPV (P = 0.861). Also, the numbers of I. canis (P = 0.196), I. ohioensis (P = 0.172), T. canis (P = 0.345), Giardia (P = 0.545), CCV (P = 0.454), and/or CPV (P = 0.384) did not influence fecal S100A12 concentrations. Age was not found to affect fecal S100A12 (P = 0.236) concentrations, but fecal score and S100A12 were inversely correlated (Spearman ρ=‐0.242; p < 0.001). We conclude that, independent of their number, the presence of parasites of the genera Isospora spp. can lead to increased fecal S100A12 concentrations, whereas Toxocara canis, Giardia sp., and the viral pathogens CCV and CPV do not affect fecal S100A12 results in these puppies. Thus, an infection with Isospora spp. appears to be an important differential for dogs with an increased fecal S100A12 concentration. Such an etiology needs to be ruled out prior to using fecal S100A12 concentration as a marker for the diagnosis/monitoring of patients with chronic GI inflammation. The association between fecal score and S100A12 levels and its clinical implication, and whether the current results translate to those in adult dogs, warrant further research. GI02 PROLONGED ADMINISTRATION OF PREDNISOLONE ALONE OR IN COMBINATION WITH AZATHIOPRINE ALTERS SERUM CANINE PANCREATIC LIPASE IMMUNOREACTIVITY (SPEC CPL™) IN DOGS Hiroshi Ohta, Tomoya Morita, Nozomu Yokoyama, Tatsuyuki Osuga, Sue Yee Lim, Keitaro Morishita, Kensuke Nakamura, Mitsuyoshi Takiguchi Graduate School of Veterinary Medicine, Hokkaido University, Sapporo, Hokkaido, Japan Canine‐specific pancreatic lipase immunoreactivity (cPLI) is considered the most specific serum biomarker for the diagnosis of pancreatitis in dogs. It has been reported that 4 weeks of oral administration of prednisone (2.2 mg/kg, q 24 h) to 6 dogs did not significantly affect serum cPLI concentrations (Steiner et al, 2009). However, the effect of long‐term exogenous glucocorticoid administration with or without azathioprine on cPLI concentration is unknown. The purpose of this study was to determine whether prolonged exogenous glucocorticoid administration with or without azathioprine alters the cPLI concentration in dogs with immune‐mediated disease. Sixteen client‐owned dogs with newly diagnosed immune‐mediated disease were enrolled between October 2012 and October 2014. Dogs with clinical signs of pancreatitis (e.g. vomiting, diarrhea, or anorexia) during the study period, increased cPLI (SPEC cPL test, SPEC) concentration at presentation, or prednisolone administration just prior to referral were excluded. Prednisolone dosage was started at 2 to 2.2 mg/kg/day, and thereafter gradually decreased every 2 to 4 weeks. Azathioprine dosage was started and maintained at 2 mg/kg/day during the study period. Pre‐ and post‐treatment sera at revisits were obtained for serial measurement of SPEC. Peak SPEC concentration during the study period was recorded. Results of peak SPEC concentrations were categorized as normal (< 200 μg/L), questionable (200‐400 μg/L), and abnormal (> 400 μg/L). Serial measurements of SPEC concentrations were stopped when SPEC concentration exceeds the upper measureable limit (1000 μg/L), or when SPEC concentration decreased from the previous revisit, or when prednisolone dosage administered was less than 0.5 mg/kg, q 24 h. Dosage of prednisolone at the time of peak SPEC concentration in each dog was also recorded. Comparison of peak SPEC concentrations between prednisolone alone (Pred) group and prednisolone with azathioprine (Pred+Aza) group was performed using Mann‐Whitney U‐test. Comparison of dosages of prednisolone between Pred and Pred+Aza group was performed using Student's t‐test. Seven dogs receiving prednisolone alone (Pred group) were diagnosed with immune‐mediated polyarthritis (IMPA, n = 2), inflammatory bowel disease (IBD, n = 2), meningoencephalitis of unknown etiology (MUO, n = 2), and inflammatory colorectal polyps (ICRP, n = 1). In combination with prednisolone, 9 dogs also received azathioprine (Pred+Aza group), and consisted of 7 dogs with MUO and 2 dogs with IMPA. Median (range) study period was 62 (14 to 105) days. In the Pred group, peak SPEC concentrations were classified as normal in 1 dog (MUO), questionable in 2 dogs (2 IBD), and abnormal in 4 dogs (2 IMPA, 1 MUO, and 1 ICRP). In the Pred+Aza group, peak SPEC concentrations were classified as normal in 6 dogs (5 MUO and 1 IMPA), questionable in 2 dogs (2 MUO), and abnormal in 1 dog (IMPA). Median (range) time to reach peak SPEC concentration in each dog was 14 (7‐49) and 35 (7‐76) days in the Pred and Pred+Aza group, respectively. Peak SPEC concentrations were significantly (p < 0.05) higher in the Pred group (median 406 μg/L, range 85‐1000 μg/L) when compared to Pred+Aza group (median 177 μg/L, range 49‐625 μg/L). Prednisolone dosages at the time of peak SPEC concentration was slightly higher in the Pred group (mean ± sd: 1.6 ± 0.46 mg/kg) when compared to Pred+Aza group (mean±SD: 1.4 ± 0.39 mg/kg), although this increase was not significant. This study demonstrated that oral prednisolone administration to dogs with naturally occurring immune‐mediated disease could cause various degrees of increase in SPEC concentration. Therefore, abnormal SPEC concentrations in dogs receiving prolonged oral prednisolone should be interpreted cautiously. Concurrent administration of azathioprine did not increase the SPEC concentrations more than the increase observed in administration of prednisolone alone. GI03 THE EFFECT OF THE PROBIOTIC SIVOY™ ON FECAL MICROBIAL COMPOSITION IN HEALTHY DOGS Blake Guard 1, Giacomo Rossi2, Joerg Steiner1, Jan Suchodolski1 1Texas A&M University, College Station, TX, USA, 2University of Camerino, Camerino, Italy Probiotic administration is often recommended for patients with gastrointestinal (GI) disease with the goal to balance microbial populations in the intestinal tract. Despite this aim, the actual impact of probiotic formulations on intestinal microbiota has not been elucidated. Various studies that have evaluated probiotic formulations containing doses ranging from 5x108 to 10x1010 colony forming units (cfu) have demonstrated only minor and transient changes in a few bacterial groups during the supplementation period. SIVOY™ is a high dose probiotic cocktail containing a mix of 9 strains of Lactobacillus spp., Bifidobaterium spp., and Streptococcus spp. for a total of 200 billion cfu (2x1011) and has shown anti‐inflammatory properties in vivo in previous studies in animal models. However, its impact on the fecal microbiome is currently unknown. Therefore, the aim of this study was to characterize the fecal microbiome in healthy dogs after having been administered the probiotic SIVOY. Healthy dogs were randomized to receive either the probiotic SIVOY (n = 10) or a placebo (n = 10). Fecal samples were collected at baseline (T0) and after 60 days (T60) of supplementation. DNA was extracted using the ZR Fecal DNA Kit™ (Zymo Research Corporation, Irvine, CA). The fecal microbiota was analyzed by Illumina sequencing of the 16S rRNA gene. Quantitative PCR was performed for Lactobacillus spp., Bifidobacterium spp., E. coli, C. perfringens, and Streptococcus spp. Sequence percentages and qPCR data between time points were evaluated using a Wilcoxon rank‐sum test. To determine differences in microbial communities between groups, the analysis of similarities (ANOSIM) function in QIIME (pipeline version 1.8) was used based on the unweighted UniFrac distance matrixes. Statistical significance was set at p < 0.05. The overall microbial communities were not significantly altered between T0 and T60 in either the placebo or the probiotic group. Bacterial species richness was also not significantly different between T0 and T60 in either the placebo or the probiotic group. Quantitative PCR revealed that Streptococcus spp. and Bifidobacterium spp. were significantly increased at T60 vs. T0 in the probiotic group (median log DNA [p‐value]: 7.33 vs. 4.61 [0.032] and 6.80 vs. 3.20 [0.039], respectively). There were no differences in bacterial abundances based on qPCR data between the two time points in the placebo group. In conclusion, this study revealed an increased abundance of Streptococcus spp. and Bifidobacterium spp. in the feces of healthy dogs that had received the probiotic SIVOY. Further studies may be warranted to evaluate the effect of this probiotic on the microbial composition within the upper GI tract. GI04 QUANTITATIVE AND QUALITATIVE IDENTIFICATION OF FELINE TRITRICHOMONAS FOETUS SURFACE ANTIGENS: PUTATIVE TARGETS FOR DIAGNOSIS AND TREATMENT Emily Gould, Mabre Brand, M. Katherine Tolbert University of Tennessee College of Veterinary Medicine, Knoxville, TN, USA Tritrichomonas foetus (TF) is a flagellated protozoal pathogen responsible for chronic diarrhea and reproductive failure in domestic felids and cattle, respectively. There is no safe and efficacious preventative or treatment for feline trichomonosis. Previous studies have identified two surface antigens (1.15 and 1.17) that are expressed by bovine TF and promote its survival and pathogenicity towards the host epithelium. Bovine and feline TF are highly similar, therefore these surface epitopes may be conserved between TF genotypes and may provide a similar pathogenic benefit to feline TF. The study aims were to determine if feline TF express 1.15 and 1.17 and if neutralization of these antigens with monoclonal antibodies (MAbs) ameliorates intestinal epithelial cell destruction. An indirect enzyme‐linked immunosorbent assay (ELISA) was developed to semi‐quantitatively evaluate the ability of two monoclonal antibodies (MAbs) that target bovine TF surface antigens (MAbs 1.15, 1.17) to recognize feline TF. ELISA positivity was defined as two standard deviations above the mean of a negative control. Qualitative assessment was performed using dot blot analysis. Co‐culture of porcine intestinal epithelial cells (IPEC‐J2) with MAb 1.15, 1.17‐treated TF was performed to evaluate the effect of these surface antigens on intestinal epithelial cytotoxicity. A bovine TF and 1 Pentatrichomonas hominis (PH) isolate were used as positive and negative controls, respectively. Assays were performed in a minimum of triplicate. Data were analyzed using Systat (p < 0.05). Feline TF were identified to express surface protein 1.15 by indirect ELISA and dot blot analysis. Results were confirmed with bovine TF in all immunodetection assays. Treatment of feline TF isolates with Mab 1.15 resulted in amelioration of TF‐induced intestinal epithelial destruction compared to IPEC‐J2 cells infected with vehicle‐treated feline TF. These studies demonstrate the presence of the surface antigen 1.15 in feline TF isolates. If antigen 1.15 participates in the pathogenicity of feline TF and is conserved across multiple isolates, it may represent a novel target for development of a vaccine and/or rapid, bedside diagnostic assay. GI05 CYTOPATHOLOGICAL FINDINGS OF FECAL OR RECTAL CYTOLOGY SAMPLES FROM DOGS IDENTIFIED WITH OR WITHOUT THE PRESENCE OF CYNICLOMYCES GUTTULATUS Jenessa A. Winston 1, Brian R. Gaines2, Adam J. Birkenheuer1, Ida Piperisova3, Scott Moroff3, Jody L. Gookin1 1North Carolina State University College of Veterinary Medicine, Raleigh, NC, USA, 2North Carolina State University Department of Statistics, Raleigh, NC, USA, 3ANTECH Diagnostics, New York City, NY, USA Cyniclomyces guttulatus (Cg) is a gastrointestinal yeast of rabbits. Cg can also be observed in the feces of dogs. It is generally presumed that Cg in dogs is a nonpathogenic “pass through” organism reflecting the dog's consumption of rabbit feces. However, the clinical significance of Cg infection in dogs remains unclear. The purpose of this study was to determine the cytopathological characteristics and clinical pathologists’ perceptions of pathogenicity of Cg when observed in fecal samples submitted to a U.S. commercial veterinary diagnostic facility. All cytology reports generated by Antech Diagnostics from 2001 to 2011 were searched using the keyword Cyniclomyces. Reports pertaining to dogs in which Cg was identified by fecal or rectal cytology were reviewed for patient sex, age, breed, season, and region. Descriptions referable to the presence of bacterial dysbiosis, RBCs, WBCs, inflammation, mucus, and other organisms, as well as the abundance (low, moderate, marked) and perceived pathogenesis (not pathogenic, pathogenic, uncertain pathogenicity) of Cg present were also recorded. For comparative purposes, similar data was extracted from fecal and rectal cytology reports from a subset of dogs in which Cg was not identified and a subset of dogs in which other enteric yeasts were identified. Fisher's exact test was used to examine the relationship between the abundance of Cg and other reported cytological findings, as well as to calculate odds ratios for comparing dogs with Cg to the two different control groups. Logistic regression models were used to estimate probabilities and odds ratios for cytological findings given the existence of Cg, after adjusting for other relevant variables. Cyniclomyces was identified in 265 canine cytology reports generated by Antech over the study interval. Sixty four percent (n = 170) of these samples were fecal/rectal cytologies. Among dogs with Cg identified by fecal or rectal cytology, there was a significant positive association between the abundance of Cg and presence of WBCs (P = 0.037). A total of 188 dog fecal/rectal cytologies without Cg reported were identified for use as controls (no yeast present, n = 116; other enteric yeast present, n = 72). The control dogs were 3 times more likely to have concurrent bacterial dysbiosis (95% CI, 2.066‐5.236, p < 0.0001) than were dogs with Cg. Dogs with Cg were 2.4 times (95% CI, 1.140‐5.088, P = 0.021) more likely to have WBCs and 3.6 times (95% CI, 1.471‐8.998, P = 0.003) more likely to have inflammation present in the fecal or rectal cytology compared to control dogs with other enteric yeast identified. No significant association was observed between the presence of Cg in fecal/rectal cytology and patient sex, age, breed, season, or region. In 94% (n = 160) of the canine fecal/rectal cytology reports in which Cg was identified, clinical pathologists opined on the pathogenicity of Cg. Expressed opinions were that Cg was nonpathogenic (81%) or was of uncertain pathogenicity (13%). Of 23 clinical pathologists that completed multiple reports, equivocation among their reports was common (43%) but did not correlate with abundance of Cg present. These results suggest that Cg is significantly associated with concurrent fecal cytological evidence of gastrointestinal inflammation in dogs. The pathogenicity of Cg however, remains an enigma. Based on these results, further investigation of Cg pathogenic effects in dogs may be justified. Such studies would aid in determining the significance of Cg and whether treatment of the infection is warranted in dogs with gastrointestinal disease. GI06 NATURAL HISTORY OF BILIARY SLUDGE IN DOGS Stefanie DeMonaco, David Grant, Martha Larson, Thomas Cecere, David Panciera, Michael Leib Virginia Maryland College of Veterinary Medicine, Blacksburg, VA, USA Biliary sludge is associated with gallbladder (GB) dysmotility and mucus hypersecretion suggesting that these factors may lead to GB mucoceles. If biliary sludge progresses to GB mucoceles then treatments to reduce the progression of sludge are warranted. The aim of this study was to determine the natural history of biliary sludge in dogs using serial ultrasonographic and biochemical examinations. Of seventy‐four healthy dogs that underwent ultrasonography forty‐two (57%) had biliary sludge and were subsequently reevaluated at three‐month intervals for one year. The following were determined: percentage of the GB that was filled with sludge (mild (0.01%‐24.4%), moderate (24.5%‐49.4%), moderate to severe (49.5%‐74.4%), severe (74.5%‐100%)), gravity dependency of sludge, GB dimensions, and biochemical indices (ALT, GGT, ALP, total bilirubin, triglycerides and cholesterol). Mixed model ANOVA, Friedman's Chi‐Square, Mantel‐Haenzsel chi‐square tests, and Kruskal‐Wallis test were performed to detect significant changes in these parameters. After 1 year of follow‐up, the percentage of the GB filled by sludge was mild (32%), moderate (47%), moderate to severe (5%), and severe (3%) with no significant differences over time. The prevalence of non‐gravity dependent sludge significantly increased (P = 0.004). Dogs had resolved (2%), decreased (19%), persistent (40%), increased (29%), or recurrent (10%) sludge at the conclusion of the study. Biochemical indices or GB volume were not significantly different over time or among groups. Biliary sludge is prevalent, remains asymptomatic, and rarely resolves in healthy dogs over a period of 1 year. Gallbladder content became less gravity dependent, which may indicate changes in consistency over time. GI07 IDENTIFICATION OF HISTAMINE RECEPTORS IN THE CANINE GASTROINTESTINAL TRACT Alyssa Sullivant, Jim Cooley, Todd Pharr, Wills Robert, Mackin Andrew, Archer Todd Mississippi State University, Mississippi State, MS, USA The role of mast cells and their major secretory product, histamine, in chronic gastrointestinal diseases such as Crohn's disease and ulcerative colitis has been intensely investigated over the last two decades in human medicine. Histamine is released following mast cell activation and exerts its action through binding to four different histamine receptors (H1, H2, H3, and H4). Histamine receptors are dispersed throughout the body, with each different receptor mediating unique responses in the body. Documentation of the presence and type of histamine receptors in the canine gastrointestinal tract will provide additional research opportunities to further explore the role of histamine and its receptors in chronic canine enteropathies as well as the therapeutic potential of histamine receptor antagonists. Full thickness gastric, duodenal, jejunal, ileal, and colonic biopsies were obtained from 6 clinically normal, healthy adult dogs immediately after humane euthanasia. Immunohistochemistry was performed using commercially available rabbit and/or goat histamine‐receptor antibodies predicted to react with canine tissues on formalin‐fixed, paraffin embedded tissue sections using a refined avidin‐biotin detection technique. Tissues were evaluated with light microscopy to identify the presence and histologic distribution of the H1, H2, H3, and H4 receptors in the gastrointestinal tract. Formalin‐fixed tissue of a grade 3 canine mast cell tumor was used as a positive control. The staining intensity of the immunoreactivity of the histamine receptors in the gastrointestinal tract was qualitatively graded as: 0, negative; 1 + , weak; 2 + , moderate, and 3 + , strong. To determine if there were differences in histamine receptor scores among the superficial and deep mucosa, submucosa, lymphoid tissue, smooth muscle and ganglia of each section in the gastrointestinal tract, a method similar to the non‐parametric Friedman's test was conducted. Specificity of each antibody was evaluated using a Western Blot technique on tissue lysates of the canine gastrointestinal tract. The presence and distribution of H1, H2, H3, and H4 receptors varied among tissues and cell types within the canine gastrointestinal tract. H1 receptors were located in the epithelial cells of the superficial mucosa and glands of the deep mucosa of the stomach, lymphoid tissue in the ileum, and the smooth muscle and ganglia of all tissues. H2 receptor staining intensity was significantly greater in the gastric glands of the deep mucosa than the epithelial cells of the superficial mucosa. H2 receptors were located in the superficial mucosal epithelial cells and deep mucosal cells of intestinal glands, lymphoid tissue, and ganglia of the duodenum, jejunum, ileum and colon. In all tissues, H3 receptors were located in the smooth muscle and ganglia. H3 receptors were also located in ileal and colonic lymphoid tissue. H3 receptor staining intensity was significantly greater in the epithelial cells of the colonic and gastric superficial mucosa compared to the deep mucosal cells of these tissues, and H3 receptors were not identified in the duodenal, ileal, or jejunal mucosa, except one dog that had weak staining in the jejunal deep mucosa. H4 receptors were located in the ganglia, lymphoid tissue, and occasionally in the gastric glands of the deep mucosa. The density and distribution of H4 receptors in the gastrointestinal tract were decreased in comparison to the other histamine receptors. Western blot demonstrated both specific and non‐specific staining of the H1 and H3 receptor antibody but good specificity of the H4 receptor antibody. The H2 receptor antibody was not compatible with Western Blot techniques, despite excellent immunohistochemical specificity and consistency. All 4 histamine receptors were identified in the canine gastrointestinal tract, with the H1, H2, and H3 receptors having a greater distribution than the H4 receptor. Validation of the immunohistochemical technique with the H4 receptor antibody was successful. Additional validation methods for the H2 receptor antibody are warranted. Despite non‐specific staining with the H1 and H3 receptor antibodies on Western blot, the distribution of H1 and H3 receptors in the canine gastrointestinal tract was similar to that seen in the human gastrointestinal tract. Further studies to compare the density and distribution of these receptors in dogs with gastrointestinal disease are needed. GI08 ALTERATIONS IN FECAL METABOLITE PROFILES OF DOGS WITH CHRONIC ENTEROPATHY J.B. Honneffer, B.C. Guard, J.M. Steiner, J.S. Suchodolski Gastrointestinal Laboratory, Texas A&M University, College Station, TX, USA Intestinal dysbiosis in dogs is defined as a relative shift in the bacterial populations of the gastrointestinal tract (GIT) and has been associated with chronic enteropathy (CE). Our group has previously reported small intestinal or fecal dysbiosis and decreased species richness in canine spontaneous inflammatory bowel disease. However, due to functional redundancy between bacterial groups, determining the biochemical changes in the GIT requires direct analysis of the biomolecules that are present. The study aim was to evaluate differences in fecal metabolite profiles between dogs with chronic gastrointestinal inflammation and healthy control dogs. Fecal samples were collected from dogs with histopathological evidence of gastrointestinal inflammation (n = 15) and from healthy control dogs (n = 15). The fecal metabolome, comprised of metabolic products of both the host and microbiota, was assessed using an untargeted approach combining LC‐MS and GC‐MS platforms. Random Forest Analysis was used to rank differentiating power of metabolites, and the Mann‐Whitney U‐test followed by the Benjamini‐Hochberg adjustment of p‐values for multiple comparisons was used to test for statistical significance (q < 0.05). A total of 797 named biochemical compounds were identified. Of these, 145 were significantly altered after adjusting for multiple comparisons (q < 0.05). Among major metabolic pathways, some metabolites associated with GIT redox homeostasis were altered, including significantly increased concentrations of precursors of glutathione: 5‐oxoproline (q = 0.004), cysteine (q = 0.024), glycine (q = 0.003), and γ‐glutamyl amino acid derivatives of lysine, phenylalanine, valine, and leucine (all q ≤ 0.015). Primary bile acids were generally increased in dogs with disease (cholate, q = 0.013; chenodeoxycholate, q = 0.055), while secondary bile acids were generally decreased in dogs with chronic enteropathy (deoxycholate, q = 0.045; lithocholate, q = 0.059), suggesting impaired bacterial conversion from primary to secondary bile acids. Metabolites within the aromatic amino acid biosynthesis pathway exhibited some of the most significant alterations. The tryptophan metabolites indoleacetate and indolepropionate were significantly decreased in dogs with disease (q = 0.008 and 0.030, respectively). 2‐oxindole‐3‐acetate, an oxidative degradation product of indoleacetate, was 100‐fold decreased in diseased animals (q = 0.001) and was top‐ranked for discriminating power out of all detected metabolites. This study demonstrates that feces are a rich sample matrix for biomolecules, representing both the host and microbial metabolite profiles. Several metabolites were identified that may be investigated as future biomarkers and may help elucidate the etiopathogenesis of canine chronic enteropathy. GI09 TARGETED SERUM AMINO ACID ANALYSIS OF HYPOCOBALAMINEMIC DOGS WITH DECREASED OR INCREASED SERUM FOLATE CONCENTRATIONS Rosana Lopes, Niels Grützner, Jonathan Lidbury, Jan Suchodolski, Jörg Steiner Texas A&M University, College Station, TX, USA Cobalamin (B12) and folate (B9) are water‐soluble vitamins that are essential components of one‐carbon metabolism. Both vitamins influence each other in biochemical processes including the homocysteine remethylation cycle and the transsulfuration pathway. It is well known that cobalamin deficiency can alter folate metabolism by decreasing intestinal absorption of folate. Also, increased serum folate concentrations have been observed in dogs with small intestinal dysbiosis. A previous study comparing changes in serum amino acid concentrations between healthy dogs, hypocobalaminemic dogs, and dogs with biochemical evidence of cobalamin deficiency has shown significant differences in serum concentrations of glycine, glutamic acid, methionine, tyrosine, tryptophan, and lysine in dogs with hypocobalaminemia. The aim of this study was to investigate changes in serum amino acid concentrations between healthy normocobalaminemic dogs and hypocobalaminemic dogs with decreased or increased serum folate concentrations. Fresh serum samples from healthy dogs (n = 10) were collected and used to perform a serum chemistry profile and measure serum cobalamin and folate concentrations. All healthy dogs had serum cobalamin and folate concentrations within the respective reference intervals of 251‐908 ng/L and 7.7‐24.4 μg/L. Surplus serum samples from 31 hypocobalaminemic (serum cobalamin <150 ng/L) dogs were prospectively enrolled. Nineteen out of 31 hypocobalaminemic dogs also had decreased serum folate concentrations (serum folate <7.3 μg/L; low B9), whereas twelve had increased serum folate concentrations (serum folate >18.7 μg/L; high B9). A previously validated gas chromatography/mass spectrometry method was used to quantify serum glycine, glutamic acid, methionine, tyrosine, tryptophan, and lysine concentrations. A Kruskal‐Wallis test and a Dunn's post‐test were performed to compare the concentrations of the six amino acids amongst healthy dogs and dogs with hypocobalaminemia with decreased or increased serum folate concentrations. Results are summarized in Table 1. Table 1. Statistical analysis of serum amino acid concentrations (μmol/L) showing mean and standard deviation or median and range for healthy and diseased dogs Amino acids Healthy dogs low B12 low B9 low B12 high B9 P‐value Glycine 20.6 (14.3–32.5)a 6.7 (2.6–86.3) b 7.0 (2.8–107.3)a P < 0.05 Glutamic acid 3.2 (2.7–4.9) a 33.9 (8.2–213.8) b 11.7 (4.6–68.9) c P < 0.0001b; P < 0.001c Methionine 5.3 (4.0–8.4) a 6.9 (4.4–53.8) a 10.5 (4.4–35.6) b P < 0.05 Tyrosine 6.6 (5.7–10.2) a 4.5 (5.3–12.0) a 5.7 (5.9–23.3) a P = 0.065 Tryptophan 168.8 (89.3–317.2) a 24.1 (4.0–68.1) b 29.9 (13.0–136.2) c P < 0.0001 b; P < 0.001 c Lysine 24.8 (±10.2) a 62.9 (±28.1) b 61.4 (±30.5) c P < 0.0005 b; P < 0.05c Diseased groups with the same letter in the same row are not statistically significantly different. A p‐value <0.05 was considered significant. In conclusion, significant differences in serum amino acid concentrations were observed between healthy dogs and hypocobalaminemic dogs with either decreased or increased serum folate concentrations. These results suggest that altered absorption, metabolism, and/or degradation of amino acids are associated with chronic intestinal diseases. GI10 SERUM AMINO ACID ANALYSIS IN HYPOCOBALAMINEMIC DOGS WITH EXOCRINE PANCREATIC INSUFFICIENCY Rosana Lopes, Niels Grützner, Jonathan Lidbury, Jan Suchodolski, Jörg Steiner Texas A&M University, College Station, TX, USA In dogs, chronic intestinal disease and exocrine pancreatic insufficiency (EPI) are the most common causes of cobalamin deficiency and have been associated with decreased intestinal absorption of cobalamin and also with a reduced availability of several amino acids. Digestibility is directly affected by the amount of pancreatic enzymes available for food breakdown in the gastrointestinal tract. Altered amino acid concentrations of various amino acids (i.e., glycine, glutamic acid, tyrosine, methionine, lysine, and tryptophan) have been described in humans with gastrointestinal and exocrine pancreatic disorders. The aim of this study was to investigate changes in serum amino acid concentrations between healthy normocobalaminemic dogs and hypocobalaminemic dogs with or without EPI. Fresh serum samples from healthy dogs (n = 10) were collected after physical examination and were used to perform a serum chemistry profile and measure serum cobalamin and canine trypsin‐like immunoreactivity (cTLI) concentrations. These healthy dogs had serum cobalamin and cTLI concentrations within the respective reference intervals of 251‐908 ng/L and 5.7‐45.2 μg/L. Diseased groups were both hypocobalaminemic (serum cobalamin concentration < 150 ng/L; low B12) and were prospectively enrolled. A group of 15 hypocobalaminemic dogs without EPI and a group of 13 hypocobalaminemic dogs with EPI (cTLI concentration of ≤ 2.5 μg/L; EPI) were enrolled. A previously validated gas chromatography/mass spectrometry method was used to quantify serum glycine, glutamic acid, methionine, tyrosine, tryptophan, and lysine concentrations of the dogs enrolled. A Kruskal‐Wallis test and a Dunn's post‐test were performed to compare the concentrations of the six amino acids amongst healthy dogs, and hypocobalaminemic dogs with or without EPI. Results are summarized in Table 1. Table 1. Statistical analysis of serum amino acid concentrations (μmol/L) showing mean and standard deviation or median and range between the 3 groups of dogs Amino acids Healthy low B12 low B12 with EPI P ‐value Glycine 20.6 (14.3–32.5)a 12.4 (2.8–107.3) a 6.9 (3.2–69.3)b P < 0.005 Glutamic acid 3.2 (2.7–4.9) a 15.5 (7.4–38.7) b 15.2 (4.6–69.0) b P < 0.0001 Methionine 5.3 (4.0–8.4) a 19.5 (4.4–43.7) b 10.2 (4.6–35.6) a P < 0.05 Tyrosine 6.6 (5.7–10.2) a 5.9 (5.7–8.0) a 5.7 (5.9–23.3) a P = 0.095 Tryptophan 168.8 (89.3–317.2) a 32.4 (11.1–136.2) b 35.0 (6.4–107.3) c P < 0.001b;0.0001 c Lysine 24.8 (±10.2) a 79.9 (±32.9) b 39.6 (±19.2) a P < 0.0001 Groups with the same letter in the same row are not statistically significantly different. A p‐value <0.05 was considered significant. In conclusion, altered serum amino acid concentrations were observed among healthy dogs and hypocobalaminemic dogs with or without EPI. The exact cause of these changes requires further investigation. But the lack of digestive pancreatic enzymes in dogs with EPI could play a role in decreased absorption of amino acids leading to altered amino acid metabolism. GI11 EFFECTS OF AN ORAL SUPPLEMENT BASED ON CHONDROITIN SULFATE, RESISTANT STARCH, AND PREBIOTICS IN DOGS WITH IBD Sergi Segarra 1, Silvia Martínez‐Subiela2, Alberto Muñoz‐Prieto2, José Cerón2 1R&D Veterinary Division, Bioiberica SA, Barcelona, Spain, 2Department of Animal Medicine and Surgery, Veterinary School, University of Murcia, Murcia, Spain Current therapeutic options for canine inflammatory bowel disease (IBD) are focused on reducing intestinal inflammation and achieving clinical remission, but often result in side effects especially due to the use of immunosuppressant drugs. In IBD patients, activation of nuclear factor kappa‐light‐chain‐enhancer of activated B cells (NF‐κB) causes intestinal inflammation. Chondroitin sulphate (CS) inhibits NF‐κB phosphorylation and therefore could help to reduce the inflammatory response associated with IBD. Large‐bowel fermentation of certain resistant starches increases butyrate production, which helps maintaining the intestinal barrier integrity. Orally administered prebiotics promote the growth of beneficial gut microflora. The aim of this study was to evaluate, in a multicentric randomized double‐blind placebo‐controlled trial in dogs with IBD, the effects of an oral nutritional supplement containing butyrogenic resistant starch, CS, β‐glucans and mannanoligosaccharides on serum biomarkers, intestinal histology, and canine IBD activity index (CIBDAI). A total of twenty‐two biopsy‐confirmed canine IBD patients received either a placebo (group A; n = 12) or the oral supplement (group B; n = 10) orally once daily for six months. During the study, all dogs were given the same special hypoallergenic diet. Pre and post treatment serum biochemical markers, histopathology and CIBDAI scores were compared. Biopsies were scored based on the WSAVA guidelines for the assessment of endoscopic intestinal biopsy samples. A one‐way analysis of variance (ANOVA) of repeated measures and an uncorrected Fisher's Least Significant Difference (LSD) test were used to compare the values in each group at different times before and during treatment. Unpaired t test was performed to compare pretreatment values between groups. Analytes that did not follow a parametric distribution were log transformed before applying the test. Dogs in group B showed increased serum concentrations of cholesterol and paraoxonase‐1 (PON1) after one month of treatment compared with pretreatment values (median: 242.9 mg/L vs 163.3 mg/L; P = 0.0094 for cholesterol and median: 2.7 UI/ml vs 2.0 UI/ml; P < 0.0001 for PON1) and increased serum concentrations of triglycerides after three months of treatment (median 82.38 mg/L vs 42.91 mg/L; P = 0.034). These changes were not seen in the placebo group, and could indicate an improvement in intestinal absorption and a decrease in oxidative damage associated with IBD in the supplemented group. Group B showed a 1.6‐fold decrease (P = 0.0011) in mean histologic severity score at the end of the follow‐up. A limitation of the study was that before initiating the treatment this parameter was significantly higher in the supplemented group (P = 0.01), compared to the placebo group. Initially, there were no significant differences (P = 0.53) in mean CIBDAI between groups (median group A: 6.5; group B: 7). A significant reduction in mean CIBDAI was observed in both groups after six months, which might be due to the special diet. However, the supplemented group experienced a faster clinical improvement, as shown by a more pronounced decrease in mean CIBDAI especially after one (1.75‐fold; P = 0.003) and two months (3.5‐fold; P = 0.0002) of treatment, compared to the placebo group (1.62‐fold and 2.6‐fold; P = 0.04 and P = 0.01, respectively). No side effects related to the treatments were reported in any of the groups. In conclusion, although larger prospective studies are warranted, our results point towards beneficial effects of this oral supplement containing butyrogenic resistant starch, CS, and prebiotics in dogs with IBD, by improving clinical and histological indexes as well as serum biochemical markers of intestinal absorption and oxidative damage with no associated side effects. GI12 GASTROINTESTINAL ENDOSCOPIC MUCOSAL LESION SCORES IN HEALTHY DOGS RECEIVING PREDNISONE, ASPIRIN, AND OMEPRAZOLE Jacqueline Whittemore 1, Emily Gould1, Shelley Newman1 1University of Tennessee, College of Veterinary Medicine, Knoxville, TN, USA Information is lacking on the impact of low‐dose aspirin, with or without omeprazole, on development of gastrointestinal disease in dogs receiving immunosuppressive doses of prednisone. The objective of this study was to compare endoscopic mucosal lesions in dogs receiving these therapies. Nine research dogs were randomized to receive 2 mg/kg/d of prednisone with placebo (Group 1), 1 mg/kg/d of aspirin (Group 2), or 1 mg/kg/d of aspirin and 1 mg/kg/d omeprazole (Group 3) for 28 days. Medications were administered in 1Tbsp food followed by 3 mL water. Endoscopic evaluation and biopsy collection were performed on days 0, 14, and 28. Mucosal lesion scoring was performed by blinded review of video and photos after completion of data collection. Mucosal lesion scores and treatment day were compared by Friedman's test and Kruskal‐Wallis. P < .05 was considered significant. Ulceration of the lower esophageal sphincter developed in 4 dogs (1, 2, and 1 per group, respectively). Gastric ulcers developed in 4 dogs. Median total endoscopic mucosal lesion scores were significantly higher during treatment for all groups but did not differ among groups (Figure; letters denote significant differences). Administration of immunosuppressive doses of prednisone is associated with significant increases in endoscopic mucosal lesion scores in healthy dogs. Lack of difference between groups could reflect Type 2 error due to the small number of dogs and wide inter‐individual variation. Esophageal and gastric ulcers can develop during glucocorticoid therapy, with or without aspirin and omeprazole. Further evaluation in more dogs, with histology and COX1/2 immunohistochemistry, is warranted and ongoing. GI13 CHARACTERIZATION OF CANINE DYSPHAGIA IN A VETERINARY UNIVERSITY HOSPITAL Cecily Bonadio1, Stanley Marks1, Philip Kass1, Rachel Pollard1, Diane Cheney1 1UC davis, davis, CA, USA Dysphagia is a relatively common disorder in dogs and can result from an abnormality in the oral, pharyngeal, pharyngoesophageal, or esophageal phase of swallowing. The most common causes of dysphagia in dogs in a veterinary university hospital have not been determined, and the utility of videofluoroscopic swallow studies in this cohort of animals has not been comprehensively evaluated. Our goals were to retrospectively evaluate the most common causes of dysphagia in dogs and to determine associations between the animal's dysphagia and the signalment, clinical signs, diagnostic tests or diagnostic procedures. A comprehensive search of medical records at the University of California, Davis VMTH for all canine patients with dysphagia was performed over a 10‐year period (January 2003 to January 2013). Dogs with dysphagia directly attributable to oral causes (dental disease, stomatitis, cleft palate) were excluded from the investigated cohort. All dogs included in the survey had thoracic radiographs as an inclusion criterion. 742 dogs were included in the study of which 474 (63.9%) underwent videofluoroscopy. Ages of dogs ranged from 0.5 to 217 months (median 56 ± 52 months). The most commonly affected purebred dogs were Labrador Retrievers, Golden Retrievers and German Shepherds. The most common causes of dysphagia included megaesophagus (n = 274, 38.3%); gastroesophageal reflux (GER) (n = 66, 8.9%); esophageal foreign body (n = 63, 8.5%); sliding hiatal hernia (n = 54, 7.3%); and cricopharyngeal dysphagia (n = 34, 4.6%). 131 dogs had chronic regurgitation of unknown cause (17.7%). Videofluoroscopy was most helpful for diagnosing GER, hiatal hernia, pharyngeal weakness, esophageal stricture, cricopharyngeal achalasia, and esophageal hypomotility. A specific cause of dysphagia was not identified in 17.7% of dogs and further studies are warranted to better elucidate the causes of dysphagia in these animals. The findings of this study should help prioritize the workup of dysphagic dogs and highlights the clinical utility of videofluoroscopic swallow studies. GI14 ABILITY OF ULTRASOUND TO PREDICT THE PRESENCE AND LOCATION OF HISTOLOGIC LESIONS IN THE SMALL INTESTINE OF CATS Talia Guttin 1, Audra Walsh1, Jennifer Reetz1, Amy Durham1, Dorothy Brown1, Mark Rondeau1s 1University of Pennsylvania School of Veterinary Medicine, Philadelphia, PA, USA Definitive diagnosis of feline infiltrative small intestinal (SI) disease is challenging, and debate exists as to which biopsy technique is optimal. Ultrasound may be useful in selecting biopsy type. The purpose of this study was to assess the ability of ultrasound examination to predict the presence and location of histologic lesions by segment and tissue layer of the SI. The records of cats who had an abdominal ultrasound and full thickness SI biopsies within 1 month were reviewed retrospectively. Ultrasonographic images were reviewed by a single radiologist and assessed for lesions such as abnormal wall thickness, layering, echogenicity, or mural masses. Pathologists graded all histopathologic samples according to WSAVA standards. For both ultrasound and biopsy, each wall layer of SI was evaluated for lesions in each section available for review. The ability of ultrasonographic lesions within a specific wall layer of SI to predict the presence of histologic change in that wall layer was assessed. One hundred sixty nine cats met the inclusion criteria. Ultrasonographic SI lesions were present in 132 cats in the duodenum (50), jejunum (67) and/or ileum (42). Histopathologic samples were obtained via laparotomy (60) or via necropsy (109). Diagnoses in cats with SI pathology included: inflammatory bowel disease (105), small cell lymphoma (16), intermediate‐large cell lymphoma (14), other neoplasia (13), and suspected feline infectious peritonitis (10). Twenty five cats had no histologic lesions. Overall ultrasonographic lesions in the SI had high positive predictive value (PPV) for histologic SI lesions (duodenum 82%, jejunum 91%, ileum 88.1%) but poor negative predictive value (NPV) (duodenum 27.1%, jejunum 27.3%, ileum 40.4%). Table 1 shows the ability of ultrasound to predict histologic lesions in specific SI wall layers. The 95% confidence intervals (95% CI) are also reported. Table 1 Sensitivity (95% CI) Specificity (95% CI) PPV (95% CI) NPV (95% CI) Duodenum, mucosal lesion 19.3% (11.4–29.4) 80.6% (62.5–92.5) 72.7% (49.8–89.3) 27.2% (18.4–37.4) Jejunum, mucosal lesion 40.8% (29.6–52.7) 90.9% (58.7–99.8) 96.9% (83.8–99.9) 18.2% (9.08–30.9) Ileum, mucosal lesion 21.3% (11.9–33.7) 100% (86.3–100) 100% (75.3–100) 34.2% (23.5–46.3) Duodenum, muscularis lesion 53.8% (25.1–80.8) 73.5% (63.9–81.8) 20.6% (8.7–37.9) 92.6% (84.6–97.2) Jejunum, muscularis lesion 86.2% (68.3–96.1) 44.1% (31.2–57.6) 43.1% (30.2–56.8) 86.7% (69.3–96.2) Ileum, muscularis lesion 53.8% (25.1–80.8) 59.5% (47.4–70.7) 18.9% (7.96–35.2) 88.0% (75.7–95.5) These results show that most cats with SI mucosal ultrasonographic lesions will have mucosal histologic changes. Mucosal biopsy may effectively identify these lesions. However, SI muscularis ultrasonographic lesions are not predictive of histologic disease in the muscularis layer, suggesting that full thickness biopsy may not be essential to obtaining a diagnosis in cats with ultrasonographic muscularis lesions. We conclude that ultrasound may help guide decisions about which biopsy type to perform in an individual cat. GI15 PROGNOSTIC FACTORS FOR SHORT‐ AND LONG‐TERM SURVIVAL IN DOGS WITH PROTEIN‐LOSING ENTEROPATHY Paola Gianella 1, Ugo Lotti2, Claudio Bellino1, Francesca Bresciani3, Aurelio Cagnasso1, Federico Fracassi3, Antonio D'Angelo1, Marco Pietra3 1University of Turin, Department of Veterinary Sciences, Grugliasco, Turin, Italy, 2Veterinary Clinic Valdinievole, Monsummano Terme, Pistoia, Italy, 3Department of Veterinary Medical Sciences, University of Bologna, Ozzano dell'Emilia, Bologna, Italy Little information is available about the outcome in dogs with protein‐losing enteropathy (PLE). The aims of this study were to evaluate the clinical and clinicopathological findings in dogs with PLE secondary to chronic enteropathy (CE) at time of diagnosis and 1 month after treatment that correlate with survival ≤ than or > 6 months (short‐ and long‐term survival), in order to identify risk factors for outcome. Records of dogs with chronic enteropathy (CE) and concurrent PLE that were diagnosed at three referral clinics, were retrospectively analyzed. The inclusion criteria were a history of chronic gastrointestinal signs (> 3 weeks), hypoalbuminemia (<2 g/dL) of gastrointestinal origin, histopathological evidence of gastrointestinal inflammation on biopsies, and available follow up. The exclusion criteria were the presence of underlying or concurrent disorders other than PLE secondary to CE, or an immunosuppressive therapy used within 2 weeks before intestinal biopsy. Dogs were classified as either short‐term (≤6 months) survivors (ST) or long‐term (>6 months) survivors (LT), and the clinical and clinicopathological variables were investigated between groups. The ability to predict the outcome was determined by multiple regression and ROC analyses. There must be no differences in treatment between groups. Sixty‐eight cases were enrolled; 23 dogs were classified as ST, 45 as LT. There was no difference in body weight, age, sex distribution, lifestyle, complaints/clinical signs, canine inflammatory bowel disease activity index (CIBDAI) and canine chronic enteropathy clinical activity index (CCECAI), or clinicopathological data between groups at the time of diagnosis. At one‐month follow up CIBDAI/CCECAI scores were higher, and serum total protein, globulin and total cholesterol were lower (P < 0.05) in STs versus LTs. Risk factors for poor outcome (≤6 months) were CIBDAI > 5 and low cholesterol determined 1 month after diagnosis. According to the receiver operating characteristic (ROC) curve analysis, CIBDAI > 5 evaluated 1 month after diagnosis was shown to be the best predictor of poor outcome. In conclusion, clinical and clinicopathological findings at time of diagnosis did not predict outcome in dogs with PLE. CIBDAI and cholesterol at 1 month follow up more predicted ST vs LT. GI16 EVALUATING QUALITY AND ADEQUACY OF GASTROINTESTINAL SAMPLES WITH REUSABLE AND DISPOSABLE FORCEPS Jennifer Sinclair1, Tracy Hill1, Sionagh Smith1, Darren Shaw1 1Royal Dick School of Veterinary Studies, The University of Edinburgh, Hospital for Small Animals, Midlothian, UK Gastroduodenoscopy is commonly performed in dogs with chronic enteropathies. Sample quality of endoscopic biopsies is of paramount importance for accurate histological diagnosis. Many veterinary specialists use reusable forceps due to decreased cost., It is unknown whether sample quality decreases with repeated use and at some point becomes inferior to disposable forceps. The hypothesis of this study was that biopsy quality with reusable forceps would deteriorate after repeated uses and become inferior to biopsy quality obtained with disposable forceps. Twenty‐four dogs undergoing gastrointestinal endoscopy for diagnostic investigations were included in a prospective, pathologist‐blinded study comparing disposable and reusable alligator standard cup biopsy forceps (Olympus 2,0 mm 1550 mm). A new disposable forceps was used for each patient; five reusable forceps were randomised for use. Samples were collected from the stomach, duodenum, ileum, and colon. A one‐sample Wilcoxon test was used to compare scores between forceps types. Mann‐Whitney tests were used to compare degradation of biopsy score over time. There was no significant difference in the adequacy, depth, or crush artifact between reusable and disposable forceps in the 4 intestinal areas for the first use of each forceps. With 5 uses of each reusable forceps, there was no apparent degradation in sample quality over that time. At least 5 dogs can be sampled with reusable forceps without a significant effect on tissue quality and adequacy compared to disposable biopsy forceps. This study suggests that because reusable biopsy forceps provide a financial advantage over disposable forceps, reusable forceps are adequate for at least 5 dogs. GI17 DOUBLE‐BLIND, PLACEBO‐CONTROLLED ULTRASONOGRAPHIC EVALUATION OF THE EFFECTS OF ANTIEMETIC DRUGS ON ANTRAL MOTILITY AND GASTRIC EMPTYING IN HEALTHY DOGS Kelli Bogard 1, Frederic Gaschen1, Erin Olson1, Marylynn Waddle1, Rhett Stout2, Lorrie Gaschen1 1Veterinary Clinical Sciences, School of Veterinary Medicine, Louisiana State University, Baton Rouge, LA, USA, 2Pathobiological Sciences, School of Veterinary Medicine, Louisiana State University, Baton Rouge, LA, USA The prokinetic effects of the most commonly used antiemetics were evaluated ultrasonographically in 14 healthy dogs after a 24 hour fast using a randomized, controlled, blinded crossover design. Starting 24 h before motility evaluation, metoclopramide (0.5 mg/kg q8 h), dolasetron (1 mg/kg q24 h), maropitant (1 mg/kg q24 h) or saline were administered subcutaneously. Cisapride (0.75 mg/kg q8 h) was given orally. Dogs were fed 1 h after they received their morning treatment. Contracted and relaxed cross‐sectional antral surface were measured in triplicate, and antral contractions (CF) were counted over 2 minutes before and at regular intervals for 7 h postprandially (pp). Motility index (MI) was calculated. Results were evaluated using a mixed model repeated measures ANOVA with dog as a random effect, treatment and time as fixed effects, and significance set at P < 0.05. After placebo MI increased over the first 30 min. pp, then reached a plateau, and started to decrease 5 h pp. Compared to placebo, metoclopramide significantly increased MI preprandially and 0, 30 min, 1, 4, 5, 6 and 7 h pp. Dolasetron also significantly increased MI preprandially, 0, 1 and 5 h pp. Maropitant significantly increased MI only preprandially. After cisapride MI was significantly higher than placebo only preprandially, 0 and 7 h pp. Of all tested drugs, metoclopramide caused the most pronounced and sustained increase in MI both before and after a meal. Fasted MI was increased for all other dugs, however postprandial MI was mostly unchanged. Metoclopramide is the antiemetic drug with the highest prokinetic activity in the gastric antrum. GI18 EVALUATION OF THE EFFECTS OF A 2‐WEEK TREATMENT WITH METRONIDAZOLE ON THE FECAL MICROBIOME OF HEALTHY DOGS Erin Olson 1, Julia Honneffer2, Marylynn Waddle1, Joerg Steiner2, Jan Suchodolski2, Frederic Gaschen1 1Veterinary Clinical Sciences, School of Veterinary Medicine, Louisiana State University, Baton Rouge, LA, USA, 2GI Lab, College of Veterinary Medicine, Texas A&M University, College Station, TX, USA The abnormal interaction between the GI microbiome and the immune system is thought to be a major contributing factor to the etiology of inflammatory bowel disease (IBD). Treatment of canine IBD often includes judicious use of antimicrobials such as metronidazole. The aim of this study was to prospectively evaluate the impact of metronidazole administration on the fecal microbiome of healthy dogs. Twenty‐four privately owned healthy dogs were enrolled and assigned to one of 3 groups of 8 dogs each: control dogs with no intervention (group 1, G1), dogs treated with a soy‐based hydrolyzed diet for 12 weeks with administration of metronidazole 15 mg/kg PO q12 h during weeks 8 to 10 (G2), and dogs maintained on their usual diet and administered the same dose of metronidazole for 2 weeks (G3). Feces and blood were collected at day (D) 0, 21 and 42 in G1; and D0, 14, 28 and 42 days in G2 and G3, with D0 coinciding with initiation, and D14 with discontinuation of metronidazole treatment. Feces were frozen at ‐80C within 4 hours of collection and analyzed in batch later. A qPCR panel targeting 8 bacterial groups of interest to gastrointestinal health was used for analysis of fecal microbiota at various phylogenetic levels (Faecalibacterium spp., Bifidobacterium spp., Lactobacillus spp., Streptococcus spp., Ruminococcaceae, C. perfringens, E. coli, and Blautia spp.). Serum cobalamin and folate concentrations were analysed by automated chemiluminescence assays. Differences in serum cobalamin and folate and in bacterial abundance among groups and time‐points were evaluated using repeated measures ANOVA followed by Tukey's post‐hoc test. Statistical significance was set at P < 0.05. Nine of 16 healthy dogs (56%) receiving metronidazole produced yellow, soft to liquid feces on one or more occasion(s) while they were medicated. This did not happen in the untreated control group. No significant changes were observed in serum concentration of cobalamin and folate in all 3 groups. In the control dogs, only E. coli showed a significant change in abundance over time between D0 and D21 (P < 0.05). In contrast, in G2 and G3 several of the bacterial groups showed significant changes in abundance over time. Because the responses in these 2 groups were similar, they were combined for statistical analysis. Significant increases in abundance compared to baseline were observed after metronidazole administration for E. coli (P < 0.001 at D14 and D28), Streptococcus spp. (P < 0.001 at D14), and Bifidobacteria spp. (P < 0.001 at day 14 and P < 0.05 at day 42). In contrast, significant decreases in abundance were observed for Faecalibacterium spp. (P < 0.001 at D14 and P < 0.05 at D28), Ruminococcaceae (P < 0.001 at day 14 and P < 0.05 at day 42), and Blautia spp. (P < 0.01 at D14). No significant changes in abundance due to metronidazole were observed for C. perfringens and Lactobacillus spp. The administration of metronidazole led to significant changes in the fecal microbiome as evaluated by qPCR. Interestingly, metronidazole led to a decrease in bacterial groups that constitute major abundant groups in the GI tract of healthy dogs and are thought to be important for maintaining intestinal health (Faecalibacterium, Blautia, and Ruminococcaceae). While most of the changes were reversed 14 days after the end of antibiotic administration, some of the evaluated bacterial groups remained significantly altered up to 4 weeks after administration. In conclusion, metronidazole has a profound effect on the fecal microbiome and leads to a more variable microbiome over time. Further studies are needed to evaluate the functional changes in the microbiome that occur due to metronidazole, to better understand the mechanisms of antibiotics in gastrointestinal disease. GI19 GASTROKINE MRNA EXPRESSION IS UP‐REGULATED IN GASTRIC TISSUE FROM DOGS WITH HELICOBACTER COLINIZATION BUT WITHOUT INFLAMMATORY CHANGE Mellora Sharman 1, Barbara Bacci1, Leilani Santos1, Caroline Mansfield1 1University of Melbourne, Werribee, Australia Gastrokines (GKNs) are bioactive substances secreted by selected gastric cells. Evidence supports functional roles for GKNs in gastric homeostasis, immune responses and tumour suppression. Down‐regulation has been reported in Helicobacter pylori associated gastritis and other inflammatory gastrointestinal conditions. The aim of this study was to develop a quantitative polymerase chain reaction (qPCR) assay for utilization to evaluate GKN gene expression in Helicobacter spp positive dogs before and after treatment. Expression of GKN‐1 and GKN‐2 mRNA was studied in endoscopic biopsy samples collected from seven dogs over three time‐points pre‐ (T1) and at 1 (T2) and 18 (T3) weeks post‐treatment for Helicobacter spp colonization. The relative expression software tool (REST) was used to provide efficiency corrected expression ratios for comparisons between groups and these results were compared to a standard 2∆∆CT methodology. Compared with T2 GKN1 and GKN2 mRNA expression was up‐regulated at T1 by a mean factor of 2.53 (SE = 1.83 – 3.54) for GKN1 (P = 0.000) and 2.85 (SE = 2.23 – 3.75) for GKN2 (P = 0.000). This difference was attenuated when comparisons were made between T1 and T3. Histopathologically evidence of gastritis was not present in any Helicobacter spp positive sample. When compared to immediate post‐eradication samples GKN gene expression is up‐regulated in the presence of Helicobacter spp. in dogs without evidence for concurrent inflammation. Further evaluation is required to identify the relevance of this finding, however given a suspected role in gastric homeostasis, upregulation could contribute to limit development of gastritis in Helicobacter spp positive dogs. GI20 FECAL SHORT‐CHAIN FATTY ACID CONCENTRATIONS IN DOGS WITH CHRONIC ENTEROPATHY Yasushi Minamoto 1, Tomomi Minamoto1, Panpicha Sattasathuchana1, Agostino Buono1, Blake Guard1, Julia Honneffer1, Jörg Steiner1, Jan Suchodolski1 1Gastrointestinal Laboratory, Texas A&M University, College Station, TX, USA Accumulating evidence has shown a significant relationship between the gastrointestinal (GI) microbiota and host GI health. Microbial metabolites are believed to play a critical role in host‐microbial interactions. Short‐chain fatty acids are major end‐products of bacterial carbohydrate fermentation in the intestinal tract. They constitute an important energy source for intestinal epithelial cells and help maintain intestinal mucosal integrity. Decreased concentrations of short‐chain fatty acids have been observed in humans with GI disease. However, large scale clinical data are lacking in small animal medicine. Therefore, the aim of this study was to evaluate fecal concentrations of short‐chain fatty acids in healthy pet dogs and dogs with chronic enteropathy. One fresh fecal sample was collected from each of 51 healthy control dogs and 65 dogs with chronic enteropathy. Fecal concentrations of short‐chain fatty acids (i.e., acetate, propionate, butyrate, and total short‐chain fatty acids) were measured using gas chromatography/mass spectrometry. Fecal short‐chain fatty acid concentrations were normalized by dry matter and compared between both groups using a Wilcoxon rank‐sum test. The Benjamini & Hochberg's False Discovery Rate was used to correct for multiple comparisons and an adjusted P < 0.05 was considered to be statistically significant. Fecal concentrations (median [range]) of acetate were significantly decreased (P = 0.024) in dogs with chronic enteropathy (185.8 [20.1‐1042.1] μmol/g of dry matter) compared to those in healthy control dogs (233.7 [87.7‐672.8] μmol/g of dry matter). Also, fecal concentrations of propionate were significantly decreased (P < 0.001) in dogs with chronic enteropathy (46.2 [0‐224.4] μmol/g of dry matter) compared to those in healthy control dogs (102.1 [1.6‐266.8] μmol/g of dry matter). Moreover, fecal concentrations of total short‐chain fatty acids were significantly decreased (P = 0.010) in dogs with chronic enteropathy (247.7 [25.9‐1378.2] μmol/g of dry matter) compared to those in healthy control dogs (397.9 [126.6‐927.0] μmol/g of dry matter). However, there was no significant difference (P = 0.178) in fecal concentrations of butyrate between healthy control dogs (31.3 [8.1‐148.1] μmol/g of dry matter) and dogs with chronic enteropathy (23.6 [0‐137.6] μmol/g of dry matter). In this study, dogs with chronic enteropathy had decreased fecal concentrations of acetate, propionate, and total short‐chain fatty acids. The relationship between fecal short‐chain fatty acid concentrations and microbial composition is under investigation. GI21 ASSESSMENT OF SERUM LIPOPROTEIN PROFILES IN DOGS WITH CHRONIC ENTEROPATHY AND HEALTHY CONTROL DOGS Tomomi Minamoto 1, Rosemary Walzem2, Suchodolski Jan1, Steiner Joerg1 1Gastrointestinal Laboratory, Texas A&M University, College Station, USA, 2Department of Nutrition and Food Sience, Texas A&M University, College Station, USA The main function of lipoproteins is to transport lipids in the blood stream. Lipoproteins are classified into 5 major classes based on their densities: chylomicrons, very low density lipoproteins (VLDL), intermediate density lipoproteins (IDL), low density lipoproteins (LDL), and high density lipoproteins (HDL). Based on the human literature, altered proportions of lipoprotein classes have been described in certain disease states, such as cardiovascular disease, metabolic disease, and inflammatory diseases, including inflammatory bowel disease and systemic lupus erythematosus (SLE). It is known that infection and chronic inflammation impair lipoprotein metabolism and cause a variety of changes in plasma concentration of lipids and lipoproteins in human patients with SLE. However, little is known about lipoprotein profiles of dogs with chronic enteropathy (CE). The aim of this study was to investigate serum lipoprotein profiles, cholesterol, and triglyceride concentrations in dogs with CE and healthy control dogs. Analysis of serum lipoprotein profiles was performed in 31 healthy control dogs and 43 dogs with CE using a density gradient ultracentrifugation. The image of each tube following ultracentrifugation was converted to a density profile using a commercially available software program (OriginPro 7.5). The area under the curve (AUC) of triglyceride‐rich lipoproteins, LDL, HDL, and total lipoproteins were calculated for analysis. In addition, serum cholesterol and triglyceride concentrations were measured using a commercially available clinical chemistry analyzer (SIRRUS®, STANBIO, Boerne, TX, USA). Data were analyzed using a Wilcoxon rank sum test or t‐test where appropriate. Significance was set at P < 0.05. Total lipoproteins, LDL, and HDL levels in healthy dogs were significantly higher than in dogs with CE (P < 0.0001, P = 0.0022, and P < 0.0001, respectively). Serum cholesterol concentrations in healthy control dogs (mean ±SD: 237 ± 59 mg/dL) were significantly higher than those in dogs with CE (mean ± sd: 180 ± 88 mg/dL, P = 0.002). This study suggests that lipoprotein profiles are altered in dogs with CE. Investigating clinical importance of lipoprotein profiling in dogs with CE is warranted. GI22 SERUM HOMOCYSTEINE CONCENTRATIONS IN GREYHOUNDS Niels Grützner 1,2, Christina M Iazbik3, Romy M Heilmann1, Rosana Lopes1, Seth C Bridges1, Jan S Suchodolski1, Guilermo C Couto4, Joerg M Steiner1 1Gastrointestinal Laboratory, College Station, USA, 2Farm Animal Clinic, Bern, Switzerland, 3Department of Veterinary Clinical Sciences, Ohio, USA, 4Couto Veterinary Consultants, Hilliard, USA In humans, hyperhomocysteinemia is a multifactorial and incompletely understood condition. Increased serum homocysteine (HCY) concentrations have been associated with cardiovascular disease, peripheral arterial occlusive disease, and venous thrombosis. Recently, cardiovascular/thrombotic disease and hypocobalaminemia have been described in Greyhounds. It was also suggested that hyperhomocysteinemia in Greyhounds with suspected gastrointestinal disease is due to hypocobalaminemia and hypofolatemia. Both conditions have also been linked to hypoalbuminemia, which may affect HCY levels, as shown previously in dogs. However, it is unknown if low levels of both micronutrients (cobalamin and/or folate) are present only in Greyhounds with gastrointestinal disease or also in healthy Greyhounds. Therefore, the aim of this study was to evaluate serum HCY, cobalamin, and folate concentrations in Greyhounds with diarrhea or thrombotic disease as well as in healthy Greyhounds. Serum samples from healthy Greyhounds (n = 16), Greyhounds with diarrhea (n = 30), or a history of thrombotic events (n = 8) were collected at the Ohio State University and Texas A&M University. Serum HCY concentrations (reference interval [RI]: 5.0‐22.1 μmol/L) were measured by gas chromatography/mass spectrometry. Concentrations of cobalamin (RI: 251‐908 ng/L) and folate (RI: 7.7‐24.4 μg/L) were measured using an automated chemiluminescence assay. Concentration of serum HCY, cobalamin, and folate were compared among the three groups of Greyhounds using an ANOVA or a non‐parametric Kruskal‐Wallis test, as applicable. Correlation analysis was performed to test for any possible correlation between HYC and the two B vitamins. Serum HCY concentrations differed among the three groups of Greyhounds (P = 0.0012). Healthy Greyhounds had significantly higher serum HCY concentrations (mean ±SD: 65.2 ± 24.2 μmol/L) than Greyhounds with diarrhea (39.7 ± 21.7 μmol/L; P < 0.01) or thrombosis (36.7 ± 22.1 μmol/L; P < 0.05). All healthy Greyhounds had serum HCY concentrations above the upper limit of the RI. Serum cobalamin and folate concentrations did not differ among the groups of Greyhounds (both: p > 0.05). A negative correlation was observed between serum HCY and serum cobalamin (ρ: ‐0.61; 95%CI: ‐0.76 to ‐0.39; P < 0.0001) or serum folate concentrations (ρ: ‐0.28; 95%CI: ‐0.52 to ‐0.01; P = 0.0386) when analyzing all samples together. However, no correlation was observed when analyzing the three groups of Greyhounds separately (for all: p > 0.05). Healthy Greyhounds had higher serum HCY concentrations than Greyhounds with diarrhea or thrombotic disease. All healthy Greyhounds were hyperhomocysteinemic, which suggests that Greyhounds could serve as a novel canine model to further investigate hyperhomocysteinemia in humans. However, further studies are warranted to characterize this model. GI23 EFFECT OF THE PROBIOTIC ENTEROCOCCUS FAECIUM SF 68 ON PRESENCE OF DIARRHEA IN WEANLING KITTENS Stanley Marks 1, Samantha McDonnel1, Carol Smith1, Lisa Fowler1, Valerie Goetting1, Christine Chaveste1, Courtney Cook1, Kelly Keating1, Philip Kass1 1University of California, Davis, School of Veterinary Medicine, Davis, CA, USA Diarrhea in kittens is a frequent malady facing veterinarians and managers of feline shelters and catteries; however, there is scant literature providing specific information on causes and management of this problem. The objectives of this study were to determine the prevalence of fecal enteropathogens in weanling kittens and assess whether administration of Enterococcus faecium SF 68 affected fecal quality. 135 kittens 1‐2 weeks of age were entered into the 6‐week trial. Kittens were maintained on PetAg KMR® milk replacer for a 1 week acclimation period prior to randomization into groups of 2 or 3 for probiotic or placebo administration. Kittens were maintained on KMR® and treated for 2 weeks prior to weaning onto a commercial canned diet at week 4. Fresh feces from all kittens were scored daily for consistency (score 1 =  liquid diarrhea; 2 =  soft, unformed; 3 =  soft, formed; 4 =  normal). Feces were evaluated for enteropathogens via centrifugation flotation, bacterial culture, toxin immunoassays, and PCR. 81 kittens completed the trial (placebo n = 36; probiotic n = 45). Enteropathogens detected included: Isospora felis (16.2%), Cryptosporidium spp. (8.2%), Clostridium difficile toxin A/B (27.3%), C. perfringens enterotoxin (2.6%), Salmonella (2.7%), Panleukopenia (1.4%), and Coronavirus (6.8%). Median fecal consistency scores during weeks 4 and 5 were ≥ 3 in both groups, but were associated with significantly improved scores in the placebo versus probiotic group (P = 0.013). Administration of the probiotic Enterococcus faecium SF 68 to weanling kittens did not result in improved fecal scores prior to or during weaning. GI24 FECAL N‐METHYLHISTAMINE CONCENTRATIONS IN DOGS WITH EXOCRINE PANCREATIC INSUFFICIENCY Joseph Cyrus Parambeth 1, Jan Suchodolski1, Joerg Steiner1 1Gastrointestinal Laboratory, Texas A&M University, College Station, Texas, USA Exocrine pancreatic insufficiency (EPI) is a syndrome characterized by inadequate synthesis and secretion of pancreatic digestive enzymes, resulting in maldigestion. Mast cells within the gastrointestinal tract contribute to the pathogenesis of canine chronic enteropathies because of their ability to release inflammatory mediators such as histamine. N‐methylhistamine (NMH) is a major metabolite of histamine, and increased fecal concentrations of N‐methylhistamine have been reported in dogs with chronic gastrointestinal disease. As EPI in dogs can be complicated by chronic enteropathies, quantification of fecal NMH may serve as a potential prognostic marker. The aim of this study was to evaluate fecal N‐methylhistamine concentrations in dogs with EPI. Surplus fecal samples from 21 dogs enrolled in an unrelated clinical trial at the Gastrointestinal Laboratory were utilized. To be included into the study, the dogs had to be at least 1 year of age, have clinical signs of EPI (i.e., polyphagia, weight loss, steatorrhea, and/or loose, voluminous and/or malodorous stools), have a serum cTLI concentration ≤2.5 μg/L, not be pregnant or lactating, and be free from any clinically apparent disease other than EPI. Three naturally voided fecal samples collected over three consecutive days were immediately frozen after collection and were used for the study. Fecal N‐methylhistamine concentrations were measured using a previously validated in‐house assay using stable isotope dilution gas chromatography/mass spectrometry (GC–MS). A liquid extract of each fecal sample (1:5 dilution) was used for NMH analysis and fecal NMH concentrations were back‐calculated for the wet weight of the fecal samples and expressed in ng/g feces. Results were compared with the established reference interval for healthy dogs and a mean three‐day fecal NMH >191 ng/g feces or a maximum fecal NMH of one individual sample of >334 ng/g feces was considered abnormal. Forty three percent (9/21) of the EPI dogs had increased fecal NMH concentrations (median [range]: 97.1 μg/g feces [0 – 9,973]). German Shepherds and German Shepherd cross breed dogs made up 43% of dogs in the study (9/21), but 78% (7/9) of these dogs had increased fecal NMH concentrations with an odds ratio of 17.5 (95% confidence interval (CI): 2 to 15.6; P = 0.0102) when compared to other breeds. This study showed that fecal N‐methylhistamine concentrations may be increased in dogs with EPI, especially when German Shepherds or German Shepherd mixes. Further studies are necessary to elucidate the mechanisms that are responsible for this finding and if increased fecal NMH concentrations are associated with an altered clinical outcome. GI25 THE EFFECT OF THE PROBIOTIC SIVOY™ ON CLINICAL AND HISTOPATHOLOGICAL PARAMETERS IN CATS WITH CHRONIC IDIOPATIC CONSTIPATION AND MEGACOLON Giacomo Rossi 1, Albert Jergens2, Matteo Cerquetella1, Sara Berardi1, Graziano Pengo3, Jan Suchodolski4 1University of Camerino, Camerino Italy Marche, Italy, 2Iowa State University, Ames, USA, 3Private Clinic S. Antonio, Madignano Cremona, Italy, 4Texas A&M University, College Station, USA The pathogenesis of chronic constipation (CC) and idiopathic megacolon (IMC) are poorly understood in humans and animals. In particular, it is unknown whether there are abnormalities involving the extrinsic nerves, the enteric nerve plexuses, interstitial cells of Cajal (ICC) or the intestinal smooth muscle. Abnormalities of any of these components could lead to gut dilatation and impaired motility. Chronic constipation and megacolon occurs more often in the cat than the dog. In humans, probiotics have been increasingly investigated in the management of these colonic motility dysfunctions, particularly their effect on gut transit time, stool output, and constipation relief. While probiotics are used frequently in small animal practice, there are no published studies regarding their clinical efficacy in cats with CC and IMC. The aim of the study was to investigate the clinical and histological effects of a commercial multi‐strain probiotic (SIVOY™), containing 200 billion lactic acid bacteria comprised of the following strains: L. acidophilus DSM24735, L. plantarum DSM24730, L. paracasei DSM 24733, L. delbrueckii subsp. bulgaricus DSM24734, L. brevis CD2 #11988, Streptococcus thermophilus DSM 24731, B. longum DSM 24736, and B. infantis DSM 24737. Ten pet cats of different breeds and ages with a diagnosis of chronic constipation, non‐responsive to medical management were selected on the basis of recurrence of clinical signs and absence of any antibiotic treatment for a month. Three of these 10 cats suffered from IMC and full thickness biopsies were sampled for histology. In all animals enrolled in the study, the colon was found to be dilated and impacted with faeces, with the most severe dilatation occurring in the transverse and descending colon, as observed via imaging. CC (n = 7) and IMC (n = 3) cats received orally 200 billion lyophilized bacteria daily for 90 days. Cats were assessed clinically, endoscopically, and histologically at baseline (T0) and after the end of treatment (T1). Histological samples were scored, then evaluated for the immunohistochemical (IHC) expression of CD117 +  ICC, enteric neurons, glial cells, and gangliar cell apoptosis (these three latter parameters were evaluated only in the full thickness biopsy obtained at T0 from cats with IMC). Concerning CD117 IHC for ICC evaluation, mast cells, which are known to express the same antigen, were differentiated by staining with Alcian blue, and numbered. Data from constipated cats were compared before and after therapy, and with those obtained from healthy control tissues (archived material from five healthy cats) using paired t‐tests or Wilcoxon matched pairs tests, where appropriate. Statistical significance was set at P < 0.05. Constipated cats displayed a significant decrease in ICC, and cats with IMC had significantly more apoptotic enteric neurons than controls. After treatment with SIVOY™, significant decreases were observed for FCEAI clinical index (P = 0.006), and histology scores (P < 0.001). In contrast, a significant increase of CD117 +  ICC was observed (P < 0.05) after probiotic therapy. Cats with CC and IMC showed significant clinical improvement after SIVOY™ treatment, and histological parameters suggest a potential anti‐inflammatory effect of SIVOY™, associated with a reduction of mucosal infiltration, and restoration of the number of ICC. The evaluation of microbiota composition after probiotic treatment is in progress to further understand the effects of probiotic therapy on chronic constipation in cats. GI26 EFFECTS OF A SYNTHETIC SERINE PROTEASE INHIBITOR, CAMOSTAT MESILATE (FOY‐305), ON MARKERS OF PANCREATIC ACINAR CELL DAMAGE, INFLAMMATION, AND FIBROSIS IN DOGS WITH SUSPECTED NATURALLY OCCURING CHRONIC PANCREATITIS Tim Kretzschmar 1, Jan Suchodolski1, Joerg Steiner1 1Texas A&M, College Station, TX, USA Chronic pancreatitis (CP) in dogs is a clinically underdiagnosed condition characterized by acinar cell atrophy, fibrosis, and loss of exocrine function. Currently, there is no approved pharmaceutical treatment for chronic pancreatitis in dogs, thus management relies heavily on supportive care. Camostat mesilate (CM; FOY‐305) is a synthetic protease inhibitor, which has been shown to have inhibitory effects on trypsin, the kinin system, as well as inflammatory and fibrotic cascades. CM has been approved for the treatment of chronic pancreatitis in humans in Japan for over 20 years. However, data on the efficacy of CM in dogs with chronic pancreatitis are lacking. Thus, the purpose of this study was to evaluate the effects of oral CM therapy on markers of acinar cell damage, inflammation, and fibrosis in dogs with suspected CP. Thirty one privately owned dogs with suspected chronic pancreatitis were enrolled into the study. Dogs were included based on two consecutive canine pancreatic lipase immunoreactivity (cPLI) measurements >400 μg/L within 12 ± 4 days and a serum triglyceride concentration below 500 mg/dL. Dogs with major concurrent disease (e.g., renal disease, icterus) were excluded. Dogs were either treated with 12 mg/kg/d (n = 12) or 24 mg/kg/d (n = 19) oral CM for 26 ± 5 days. A complete blood count, serum biochemistry profile, and assays for the measurement of cPLI, canine trypsin‐like immunoreactivity (cTLI), cobalamin, and folate concentrations were performed before and after treatment. In addition, serum TGF‐β1, S100A12, and C‐reactive protein (CrP) concentrations were evaluated after completion of the trial as a batch measurement. Pre‐ and post‐treatment results were compared using a paired t‐test for normally distributed data sets and a Wilcoxon matched‐pairs test when Gaussian distribution could not be assumed. The significance threshold was set at P = 0.05 Serum cPLI concentrations significantly decreased from pre‐treatment (median 690 μg/L; range 344 to 6,052 μg/L) to post‐treatment (median 514 μg/L; range 125 to 2,836 μg/L; P = 0.0409) in both treatment groups combined and in the subgroup of dogs treated with 24 mg/kg/d oral CM from pre‐treatment (median: 847 μg/L; range: 414 to 2,024 μg/L) to post‐treatment (median: 520; range: 193 to 2,580 μg/L; P = 0.016). In dogs treated with 12 mg/kg/d oral CM, serum cPLI concentration did not change significantly between pre‐treatment (median 501 μg/L; range 344 to 6,052 μg/L) and post‐treatment (median 453 μg/L; range 125 to 2,836 μg/L; P = 0.9697) measurements. Serum cobalamin concentrations decreased significantly between pre‐treatment (mean: 843 ng/L; SD: ± 198 ng/L) and post‐treatment (mean: 681 ng/L; SD: ± 239 ng/L; P = 0.003) in both treatment groups combined, as well as both subgroups (12 mg/kg/d oral CM; pre‐treatment: mean 868 ng/L; SD: ± 108 ng/L; post‐treatment: mean 695 ng/L; SD: ± 223 ng/L P = 0.0363 and 24 mg/kg/d oral CM; pre‐treatment: mean 749 ng/L; SD: ± 229 ng/L; post‐treatment: mean 641 ng/L; SD: ± 252 ng/L: P = 0.045). Treatment resulted in no significant differences in serum concentrations of cTLI, folate, TGF‐β1, S100A12, or CrP in all dogs combined or either subgroup. The decrease in cPLI concentrations would suggest an attenuation of acinar cell damage but evaluation of efficacy requires further controlled studies. Despite marked individual decreases, no significant anti‐inflammatory or anti‐fibrotic properties were identified. Oral CM therapy at doses up to 24 mg/kg/d for up to 31 days was found to be safe in dogs with suspected chronic pancreatitis, with only mild adverse effects reported. GI27 BETA‐HYDROXYBUTYRATE CONCENTRATIONS IN DOGS WITH ACUTE PANCREATITIS AND WITHOUT DIABETES MELLITUS Frances Hurrell 1, Kenneth Drobatz1, Rebecka Hess1 1Matthew J. Ryan Veterinary Hospital, University of, Philadelphia, PA, USA The purpose of this study was to investigate β‐hydroxybutyrate (BOHB) concentrations in dogs with acute pancreatitis (AP) and without diabetes mellitus. Dogs without diabetes mellitus were prospectively enrolled into one of three groups: dogs with AP, sick dogs without AP, and fasted dogs. Dogs with AP (43) were enrolled if they had vomiting, diarrhea, or anorexia and ultrasonographic findings consistent with AP including an enlarged, irregular, hypoechoic pancreas and hyperechoic mesentery or an increased pancreatic lipase immunoreactivity (PLI>200 mcg/L). Sick dogs without AP (71) were enrolled if they had vomiting, diarrhea, or anorexia and a normal pancreatic ultrasound or a PLI≤200 mcg/L. Dogs fasted for 12 hours were enrolled if they had no vomiting, diarrhea, or anorexia (42). BOHB was measured on whole blood with a portable ketone‐meter. The Kruskal Wallis test was performed to confirm that BOHB was significantly different in the three groups (P = 0.0001). Pairwise comparisons were performed using the Mann‐Whitney test and Bonferroni corrected p‐values are reported. Median BOHB was significantly higher in dogs with AP (0.3 mmol/L) compared to sick (0.2 mmol/L, P = 0.004) or fasted dogs (0.1 mmol/L, P = 0.0003), and in sick compared to fasted dogs (P = 0.0003). No significant associations were detected between BOHB and duration of anorexia, body condition, body temperature, white blood cells, glucose, cholesterol or lactate concentrations in sick or AP dogs. It is concluded that BOHB is significantly higher in dogs with AP compared to other dogs without diabetes mellitus. The diagnostic and prognostic utility of this finding remains to be investigated. GI28 ENDOSCOPIC RETROGRADE CHOLANGIOGRAPHY AND BILIARY STENT PLACEMENT FOR BILIARY OBSTRUCTION IN DOGS AND CATS Allyson Berent1, Chick Weisse1, Mark Schattner2, Pari Shah2, Hans Gerdes2 1The Animal Medical Center, New York, NY, USA, 2Memorial Sloan Kettering Cancer Center, New York, NY, USA Surgical treatment for extrahepatic bile duct obstruction (EHBDO) in veterinary patients has a high morbidity and mortality. Endoscopic retrograde cholangiography (ERC) and endoscopic biliary stent placement (EBS) is a minimally‐invasive alternative used in humans with high success and few complications. The purpose of this study was to describe the technical and clinical outcomes using ERC/EBS for the treatment of EHBDO in dogs and cats. ERC/EBS was performed using a 7.5 or 11.5 mm side‐viewing duodenoscope. A sphincterotome cannulated the common bile duct (CBD) through the major duodenal papilla (MDP). Contrast fluoroscopic‐cholangiography confirmed EHBDO. A guidewire was advanced up the CBD. A stent (plastic or metallic) was advanced through the endoscope, over the guidewire, into the CBD, crossing the MDP. Patients were followed for biochemical, imaging, and overall outcomes. Five dogs (2.24‐50 kg) and 3 cats (2.9‐4.7 kg) were included with a median Tbili of 14.5 mg/dL (4.9‐26.7): pancreatitis‐induced stricture (n = 5), choledocholithiasis (n = 2), granuloma (n = 1). Pyloric cannulation was successful in all animals, ERC in 4/5 dogs and 2/3 cats, and EBS in 4/5 dogs and 1/3 cats. Catheter perforation of the duodenum occurred in one cat. A surgically‐assisted fluoroscopic‐guided stent was placed in 2 cats and a subcutaneous‐intestinal biliary‐bypass device in 1 dog that failed ERC/EBS. All patients survived to discharge and had resolution of EHBDO. Median follow‐up time was 82 days (30‐772 days) with 62.5% still alive. None died from EHBDO. ERC and EBS are possible and effective in dogs and cats with EHBDO. This procedure is technically difficult. This technique may be a potential alternative to traditional surgery in the future. GI29 EFFECT OF BIFIDOBACTERIUM ANIMALIS AHC7 AND TYLOSIN ON THE MICROBIOME DURING THE QUARANTINE PERIOD OF MILITARY WORKING DOGS Sarah Cooper 1, Nolie Parnell2, Anitha Isaiah3, Jorg Steiner3, Jan Suchodolski1 1United States Army, Fayetteville, NC, USA, 2Purdue University, West Lafayette, IN, USA, 3Texas A&M Gastrointestinal Laboratory, College Station, TX, USA The canine gastrointestinal microbiome is a diverse population of microorganisms that provides a number of host benefits. A stable microbiome is important for overall host health, and alterations in the microbiome have been reported during periods of stress. Military Working Dogs are often exposed to stressful events including their initial acquisition period in which they are procured from Europe and placed in quarantine for ten days after arrival to the United States. This quarantine period has historically been associated with a high incidence of diarrhea. The aims of this study were to compare the effects of Bifidobacterium animalis AHC7, tylosin, and placebo on the microbiome during this ten‐day quarantine period. In this double‐blinded study, Military Working Dogs (n = 48) were randomly assigned to one of three treatment groups: placebo, probiotic (Bifidobacterium animalis AHC7), or tylosin. Daily routine during the quarantine period, including diet and handlers, was the same for all dogs. A fecal sample was collected at the beginning and the end of the quarantine period for fecal 16S rRNA analysis by Illumina sequencing. Fecal scores were recorded daily, and dogs were characterized as having diarrhea (fecal score 6‐7) or not having diarrhea (fecal score 1‐5). Alpha‐ and beta‐diversity measures were compared between the three treatment groups before and after treatment and between dogs with and without diarrhea. Differences in bacterial communities were analyzed using unweighted unifrac distance metric. Analysis of similarity (ANOSIM) test was used to determine if any of the groups studied had significantly different bacterial communities. Linear discriminant analysis (LDA) effect size (LEfSe) was performed to identify bacterial groups that were significantly associated with treatment. Alpha‐diversity analysis of the pre‐treatment samples showed a significantly decreased species richness and Shannon diversity in the tylosin group compared to the probiotic group (P = 0.0131 and P = 0.0210, respectively). There was no significant difference in alpha‐diversity analysis before treatment between dogs with and without diarrhea during the quarantine period. When comparing the groups before and after treatment, regardless of fecal score, there was a significant difference in the alpha‐diversity and Shannon diversity in both the probiotic (P < 0.0001 and P < 0.0001, respectively) and tylosin (P < 0.0001 and P < 0.0006, respectively) groups but not the placebo group. Principal coordinates analysis plots also showed a significant difference in clustering in the tylosin group compared to the placebo (P = 0.001) and probiotic (P = 0.001) groups after treatment. There was also significant difference in clustering before and after treatment for all groups: placebo (P = 0.001), tylosin (P = 0.001), and probiotic (P = 0.001) with the largest effect seen in the tylosin group (R = 0.508). Bifidobacterium spp. increased in all groups after treatment, and Clostridium spp. decreased in all groups after treatment. In this study, bacterial diversity decreased during the quarantine period in all groups with the largest change seen in the tylosin group. Bifidobacterium spp. also increased in all treatment groups during the study period while Clostridium spp. decreased in all groups. These findings would suggest that while the microbiome during the quarantine period of Military Working Dogs is influenced by treatment, factors other than treatment can impact the microbiome. The views expressed in the article are those of the author and do not reflect the official policy or position of the Department of the Army, Department of Defense, or the US Government. GI30 ASSOCIATION OF TWO UNSTABLE HAS2 GENE DUPLICATIONS WITH COBALAMIN DEFICIENCY IN THE CHINESE SHAR‐PEI Niels Grützner 1,2, Jennifer Meadows3, Mia Olsson4, Romy M Heilmann1, Jan S Suchodolski1, Kerstin Lindblad‐Toh3, Joerg M Steiner1 1Gastrointestinal Laboratory, College Station, USA, 2Farm Animal Clinic, Bern, Switzerland, 3Department of Biomedical Biochemistry & Microbiology, Uppsala, Sweden, 4Faculty of Medicine, University of British Columbia, Vancouver, Canada A recent survey based on an illustration showed that in the USA cobalamin deficiency occurs more frequently in the traditional type Chinese Shar‐Pei (Shar‐Pei) than in the meatmouth type Shar‐Pei. Also, altered copy numbers of a 16.1 kb‐duplication and a 14.3 kb‐duplication have been described in different types of Shar‐Peis (e.g., meatmouth and traditional type). Both duplications are located in close proximity to the HAS2 gene and have not been investigated in Shar‐Peis with cobalamin deficiency. Therefore, the copy numbers of the two duplications were determined in cobalamin‐deficient and normocobalaminemic Shar‐Peis. Genomic DNA extracts from 20 cobalamin‐deficient and 36 normocobalaminemic Shar‐Peis were used. Genotyping was performed using two assay sets, 1) the CNV‐Eastern assay [16.1 kb‐duplication] and 2) the CNV_23.759 assay [14.3 kb‐duplication] normalized to assay C7orf28B. Copy numbers obtained were compared between cobalamin‐deficient and normocobalaminemic Shar‐Peis by use of a Mann‐Whitney U test. A Fisher's exact test was used and the odds ratios (OR) calculated to evaluate whether low or high copy numbers of the two duplications are associated with cobalamin deficiency. Cobalamin‐deficient Shar‐Peis had higher CNV‐Eastern assay copy numbers (median [range]: 5.0 [2.0‐7.0]) and lower CNV_23.759 assay copy numbers (2.1 [1.0‐8.6]) than normocobalaminemic Shar‐Peis (2.5 [1.7‐7.1] and 5.3 [2.4‐11.6], respectively; both: P < 0.0001). Receiver‐operating characteristic curves were used to determine the optimal copy number that distinguished cobalamin‐deficient from normocobalaminemic Shar‐Peis. High copy numbers (≥4.5) in the CNV‐Eastern assay was detected in 12 (60%) of the cobalamin‐deficient Shar‐Peis, and cobalamin deficiency was associated with a high CNV‐Eastern copy number (OR: 25.5 [CI: 4.7‐137.4]; P < 0.0001). The CNV_23.759 assay revealed low copy numbers (≤2.5) in 12 (60%) of the cobalamin‐deficient Shar‐Peis, and cobalamin deficiency was associated with a low CNV_23.759 copy number (OR: 52.5 [CI: 5.9‐464.6]; P < 0.0001). A high CNV‐Eastern copy number, low CNV_23.759 copy number, and high CNV‐Eastern or low CNV_23.759 copy number (OR: 81.7 [CI: 9.0‐741.4]; P < 0.0001) distinguished cobalamin‐deficient Shar‐Peis from normo‐cobalaminemic Shar‐Peis with a sensitivity of 60, 60, and 70%, respectively, and a specificity of 94, 97, and 97%, respectively. Both copy number assays identified cobalamin‐deficient Shar‐Peis with a moderate sensitivity and a high specificity, and a positive cut‐off signal on at least one assay had the highest diagnostic accuracy. GI31 THE FECAL MICROBIOME OF COYOTES (CANIS LATRANS) AND A COMPARISON WITH HEALTHY PET DOGS (CANIS CANIS) Felipe Pierezan1, Julia Honneffer 2, Aline Rodrigues Hoffmann2, Jörg Steiner2, Jan Suchodolski2 1Universidade Federal de Minas Gerais, Belo Horizonte, Minas Gerais, Brazil, 2Texas A&M University, College Station, Texas, USA The characterization of the fecal microbial composition and diversity of different species can demonstrate species‐specific associations and provide insights on how self and external factors shape the bacterial communities inhabiting the gastrointestinal tract. The comparison of data from two host species of the same genus, such as coyotes (Canis latrans) and dogs (Canis canis), with known feeding and environmental behaviors, is a useful model to access these aspects, which, in the future, will help to understand the influence of the microbiota on host development, nutrient assimilation and susceptibility to diseases. This study aim was to characterize the fecal microbiome of coyotes and compare it to healthy pet dogs. DNA extracted from fecal samples collected from the rectum of coyotes (n = 10), 7‐20 hours post mortem, and from healthy pet dogs (n = 15) were used for Illumina sequencing of the bacterial 16S rRNA gene. Quantitative Insights Into Microbial Ecology software was used to characterize the samples. The differences in the proportions of bacterial taxa between the species were evaluated using Kruskal‐Wallis tests. A total of 2,044,203 (median: 80,546; range 57,429–108,532 sequences per sample) were obtained. The fecal microbiome of coyotes was composed predominately by bacteria belonging to phyla Bacteroidetes (median 36.5%), Firmicutes (31.7%), Fusobacteria. (31.7%) and Proteobacteria (1.0%). The most common classes identified included Bacteroidia (36.5%), Fusobacteria (31.7%), Clostridia (20.7%), and Erysipelotrichi (1.4%). At the family level, Fusobacteriaceae (31.7%), Bacteroidaceae, (28.2%), Paraprevotellaceae (8.2%), Lachnospiraceae (5.7%) and Erysipelotrichaceae (1.4%) were the most common bacterial groups identified. Bacteria belonging to the phyla Bacteroidetes (36.5% versus 15.0%; P < 0.0458), Fusobacteria (31.7% vs. 4.3%; P < 0.0009), and Proteobacteria (1.0% vs. 0.0%; P < 0.0003) were significantly more frequently identified in coyotes than in healthy pet dogs samples. The higher proportions of these phyla in the coyotes samples was mostly due to significantly increased proportions of the families Bacteroidacea (36.5% vs. 15.1%; P < 0.04) and Fusobacteriaceae (31.7% vs. 4.3%; P < 0.009). Within Proteobacteria, class Betaproteobacteria and class Gammaproteobacteria were significantly more present in samples from coyotes. The families Paraprevotellaceae, Prevotellaceae, Odoribacteraceae, and S24‐7 of the phylum Bacterioidetes, as well as, Alcaligenaceae and Enterobacteriaceae of the phylum Proteobacteria, were significantly increased in coyotes samples. Fecal samples from healthy pet dogs showed higher proportions of bacteria belonging to the phylum Firmicutes (78.4% vs. 31.7%; P < 0.0009), influenced by significantly increased proportions of bacteria of the classes Clostridia (58.9% vs. 20.7%; P < 0.0004) and Erysipelotrichi (6.8% vs. 1.4%; P < 0.001). Consequently, there were significant higher proportions of the families Turicibacteraceae, Clostridiaceae and Lachnospiraceae. The Shannon diversity index (5.31 vs. 4.49), the Chao1 metric average, and number of observed species (median 1242 vs. 1042) were not significantly different between dogs and coyotes. Principal coordinate analysis plots showed significant clustering between samples from healthy pet dogs and coyotes. This study demonstrates the fecal microbiome of coyotes and dogs are much more diverse than that demonstrated by studies using culture‐based methods. Significant differences between the fecal microbiome of coyotes and healthy pet dogs were observed for numerous bacterial groups at different phylogenetic levels. Most intriguingly, the fecal microbiome of coyotes contains increased proportions of the phyla Bacteriodetes and Fusobacteria, while the fecal microbiome of healthy dogs is composed mainly by the phylum Firmicutes, influenced by high proportions of the class Clostridia. Differences on the type of diet (raw vs. industrialized food) and on the environment habits (outdoor vs. indoor), as well genetics, must play an important role on the microbiome of these two species. Of these factors, diet is the most influencing and, for coyotes, its composition mostly by raw meat, as well by fruits, insects and grass, may have impacted on the predominance of some bacterial groups. Further studies are required to verify the importance of these and identify more factors that shape the bacterial communities in these species. GI32 FOCAL INTESTINAL LIPOGRANULOMATOUS LYMPHANGITIS IN 10 DOGS Alexis Lecoindre 1, Patrick Lecoindre2, Jean‐Luc Cadoré1, Michelle Chevallier3, Sylviane Guerret3, Sean P. Mcdonough4, Kenneth W. Simpson4 1Vetagro sup, Lyon, France, 2CVC veterinary clinic, Lyon, France, 3Laboratoire Biomnis, Lyon, France, 4Cornell University, Ithaca, USA Focal lipogranulomatous lymphangitis (FLL) is a rare form of canine IBD characterized by transmural granulomatous inflammation that involves frequently the distal ileum and ileo‐colic junction and results in a stenosing, mass‐like thickening of these regions of the intestine. The aim of this retrospective study is to describe the clinical and histopathological features, with comparison to human IBD, to determine presence or absence of invasive bacteria, and response to treatment. Ten dogs with definitive histological diagnosis of focal lipogranulomatous lymphangitis were included. Histopathology of endoscopic biospies and surgical resections were reviewed. Bacterial colonization was evaluated using FISH for eubacteria and E. coli. A two‐year follow‐up was available for 10 dogs (only from record in 4 dogs). Eight male and 2 female small‐breed dogs were included. Biochemistry, ultrasound and colonoscopic abnormalities were evaluated. An enterectomy of the abnormal intestinal zone was performed and medical treatment was standardized in all dogs with: metronidazole (10‐15 mg/kg BID), immunosuppressive doses of prednisone with tapering. All biopsies were evaluated by human and veterinary pathologists. Histology revealed mostly non mucosal colitis or/and ileitis lesions with inflammation centered around vessels and mainly concerning the muscularis and serosa with extension into the mesentery. Four dogs remained in remission over a period 3‐30 months, 4/10 presented with persistent diarrhea and 2 dogs were euthanized. There was no FISH based evidence for invasive bacteria. FLL is a deep parietal affection of the intestine with preferential localization to the ileum and ileocolic junction in small‐breed dogs. Bacterial implication seems unlikely. GI33 SERIAL EVALUATION OF SPECIFIC CANINE PANCREATIC LIPASE IMMUNOREACTIVITY AND C‐REACTIVE PROTEIN IN DOGS WITH CERULEIN‐INDUCED ACUTE PANCREATITIS Sue Yee Lim 1, Kensuke Nakamura2, Keitaro Morishita2, Hiroshi Ohta2, Masahiro Yamasaki3, Mitsuyoshi Takiguchi2 1Universiti Putra Malaysia, Serdang, Selangor, Malaysia, 2Hokkaido University, Sapporo, Hokkaido, Japan, 3Iwate University, Morioka, Iwate, Japan Specific canine pancreatic lipase immunoreactivity (Spec cPL) and C‐reactive protein (CRP) are biomarkers of pancreatic inflammation and nonspecific inflammation in dogs, respectively. Both biomarkers increase in dogs with acute pancreatitis (AP). However, concurrent serial measurements of Spec cPL and CRP to monitor progression of AP have not been evaluated. This study was performed to serially evaluate changes in Spec cPL and CRP in dogs with cerulein‐induced AP. 6 dogs received 2 hours of cerulein infusion at 7.5 μg/kg/h, IV to induce AP. Blood samples were obtained before, and at 2, 4, 6, 12, 24, and thereafter at 24‐hour intervals after cerulein infusion, until Spec cPL and CRP returned to reference intervals (RI; Spec cPL ≤200 μg/L, CRP 0‐1 mg/dL). Spec cPL and CRP increased rapidly after cerulein infusion, and progressively decreased with time. Spec cPL reached median (range) peak concentration significantly earlier than CRP [3 (2‐4) versus 18 (12‐48) hours; P = 0.03]. Although not statistically significant, Spec cPL returned to RI earlier than CRP [48 (24‐96) versus 72 (24‐144) hours; P = 0.38]. Median peak concentrations of Spec cPL and CRP were 4,291 (2,736‐6,924) μg/L and 2.3 (0.6‐5.1) mg/dL, respectively. All 6 dogs showed peak Spec cPL above the RI, while 5/6 dogs showed peak CRP above the RI. These findings suggest that Spec cPL and CRP may be useful in monitoring progression of dogs with AP. Without the presence of complications, Spec cPL and CRP will progressively return to RI. Spec cPL peaked and normalized faster than CRP. GI34 VALIDATION OF A URINARY 5‐HYDROXYINDOLEACETIC ACID ASSAY, AND PRELIMINARY USE IN DOGS WITH ENTEROPATHIES Jessica Rigling 1, Craig Ruaux1 1Clincial Sciences, Oregon State University, Corvallis, Oregon, USA The neuroendocrine system of the gut plays a significant role in regulating intestinal motor and sensory functions, as well as modulating inflammation in the intestine. Serotonin (5‐HT) acts locally on the gastrointestinal tract, and excess 5‐HT has been shown in human beings to have a role in the pathophysiology of several gastrointestinal diseases. 5‐Hydroxyindoleacetic acid (5‐HIAA), the main stable metabolite of 5‐HT, is excreted in the urine. Measurement of this compound can be used to determine degree of serotoninergic system activity in an individual, making it a potential marker of serotoninergic system activation in gastrointestinal disease. The aim of this study was to validate a commercially available 5‐HIAA ELISA assay for use with canine urine, establish a reference range for 5‐HIAA:Creatinine in urine samples from 20 clinically healthy dogs, and compare those values to those of dogs with enteropathic disease. A commercially available competitive ELISA assay for 5‐HIAA was validated using acidified canine urine as the sample matrix. Validation included determination of sensitivity, dynamic range, and accuracy via spiking recovery, dilutional parallelism, and assessment of inter‐ and intra‐assay precision. Urine samples from 33 dogs were collected, 20 clinically healthy and 13 showing signs of gastrointestinal disease. Per the manufacturer's guidelines, the samples were acidified with 6M HCl to a pH ≤3.0, and methylated prior to quantification using the ELISA assay. We observed a significant matrix effect from pooled canine urine, dilution studies showed this effect was overcome when using a 1:8 dilution of samples. Urinary creatinine concentrations were determined by the Oregon State University Veterinary Diagnostic Lab using non‐acidifed aliquots. A preliminary reference range for urine 5‐HIAA was derived from the central 95th percentile for urinary 5‐HIAA:Creatinine in the clinically healthy dogs. Sensitivity was 2.0 mg/dL. Observed to expected (O/E) ratios for 3 serial dilutions of urine samples averaged 110.00%, ranging from 80.73 to 124.75%. Spiking recoveries in 3 replicates of 3 urine samples of varying specific gravities averaged 132.66%, ranging from 118.87 to 143.60%. Coefficients of variations for intra‐assay variability ranged from 3.45 to 5.19%, while coefficients of variation for inter‐assay variability ranged from 4.61 to 7.27% in three urine samples of varying specific gravities. The reference range for urine 5‐HIAA:Creatinine in 20 clinically healthy dogs was 0.01 – 0.12. Urinary 5‐HIAA:Creatinine values in dogs with gastroenteropathies were significantly greater than in healthy dogs (median 5‐HIAA:Creatinine 0.04 vs 0.08, healthy dogs vs dogs with enteropathy, P = 0.0034, Mann Whitney test), and showed a significant linear relationship (P = 0.0002, R2 = 0.369) between urinary 5‐HIAA:Creatinine and CIBDAI, a scoring system for severity of canine gastrointestinal disease. We conclude that the commercially available 5‐HIAA ELISA described here is sufficiently sensitive, linear, accurate, precise, and reproducible for use in diluted canine urine samples. Further, we conclude that there is a correlation between 5‐HIAA excreted in urine and severity of gastroenteropathic disease in canines as assessed by the CIBDAI scoring system. GI35 ANALYTICAL VALIDATION OF A GAS CHROMATOGRAPHY/MASS SPECTROMETRY METHOD FOR THE QUANTIFICATION OF 3‐BROMOTYROSINE CONCENTRATIONS IN FECAL SAMPLES FROM DOGS Panpicha Sattasathuchana 1, Hayley J Ask2, Jan S Suchudolski2, Jörg M Steiner2 1Gastrointestinal Laboratory, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University, College Station, TX USA/ Department of Companion Animal Clinical Sciences, Faculty of Veterin, Bangkok, Thailand, 2Gastrointestinal Laboratory, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University, College Station, Texas, USA 3‐Bromotyrosine (3‐BrY) has been used as a marker for eosinophil mediated diseases in humans. In a recent study, serum 3‐BrY concentrations were shown to be increased in dogs with chronic enteropathy. However, fecal samples may be preferable to serum samples for evaluating the activation of eosinophils within the gastrointestinal tract. A method for the measurement of 3‐BrY concentrations in fecal samples from dogs has not yet been established. Therefore, the aims of this study were to (1) develop and analytically validate a method for measuring 3‐BrY concentrations in fecal samples from dogs using electron impact ionization gas chromatography/mass spectrometry and (2) establish a reference interval for fecal 3‐BrY concentrations in healthy control dogs. Surplus canine fecal samples were prepared for analysis by extraction with phosphate buffered saline, 5% newborn calf serum, 1% Triton X‐100, and 0.25 mM thimerosal. For analytical assay validation, the limit of blank (LOB), the limit of detection (LOD), linearity, accuracy, precision, and reproducibility were determined. To determine the reference interval, 3 consecutive fecal samples were collected from 40 healthy control dogs (median age 4.0 years [range: 0.8‐15]; sex: female [n = 19] and male [n = 21]; breed size: small‐sized breed [n = 14], medium‐sized breed [n = 9], and large‐sized breed [n = 17]). The lower 97.5th percentile was used to determine the reference interval of the 3‐day mean or maximum fecal 3‐BrY concentration. The LOB and LOD of the assay were 2.5 and 3.7 mmol/g feces, respectively. The observed‐to‐expected (O/E) ratios for dilutional parallelism (5 different fecal samples diluted at 1/2, 1/4, and 1/8) ranged from 90.0% to 125.2% (mean ±standard deviation (SD): 104.4 ± 10.6%). The O/E for spiking recovery ranged from 85.4% to 119.5% (mean ± sd: 105.4 ± 10.4%) for five different fecal samples that had been spiked with 3‐BrY standards (2.5, 5, 10, or 20 μmol/L). Intra‐assay coefficients of variation (%CV) for five different fecal samples were 4.8%, 5.4%, 12.0%, 13.2%, and 15.2%. Inter‐assay %CVs for four different fecal samples were 7.5%, 9.1%, 12.0%, 13.3%, and 14.2%. The reference intervals for the 3‐day mean and the maximum fecal 3‐BrY concentrations in healthy control dogs were established as ≤23.0 mmol/g feces and ≤38.1 mmol/g feces, respectively. In conclusion, measurement of 3‐BrY using electron impact ionization gas chromatography/mass spectrometry is sensitive, linear, accurate, precise, and reproducible for use with fecal sample from dogs. Studies assessing the clinical utility of measurement of 3‐BrY in fecal samples from dogs are warranted and in progress. GI36 SERUM 3‐BROMOTYROSINE CONCENTRATIONS IN DOGS INFECTED WITH SPIROCERCA LUPI Panpicha Sattasathuchana 1, Paolo Pazzi2, Amelia Goddard2, Jan S Suchodolski3, Jörg M Steiner3, Eran Dvir2 1Gastrointestinal Laboratory, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University, College Station, TX USA/ Department of Companion Animal Clinical Sciences, Faculty of Veterin, Bangkok, Thailand, 2Department of Companion Animal Clinical Studies, Faculty of Veterinary Science, University of Pretoria, Onderstepoort, South Africa, 3Gastrointestinal Laboratory, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University, College Station, TX, USA Spirocerca lupi is a nematode that infects the caudal esophagus in dogs, which serve as the final host. S. lupi induces the formation of fibro‐inflammatory, non‐neoplastic nodules in the esophagus of which 25% transform into neoplastic lesions. Eosinophilic infiltration is rarely found within nodules and only early and transient eosinophilia occurs experimentally. 3‐Bromotyrosine (3‐BrY) is a biomarker of eosinophilic activation. We hypothesized that serum 3‐BrY concentrations may be increased in dogs infected with S. lupi and that 3‐BrY may be useful to discriminate between benign and malignant nodules. The objective of this study were to compare serum 3‐BrY concentrations in dogs infected with S. lupi with those in healthy dogs and to compared serum 3‐BrY concentrations in dogs with benign and those with malignant nodules. Serum 3‐BrY concentrations were measured in 18 dogs infected with S. lupi (10 with a benign nodule and 8 with a malignant nodule) and 45 healthy control dogs using electron impact ionization gas chromatography/mass spectrometry. A Mann‐Whitney U test was used to compare serum 3‐BrY concentrations between dogs with S. lupi infection and healthy control dogs. A comparison between the dogs with spirocercosis based on malignancy of the nodule was also made using a Mann‐Whitney U test. The significance was set at P < 0.05. Serum 3‐BrY concentrations were significantly higher in dogs with S. lupi (median [range]: 3.7 μmol/L [≤0.6‐31.5]) than in healthy control dogs (median [range]: ≤0.6 μmol/L [≤0.6‐1.1]; P < 0.0001). However, serum 3‐BrY concentrations were not significantly different between dogs with benign nodules (median [range]: 3.6 μmol/L [≤0.6‐18.1]) and those with malignant nodules (median [range]: 3.7 μmol/L [≤0.6‐31.5]; P = 0.342). In conclusion, serum 3‐BrY concentrations were significantly higher in dogs with esophageal infestation with S. lupi. However, serum 3‐BrY concentrations were not different between dogs with benign or malignant esophageal nodules. These results suggest that eosinophilic involvement in canine spirocercosis might have been overlooked previously. A limitation of this study was the lack of data on the prevalence of other gastrointestinal parasites. Thus, further prospective studies are warranted to determine serum 3‐BrY concentrations in dogs with other gastrointestinal parasitic infestations with or without S. lupi infection. GI37 PROSPECTIVE EVALUATION OF A ONE‐STAGE ESOPHAGEAL BALLOON DILATION FEEDING TUBE FOR BENIGN ESOPHAGEAL STRICTURES IN DOGS: INITIAL RESULTS Chick Weisse 1, Allyson Berent1 1Animal Medical Center, New York, NY, USA Dysphagia due to benign esophageal strictures (BES) is an important cause of morbidity. Mechanical dilation may improve dysphagia, however repeated costly treatments are common and few animals regain normal function. The purpose of this study is to determine if the esophageal balloon‐dilation feeding tube (EBDFT) could provide a more effective, single‐procedure alternative for the treatment of BES. A balloon mounted on an esophageal feeding tube would be placed similar to an E‐tube, and remain in place for approximately one month. The owner performed twice daily balloon dilations at home. Dogs diagnosed with BES were included following client consent; exclusion criteria included comorbidities preventing general anesthesia or lack of follow‐up communication. Four female spayed dogs with confirmed BES were included in this preliminary study; each had dysphagia scores of 3/4 (only able to swallow liquids). Prior to EBDFT placement, these patients received between 1 and 10 previous esophageal balloon dilations. All patients were discharged the same or the following day. Post‐operative complications occurred in 2 dogs including premature tube removal by the dog (1), and prototype EBDFT failure requiring exchange (1). One dog was ultimately euthanized due to failure to improve and unrelated comorbidities. One dog still has the tube in place. The other two dogs have final dysphagia scores of 0/4 (normal eating) and 1/4 (able to swallow some kibble and canned food). EBDFT placement in dogs is feasible and warrants further investigation to determine if this technique can provide improved outcomes for BES. HM01 THE INDIVIDUAL AND COMBINED EFFECTS OF LONG‐CHAIN N‐3 POLYUNSATURATED FATTY ACIDS AND LOW‐DOSE ASPIRIN ON PLATELET FUNCTION IN HEALTHY DOGS Shannon Westgarth 1, Shauna Blois1, Adronie Verbrugghe1, David Ma1, Darren Wood1, Alice Defarges1, Sean Lang1 1Ontario Veterinary College, Guelph, Ontario, Canada Thromboembolic events are a common complication of many diseases in veterinary patients, and anti‐platelet therapies can be ineffective in their prevention. n‐3 polyunsaturated fatty acids (n‐3 PUFA) can inhibit platelet function in people, and enhance efficacy of anti‐platelet therapy. The objectives of this study were to evaluate the effects of low‐dose acetylsalicylic acid (ASA; 1 mg/kg/day) alone, n‐3 PUFA (100 mg/kg/day) alone, and combination ASA and n‐3 PUFA therapy on platelet function and activation markers in healthy dogs. Platelet function was measured using platelet function analyzer (PFA) closure time and whole blood aggregometry. Flow cytometry was used to measure platelet activation (P‐selectin) and platelet leukocyte aggregates. Above tests were evaluated at baseline, during therapy with ASA and n‐3 PUFA alone, and during combination therapy to assess effects on platelet function. This study demonstrated that n‐3 PUFA alone did not cause a significant change in platelet function tests. ASA alone decreased platelet function as measured by PFA‐epinephrine following one week of therapy (P < 0.0001). ASA plus n‐3 PUFA decreased platelet function significantly more versus ASA alone when measured by whole blood aggregometry (agonists: arachidonic acid P = 0.008; collagen P = 0.006). No change in platelet activation was detected via flow cytometry after any therapies. In conclusion, ASA plus n‐3 PUFA appear to have a synergistic effect on inhibition of platelet function in healthy dogs. This combination therapy may increase efficacy of anti‐platelet therapy in dogs at risk of thromboembolic complications. HM02 DOGS CAST NETS TOO: NEUTROPHIL EXTRACELLULAR TRAPS IN HEALTH AND IMMUNE‐MEDIATED HEMOLYTIC ANEMIA Unity Jeffery 1, Kayoko Kimura1, Bryan Bellaire1, James Roth1, Dana Levine1 1College of Veterinary Medicine, Iowa State University, Ames, IA, USA Neutrophil extracellular traps (NETs) are webs of DNA and proteins that protect against infection, but are pro‐thrombotic. Mortality from canine immune‐mediated hemolytic anemia (IMHA) largely results from thrombosis, and IMHA is associated with two triggers of NET formation: heme and hypoxia. Therefore, NETs may be prognostic markers or therapeutic targets in IMHA. However, NETs have not been previously demonstrated in dogs. Study aims were (1) to develop a method for in vitro generation of NETs by canine neutrophils and (2) to determine if dogs with primary IMHA have higher circulating nucleosomes (a marker correlated with NETs) than healthy dogs. Neutrophils were isolated from 5 healthy pet dogs and cultured with increasing concentrations of phorbol 12‐myristate 13‐acetate (PMA) or platelet‐activating factor (PAF). Extracellular DNA release was measured using Sytox green fluorescence. Confocal, immunofluorescent (IF) imaging was performed to visualize NETs using DAPI and Cy5 labeled anti‐elastase antibodies. Nucleosomes were measured in serum from 20 healthy dogs and 11 IMHA dogs using a commercial ELISA. Four IMHA dogs were excluded due to positive interferences from hemolysis and icterus. Cells stimulated with PAF of ≥31 μM or PMA of ≥0.1 μM released significantly more DNA than unstimulated cells (P < 0.05). DNA release was maximal by 1 hour for PAF and 2 hours for PMA. IF imaging confirmed NET formation. IMHA dogs had significantly higher nucleosomes than controls (median absorbance [arbitrary units] 0.90 and 0.12 respectively, P = 0.01). In conclusion, we show that canine neutrophils make NETs and provide preliminary evidence that NETs are elevated in IMHA. HM03 EFFECTS OF LEUKOREDUCTION ON COAGULATION FACTORS IN CANINE FRESH FROZEN PLASMA Michelle Foote 1, Marjory Brooks2, Todd Archer1, Andrew Mackin1, John Thomason1 1Mississippi State University, Mississippi State, MS, USA, 2Cornell University, Ithaca, NY, USA Leukoreduction is performed in the processing of packed cells and fresh frozen plasma (FFP) for human transfusion to decrease the risk of immune and febrile reactions. However, leukoreduction has been shown to reduce the coagulation factor content of FFP. Since FFP is the preferred transfusion product for dogs with coagulopathies, we investigated the effects of leukoreduction on canine FFP. In a cross‐over study, whole blood units were collected on two separate occasions from 8 healthy dogs, and half the units were leukocyte‐ and platelet‐depleted using leukoreduction filters (Terumo). The leukoreduced (LR) and non‐leukoreduced (NLR) units were then centrifuged to prepare FFP for comparison of the following hemostatic parameters: prothrombin time (PT), partial thromboplastin time (PTT), fibrinogen, coagulation factors V, VII, VIII, X, and XI, antithrombin, and protein C. The assays measured functional, procoagulant or anticoagulant factor activities using pooled normal canine plasma as a reference standard. Compared with NLR FFP, leukoreduction resulted in a relative prolongation of PT and PTT by 9.5% and 2.4%, respectively. We also found that LR FFP had a relative decrease in activities of Factor V (by 6.6%), Factor VIII (4.5%), Factor X (26.3%), and an increase in activities of Factors VII (4.9%) and Factor XI (9.1%). Our study revealed some variation in the coagulation factor content of LR FFP, however these changes were considered unlikely to influence the efficacy of canine FFP for factor replacement. Transfusion trials of LR FFP are warranted to assess its benefit in reducing adverse transfusion reactions in dogs with coagulopathies. HM04 PRODUCTION AND CHARACTERIZATION OF ANTI‐DAL ANTIBODIES FOLLOWING SENSITIZATION OF A DAL‐NEGATIVE DOG Stéphanie Goulet 1, Marie‐Claude Blais1 1Faculté de médecine vétérinaire, Université de Montréal, St.‐Hyacinthe, Canada The Dal blood type is a high frequency antigen associated with the production of anti‐Dal antibodies, which may result in ineffective transfusions or in hemolytic transfusion reactions. As for most DEA, testing for Dal remains dependant on the availability of polyclonal anti‐Dal antibodies obtained following sensitization of a Dal‐negative dog. In the literature, little is known on the time‐course of antibodies post‐transfusion, which led to the present study. The purpose of this study was to produce and characterize anti‐Dal antibodies following a standard blood transfusion, notably by determining its rate of appearance, agglutination titer over time and immunoglobulin class. A secondary objective was to produce polyclonal anti‐Dal antibodies to increase the availability of Dal blood typing. Following Dal blood typing of 57 adult beagles, one Dal‐negative individual was identified. This index dog was healthy based on physical exam, hematology and biochemistry profiles. The index dog and potential blood donors were extensively blood typed for DEA 1.1, 1.X, 3, 4, 5, 7 and Dal using tube and gel column techniques, as well as Alvedia DEA 1 lab test. To avoid fluid overload, 15 ml/kg of blood was first collected from the index dog following mild sedation (butorphanol 0,2 mg/kg). A transfusion of Dal+ packed red blood cells was administered over 4 hours (80 ml; 10 ml/kg). During and following the transfusion (1 week), the index dog was closely monitored, including for PCV/TS, bilirubinemia, hematuria and hemoglobinuria. Blood samples were collected post‐transfusion at days 1, 2, 3, 4, 6, 7, 14, 21, 28 (6 ml/sample), and then once a month (100 ml) for one year. Several Coomb's tests were performed (day 1, 6 and 7; week 10). As soon as anti‐Dal antibodies were detected, agglutination titers, which is define as the highest serum dilution where a positive agglutination reaction is observed, were performed in duplicate using two‐fold dilutions of fresh serum (gel column technique). The immunoglobulin type was determined at week 12 following sulfhydryl treatments. Agglutination titers were also performed on frozen serum aliquots over time to evaluate the stability of the polyclonal anti‐Dal antibodies through freezing. Finally, 65 dogs were blood typed using the polyclonal anti‐Dal antibody produced by this index dog as well as previously documented anti‐Dal reagent. The index dog was blood typed DEA 1 +  and 4 +  (Dal‐), while the blood donor was DEA 4 +  and Dal+. The transfusion was uneventful with no clinical and/or laboratory signs of hemolytic transfusion reaction observed. The PCV/TS remained stable over the time, except for an expected increased PCV 2 hours post‐transfusion (55%), which returned to its initial value 24 hours later (38%). All Coombs' tests were negative. A weak positive anti‐Dal agglutination reaction was first observed at day 4 (titer of 1:1). The agglutination titer progressed thereafter, reaching a maximum of 1:64 at week 8, followed by a gradual decrease one year post‐transfusion to 1:8. The anti‐Dal antibody appeared to be of the IgG class. Anti‐Dal antibodies produced prior to day 28 were no longer detected following a 6‐months freezing period. However, freezing did not cause any decrease in agglutination titer of anti‐Dal antibodies collected on day 28 and after, i.e. the titer level was thereafter useful for Dal‐typing purposes. All 65 dogs blood typed for Dal had concordant results using both polyclonal anti‐Dal antibodies (25 Dal‐ and 40 Dal+). In this study, a Dal‐negative dog was sensitized to Dal via a first blood transfusion. No signs of acute or delayed hemolytic transfusion reaction were observed. Anti‐Dal antibodies, which are likely IgG, were detected as early as 4 days post‐transfusion, with the highest agglutination titer reached at week 8. The anti‐Dal titers decreased over time, but remained detectable one year post‐transfusion. HM05 EFFECTS OF LEUKOREDUCTION AND STORAGE ON PHOSPHATIDYLSERINE EXPRESSION ON CANINE PACKED RED BLOOD CELLS Samantha Muro 1, John Stokes1, Todd Archer1, Andrew Mackin1, John Thomason1 1Mississippi State University, Mississippi State, MS, USA Storage of blood products creates an unnatural environment that can accelerate red blood cell (RBC) degradation. One well recognized indicator of RBC degradation is translocation of phosphatidylserine to the outer cell membrane layer. In units of stored human blood, an increase in phosphatidylserine expression is associated with a subsequent reduction in the circulating survival time of transfused RBCs, and an increased incidence of transfusion‐related complications. With the addition of a leukoreduction step which extracts leukocytes and platelets from blood products prior to storage, these removed cells are unable to influence the degradation of the RBCs within the unit. In dogs, it is unknown how the leukoreduction and subsequent storage of blood products influences the expression of phosphatidylserine on RBCs. In this cross‐over study, units of fresh whole blood were collected from 8 healthy dogs on two separate occasions. Half were then leukoreduced using a leukoreduction filter prior to centrifugation to create units of packed RBCs. All units were refrigerated and stored for 10 or 21 days. Before and after storage, blood was collected from each unit and RBC phosphatidylserine expression was quantitated via flow cytometry. Samples were collected both after removal from refrigeration and again after five hours at room temperature to simulate transfusion conditions. In the non‐leukoreduced units, median RBC expression of phosphatidylserine decreased by 6% and increased by 2% on Days 10 and 21 of storage, respectively. In the leukoreduced units, phosphatidylserine expression decreased by 5% and increased by 3.5% on Days 10 and 21 of storage, respectively. After being held at room temperature for 5 hours, phosphatidylserine expression in the non‐leukoreduced units decreased by 2% on Day 10 and 6% on Day 21. In the leukoreduced units, there was a 3% and 1% decrease on Days 10 and Day 21, respectively. Our study suggests that, in canine packed RBCs, phosphatidylserine expression is not significantly affected by leukoreduction, storage, or warming for transfusion. HM06 PRE‐OPERATIVE EVALUATION OF HEMOSTATIC BIOMARKERS IN STABLE DOGS UNDERGOING SPLENECTOMY FOR SPLENIC MASSES AM Lynch 1, TE O'Toole1, CRL Webster1, E McCobb1, AM de Laforcade1 1Cummings School of Veterinary Medicine at Tufts University, North Grafton, MA, USA Portal vein thrombosis (PVT) is a rare but potentially fatal complication of splenectomy in dogs. The mechanism behind postoperative PVT development is unclear but may include alterations of portal blood flow following surgery, acquired hypercoagulability and endothelial dysfunction. The aim of the study was to evaluate hemostatic biomarkers in hemodynamically stable (heart rate <130 beats/minute, blood lactate < 2.5 mMol/L) and non‐anemic (hematocrit >35%) dogs prior to splenectomy for splenic masses. We hypothesized that these dogs would have no pre‐operative coagulation derangements present to risk postoperative thrombosis. Pre‐operatively abdominal ultrasonography was performed and blood was collected for platelet enumeration, prothrombin time (PT), activated partial thromboplastin time (aPTT), kaolin‐activated thromboelastography (TEG), fibrinogen, von Willebrand factor activity (vWF:Ag), antithrombin and thrombin‐antithrombin complex (TAT). Histopathological diagnosis and 30‐day survival were recorded. None of the 15 enrolled dogs had pre‐operative sonographic evidence of PVT. Three of fifteen dogs were thrombocytopenic, three had thrombocytosis, three were hyperfibrinogenemic, one had low vWF:Ag, three had mild prolongations of PT and none had abnormal aPTT. Based on the TEG G value, 13/15 dogs were hypercoagulable (mean±SD 13.5 ± 5.4kd/s). Antithrombin deficiency was identified in 9/15 dogs (mean ± sd 68.7 ± 22.7%) with 5/9 having concurrently elevated TAT suggesting active thrombin generation. There was no significant difference in hemostatic biomarkers or outcome for the 8/15 dogs with malignancy compared to those with benign lesions (p > 0.05). No dogs developed PVT and all achieved 30‐day survival. Pre‐operative hypercoagulability was recognized commonly and may in part contribute to the risk of postoperative PVT in dogs after splenectomy. HM07 ACTIVATED CLOTTING TIME IN HEALTHY SEDATED CATS USING THE MAX‐ACT SYSTEM Anthony Abrams‐Ogg1, Kimberly Ho1, Shauna Blois1, Karol Mathews1, Marie Holowaychuk 1, Alexa Bersenas1, Darren Wood1 1University of Guelph, Guelph, Ontario, Canada Activated clotting time (ACT) in cats, based on visual first clot at 37°C, using a tube no longer available, was reported as mean 99.2 sec, median 95 sec, and 95% CI 55‐165 sec, with no effect of sedation (1 Am J Vet Res 2000;61:750‐753). Using the currently available MAX‐ACT tube, visual end clot formation at 37°C was reported as mean 66 sec, and range 55–85 sec (2 Aust Vet J 2009; 87: 292‐295). In the present study, MAX‐ACT was measured in 49 healthy client‐owned cats with normal hematology and hemostasis. Cats were sedated with ketamine 2 mg/kg and butorphanol 0.2 mg/kg, given via a saphenous vein. Using jugular venipuncture with a 21ga butterfly needle, after blood draw into vacuum tubes for hematology, blood was collected by syringe. Blood draw quality was graded from 1(easy) to 3 (difficult) 1. 0.5 mL was dispensed into a MAX‐ACT tube, warmed by axillary incubation (35.4‐36.6 °C), and after 60 sec visually inspected q10sec for first clot formation, and then q5sec until end clot formation. Another 0.5 mL was synchronously dispensed into a second MAX‐ACT tube, and incubated at 38.4°C in an instrument that mechanically detects end clot formation (Actalyke MINI II). Blood collection quality was scored as 1 in 35 cats, and as 2 in 7 cats. Actalyke values were not obtained in 5 cats because of insufficient quantity. ACT medians, geometric means, and variances did not differ for quality 1 and 2, and these values were combined to describe normal ranges (ie. not 95% CI, Table 1). Visual end clot and Actalyke values were correlated (rho=0.82, P < 0.001), but geometric means (113 and 117sec, respectively) and variances differed (P = 0.002), with 29 Actalyke values > Visual end clot values. Factor XII activity was measured in all cats with MAX‐ACT>120 sec and was normal. Blood collection quality was scored as 3 for 7 cats. The Actalyke failed, ending incubation in <60 sec in all 7 cats. Visual and Actalyke values for these cats were considered outliers and removed from analysis. Visual values for 5 of these 7 cats were below normal range, and for 2 cats were above normal range. For 36 other healthy facility cats, MAX‐ACT was measured on days 1, 4 and 8 (Table 1). All collections were scored as 1 or 2; all values were within normal ranges. Overall medians, means and variances did not differ between days, although individual values did vary. For visual clotting, 33% of paired observations were identical, 66% differed by ≤ 5 sec and 88% by ≤10 sec. For Actalyke, 48% of paired observations differed ≤ 5 sec and 79% differed by ≤10 sec. Intraclass correlation coefficient was 35%. In this study MAX‐ACT was similar to the first report of ACT in cats1 and repeatability was high within a range ± 10 sec. Visual clotting was as repeatable as Actalyke. Cat age (yrs) Median(range) Visual 1st Clot (sec) Median(range), n Visual End Clot (sec) Median(range), n Actalyke (sec) Median(range), n Normal 4 (1–14) 100 (85–150), 42 110 (90–160), 42 110 (93–163), 37 Day 1 6 (3–10) 95 (85–130), 33 100 (90–140), 33 110 (95–141), 19 Day 4 6 (3–10) 95 (85–125), 33 105 (90–130), 33 109 (93–140), 28 Day 8 6 (3–10) 95 (80–130), 35 100 (90–140), 35 108 (93–143), 30 HM08 CYTOPENIAS ASSOCIATED WITH PROCARBAZINE THERAPY IN 8 DOGS A. Courtney Crane 1, Amanda Abelson1, Alex Lynch1, Elizabeth Rozanski1, Laura Harvey1, Dominik Faissler1 1Tufts University, North Grafton, MA, USA Procabarbazine is an alkylating agent used in the treatment of granulomatous meningoencephalitis(GME). Side effects of procarbazine include gastrointestinal and hematological side effects. The purpose of this study was to describe procarbazine associated cytopenias in dogs. Medical records were retrospectively reviewed. Data collected included signalment, dosage and duration of procarbazine therapy, clinicopathological findings, and outcome. Eight dogs (five castrated male and three spayed female) were included in the final analysis. Seven dogs received procarbazine for GME and one for an inflammatory myositis. All dogs were small breeds [Pug (3), Chihuahua (1), Havanese (1), Corgi (1), Fox Terrier (1), and Rat Terrier (1)]. The dosage of procarbazine ranged from 38‐50 mg/m2 twice daily to every other day. The median duration of procarbazine therapy was 28 weeks (range 4‐136 weeks) before onset of clinical signs, that included fever and lethargy. The median white cell count at the time of diagnosis was 560 (range 300‐5800/ul), median platelet count was 5000 (range 1000‐156,000/ul), and median hematocrit was 19 (range 7‐27%). Procarbazine was discontinued in all dogs and antibiotics were initiated in each case. Additional treatments (e.g. blood transfusions) were at the discretion of the primary clinician. Seven of the eight dogs did not survive to discharge, while one patient was discharged but was lost to follow‐up. Adverse hematologic effects associated with procarbazine toxicity were severe and commonly fatal in these dogs. It is unclear if bone marrow recovery is possible in these dogs and the time required to achieve bone marrow recovery is not known. HM09 EVALUATION OF THE ENTERPRISE POINT‐OF‐CARE BLOOD GAS ANALYZER Lyndsay R. Kong 1, Elisabeth C.R. Snead2, Krystina M. Musil2, Hilary J. Burgess2, Tasha Y. Epp2 1Atlantic Veterinary College, Charlottetown, PE, Canada, 2Western College of Veterinary Medicine, Saskatoon, SK, Canada Blood gas analyzers are commonly used to assess respiratory function and metabolic derangements in small animal patients. Point‐of‐care analyzers are becoming increasingly popular as a way to provide fast patient‐side results in emergency situations. The Enterprise Point‐of Care (EPOC) blood gas analyzer is newly available, but has yet to be validated in canine patients. The objective of this study was to compare the analytical performance of the EPOC analyzer to the bench top RapidLab® 1265 blood gas analyzer. Arterial blood gas samples were prospectively collected from 76 healthy dogs of variable age and analyzed using the EPOC and the RapidLab® 1265 analyzer. Concordance correlation and Bland‐Altman plots were used to assess agreement and bias between the two analyzers. Concordance correlation (CCC) showed excellent agreement for only three analytes: lactate, PaO2 and PaCO2 (CCC) 0.98, 0.91 and 0.82 respectively). For all other analytes (pH, hematocrit, sodium, potassium, ionized calcium, and glucose), agreement varied from slight to substantial (CCC < 0.80). Glucose was the only analyte on Bland‐Altman plots to have measurement bias between the two analyzers. Sodium and haematocrit had wide variance that could have potential clinical significance. The EPOC analyzer was easy to use, reliable, and portable. Although the CCC showed unsatisfactory agreement for most analytes, the majority had acceptable clinical variance on Bland‐Altman plots. In conclusion, the EPOC analyzer is suitable for use in small animal patients for all analytes (except sodium and hematocrit). Further studies are warranted evaluating the EPOC in non‐healthy animals. HM10 OSMOTIC FRAGILITY TEST – A USEFUL DIAGNOSTIC TEST IN CATS WITH ANEMIA? Christiane Weingart 1, Nina Merten1, Roswitha Merle2, Barbara Kohn1 1Clinic of small animals, Berlin, Germany, 2Department of Biometry and Information Processing, Berlin, Germany The osmotic fragility (OF) of erythrocytes determines their fragility in hypotonic solutions. The goals of this study were to determine the mean OF (MOF) in cats with anemia and to evaluate the suitability of this test for diagnosing anemic patients. The study included 108 anemic cats with various forms of anemia. Anemia was categorized into the following groups: acute/chronic blood loss anemia (BA), hemolytic anemia (HA), anemia of inflammatory and neoplastic diseases (AID), intramedullary (INR) or extramedullary nonregenerative anemia (ENR). OF was performed by mixing red blood cells in a serial dilution of NaCl solution (0 – 0.85%). MOF was determined from the lysis curve as concentration of NaCl at which 50% of the erythrocytes were hemolysed. The test was assessed by three different people. Statistical analysis was done using SPSS (IBS, Version 22). The 115 cats used as controls exhibited an MOF between 0.39 – 0.62% (mean 0.51). There was a significant difference in measurements of MOF among the three examiners (Kruskal‐Wallis‐Test, p‐value <0.001). Therefore the 95%‐percentile was used as reference value (0.44 – 0.58%), i.e. MOF above 0.58% was regarded as increased, below 0.44% as decreased. Cats with HA (n = 38) exhibited the highest MOF (0.44 – 0.85%, median 0.72); two cats with chronic BA had the lowest MOF values (0.38 – 0.40%). Overall, each of the groups exhibited the following rates of increased MOF: HA 86%, INR 67%, ENR 62%, AID 46%, and acute BA 28% (n = 38, 9, 13, 28, and 18, respectively). MOF may be a useful tool in diagnosing HA and BA and helpful to differentiate nonregenerative immune‐mediated hemolytic anemia from other forms of nonregenerative anemia. Low MOF values can be an indicator of a chronic BA. HP01 MEASUREMENT OF PLASMA FIBRINOGEN IN DOGS WITH HEPATOBILIARY DISEASE Sara Wennogle 1, Allison Bradley1, Christine Olver1, David Twedt1 1Colorado State University, Fort Collins, CO, USA Hypofibrinogenemia has been described in humans with advanced liver disease and dogs with chronic hepatitis. Additionally, hypofibriongenemia has recently been correlated with peri‐operative bleeding during liver transplantation in humans. Plasma fibrinogen concentrations have not been compared among different categories of liver disease in dogs. The goal of this study was to retrospectively evaluate fibrinogen concentrations in dogs with various histologic types of liver disease. Fibrinogen was measured in stored citrated plasma from 41 dogs that underwent liver biopsy at Colorado State University from June 2013‐November 2014. The fibrinogen assay was performed on the Destiny AMAX Plus™ analyzer and results reported in mg/dL. The reported normal reference range for this assay is 117 to 392 mg/dl. Based on the histological diagnosis dogs were grouped into one of the following categories: non‐specific reactive hepatopathy/NSR (n = 13); chronic hepatitis/CH (n = 9); hydropic/vacuolar hepatopathy (n = 8); cholangiohepatitis (n = 4); neoplasia (n = 4); or other (n = 3). Groups were compared via one‐way ANOVA with Tukey adjusted pairwise comparisons and significance was defined as P < 0.05. Plasma fibrinogen summary data are shown in the figure below. Dogs with histologically confirmed CH had significantly lower mean fibrinogen concentrations than other histological categories: NSR (P = 0.004), hydropic/vacuolar (P < 0.0001), cholangiohepatitis (P = 0.0001), neoplasia (P = 0.0009) or other (P = 0.0328). Mean fibrinogen levels were not different among other histological types of liver disease. Hypofibrinogenemia is more likely to occur in dogs with CH than other types of hepatobiliary disease. Plasma fibrinogen concentrations may be a useful indirect indicator of hepatic function and hypofibrinogenemia may also predispose a patient to peri‐operative bleeding. Further studies are needed to better characterize hypofibrinogenemia in dogs with hepatobiliary disease. HP02 SERUM MARKERS OF VITAMIN B METABOLISM IN CATS WITH HEPATIC LIPIDOSIS Alexandra Hamilton 1, Rosana Lopes2, Jan Suchodolski2, Jörg Steiner2, Steve Hill1, Jonathan Lidbury2 1Veterinary Specialty Hospital of San Diego, San Diego, CA, USA, 2Gastrointestinal Laboratory, Texas A&M University, College Station, TX, USA The prevalence of cobalamin deficiency in cats with hepatic lipidosis (FHL) is unknown. Cobalamin serves as a cofactor in metabolic reactions responsible for the generation of carnitine, which promotes the oxidation of fatty acids, and the antioxidant glutathione. Cobalamin deficiency could therefore play a role in the pathogenesis of FHL. The aims of this study were to determine the prevalence of hypocobalaminemia in cats with FHL and to compare serum methylmalonic acid (MMA) and homocysteine (HCY) concentrations between healthy cats and cats with FHL. Sixteen cats with cytologically diagnosed FHL were divided into two groups: Group 1: 9 cats with no definitively diagnosed underlying disease (4−11 years; 6 FS; 3 MC) and Group 2: 7 cats with a definitively diagnosed underlying disease (3‐12 years; 4 FS; 3 MC). Serum folate, cobalamin, MMA, and HCY concentrations were measured at the time of diagnosis for all FHL cats as well as for a group of 20 healthy control cats (Group 3: 1−13 years; 9 FS; 11 MC). Serum folate and cobalamin concentrations were measured using an automated chemiluminescence assay. Serum MMA and HCY concentrations were measured using in‐house GC/MS assays. The prevalence of hypocobalaminemia (reference interval: 290−1,500 ng/L) and hypofolatemia (reference interval: 9.7−21.6 μg/L) were compared among the three groups using χ2 tests. Serum MMA and HCY concentrations were compared amongst the three groups using ANOVA or Kruskal‐Wallis tests, as appropriate. Statistical significance was set as P < 0.05. The prevalence of hypocobalaminemia was as follows: 6%, 13%, and 0% for Groups 1, 2, and 3, respectively (P = 0.06). The prevalence of hypofolatemia was as follows: 31%, 25%, and 10% for Groups 1, 2 and 3, respectively (P = 0.12). Serum cobalamin concentrations (mean ±SD) were lower in Group 2 (606 ± 147 ng/L) than in Group 3 (2,192 ± 1,381 ng/L; P < 0.05). Serum folate concentrations (mean ± sd) were lower in Group 1 (9.6 ± 6.2 μg/L) and Group 2 (11.3 ± 6.3 μg/L) than in Group 3 (19.4 ± 7.9 μg/L; P < 0.05). Serum MMA concentrations (median [minimum−maximum]) were higher in Group 1 (239 [145−1,873] nmol/L) and Group 2 (559 [400−2,718] nmol/L) than in Group 3 (147 [104−304] nmol/L; P < 0.05 and P < 0.01, respectively). Serum Hcy concentrations were not significantly different among the three groups (P = 0.13). The overall prevalence of hypocobalaminemia in the cats with FHL was relatively low (19%) and was not associated with the presence of a definitely diagnosed underlying disease. However, cats suffering from FHL, with and without a definitively diagnosed underlying disease, had higher serum MMA concentrations than healthy cats, possibly indicating that some of them had cobalamin deficiency at a cellular level. The overall prevalence of hypofolatemia in cats with FHL was higher (56%) and was not associated with the presence of a definitively diagnosed underlying disease. HP03 ASSOCIATION BETWEEN BILIARY CYTOLOGY AND MICROBIOLOGY IN DOGS AND CATS: 52 CASES (2004–2014) Medora Pashmakova 1, Whitney Nelson1, Julie Hilligas1, Micah Bishop2 1Texas A&M University, College Station, TX, USA, 2Animal Specialty Hospital of Florida, Naples, FL, USA Diagnostic cholecystocentesis is frequently performed for the evaluation of hepatobiliary disease in small animals. In our clinical experience, cytologic bactibilia is not always associated with a positive microbiologic culture and large studies describing cytologic findings of bile are lacking in the veterinary literature. The aim of this retrospective study was to evaluate the agreement between cytologic and microbiologic findings in the bile of dogs and cats with hepatobiliary disease. Electronic medical records at Texas A&M University were searched for dogs and cats that had a cholecystocentesis with concurrent bile cytology and bile culture performed between 2004 and 2014. Cohen's kappa coefficient (κ) was used to measure agreement between different diagnostic techniques. Fifty‐two patients met inclusion criteria. Bactibilia was cytologically identified in 17/52 samples (32.7%), while bacterial culture growth was identified in 11/52 samples (21.2%). One negative cytology resulted in positive bacterial growth. There was good agreement between bile cytology and culture growth in 84.6% of samples (κ  =  0.616; 95% CI: 0.383 ‐ 0.848). Enterococcus spp. was the most common isolate (7/11 samples) followed by Escherichia coli (5/11 samples). There was fair agreement between cytologic evidence of inflammation and cytologic evidence of bactibilia (κ  =  0.253; 95% CI:0.008 ‐ 0.497). Finally, there was poor agreement between cytologic evidence of inflammation and positive bacterial growth (κ  =  0.136; 95% CI: ‐0.160 ‐ 0.431). In this patient population, good agreement existed between the presence or absence of bacteria in bile cytology and positive or negative culture results, respectively. Evidence of inflammation alone was poorly associated with culture and bactibilia. Empiric antimicrobial therapy targeting Enterococcus spp. and E.coli isolates may be considered in clinically ill patients with cytologic bactibilia while confirmatory bacterial culture results are pending. HP04 HISTOLOGIC SCORING AND CLINICAL OUTCOMES OF HEPATITIS IN 42 STANDARD POODLES Allison Bradley 1, Tawfik Aboellail1, Allison Cotter2, David Twedt1 1Colorado State University, Fort Collins, CO, USA, 2VCA Emergency Animal Hospital and Referral Center, San Diego, CA, USA Standard Poodles have been suggested to suffer from a breed‐related chronic hepatitis, and the Poodle Club of America lists chronic hepatitis as a health issue. The purpose of this prospective study was to characterize the liver histology and clinical course of Standard Poodles suspected to have chronic hepatitis. Liver tissue from Standard Poodles suspected to have chronic hepatitis were solicited over a period of approximately three years. Patients receiving immunosuppressive therapy were excluded. Veterinarians and owners were contacted for long‐term outcome information. Liver samples were submitted for histologic diagnosis, and case histories were reviewed. Inflammation was grouped into the following categories: Acute Hepatitis, Chronic Hepatitis (including Copper‐associated Hepatitis and Cirrhosis), and Reactive Hepatitis. Lesions were also assigned a total score out of 30 points for 7 histologic features: lobular collapse; bile duct hyperplasia; interface hepatitis; focal lytic necrosis/apoptosis/random inflammation; portal inflammation; zonal necrosis; and fibrosis. Hepatic metal quantification (copper, zinc, and iron) was performed when adequate tissue was available, and in the remainder, copper and iron content was assessed semi‐quantitatively through histochemical staining. 42 eligible liver biopsies were received, 35 of which were available for quantitative histologic review, and the results are summarized below. Disease Category # of Cases Male / Female Age * (years) Histologic Score* (1‐30) % Deceased % with Gastrointestinal, Pancreatic, or Biliary Disease Acute Hepatitis 2 1/1 3.7 (3.3–4.1) 5 (5–5) 0 100 Chronic Hepatitis 20 12/8 7.9 (2.9–12.7) 10 (3–18) 35 35 Copper‐associated 1 0/1 7 (N/A) 9 (9–9) 0 0 Cirrhosis 3 2/1 9.5 (7.3–13.5) 20 (16–21) 100 0 Reactive Hepatitis 16 6/10 9.1 (3.9–12.8) 8 (2–20) 6 50 * median (range) Chronic hepatitis was the most common diagnosis (57% of submissions) in this group of Standard Poodles with clinicopathologic evidence of liver disease. In general, these patients responded well to immunomodulatory therapy; however, patients with a total histologic score greater than 18 had a poor prognosis. Nonspecific reactive hepatitis was also common (38% of submissions), and dogs in this group were frequently also affected by gastrointestinal or pancreaticobiliary disease.  In this group of Standard Poodles, both chronic and reactive hepatitis were frequently identified. Reactive hepatitis was non‐specific and often associated with concurrent extra‐hepatic disease. There were no consistent histologic features of chronic hepatitis that appear unique to the breed, and primary copper‐associated hepatitis was not the main etiology in this group of dogs. Whether chronic hepatitis is a breed‐specific condition in the Standard Poodle remains to be elucidated through further studies. HP05 POSSIBLE ASSOCIATION OF GALLBLADDER MUCOCELE HISTOLOGIC DIAGNOSIS WITH SELECTED PREVENTATIVE DRUG USE IN DOGS: A MATCHED CASE–CONTROLLED STUDY Jody Gookin 1, Maria Correa1, Allison Peters1, Amanda Malueg1, Kyle Mathews1, John Cullen1, Gabriela Seiler1 1North Carolina State University, Raleigh, North CarolinA, USA The cause of gallbladder mucocele formation in dogs is unknown. Drugs used for the prevention of flea and tick infestation, heartworm infection, and those commonly used for adjunctive treatment of degenerative joint disease, represent a newer‐generation of xenobiotics to which dogs are chronically exposed. We speculate that pharmacogenomic differences in drug metabolism may be a reason for gallbladder mucoceles to develop. In this study, we examined the medical records of dogs diagnosed at our institution with a gallbladder mucocele and matched controls (1 to 2 ratio) based on age, breed, and admission date. The aim was to determine if there was an association between gallbladder mucocele diagnosis and reported administration of preventative drugs. Seventy seven dogs with a histological diagnosis of gallbladder mucocele and 154 matched controls from a single referral institution were included in the study; selection of cases and controls was done through medical record accessions review for the 2001‐2011 period. Owners were contacted for verification of drug history through a standard mail questionnaire. Reported use and type of heartworm, flea, and tick preventatives; non‐steroidal anti‐inflammatory drugs, oral analgesics, corticosteroids, or medications used for adjunctive treatment of osteoarthritis, hyperadrenocorticism, hypothyroidism, or Diabetes mellitus were recorded. Data were described using basic statistics and further analyzed using conditional logistic regression to identify univariate associations between gallbladder mucocele diagnosis and the use of any of the above mentioned medications. Odds ratios and 95% Confidence Intervals for the odds ratios were used to identify variables for further analysis. Dogs diagnosed with gallbladder mucocele ranged in age from 4.0 to 15.2 years (median, 10.3 years). Shetland sheepdogs, Cocker spaniels, and Chihuahuas were significantly overrepresented relative to the general hospital population. Dogs with gallbladder mucocele were 2.6 times as likely to have a report of treatment with thyroxine (proxy for a diagnosis of hypothyroidism) as controls dogs (OR 2.549, 95% CI 1.124‐5.796, P = 0.0256); 2.6 times as likely to have a report of treatment for Cushing's disease (OR 2.605, 95% CI 0.988‐6.861, P = 0.0529); and, 2.079 times as likely to have reported the use of products containing imidacloprid (OR 2.079, 95% CI 0.985‐4.389, P = 0.0549). The lower side of the 95% CI for imidacloprid crosses 1 and results should be interpreted with caution. Based on unique pharmacogenomic differences in Shetland sheepdogs (e.g. ABCB1 gene mutation), we analyzed two data subsets considering all Shetland sheepdogs in one group (18 cases and 36 controls) and the other breeds in a different group (no multivariate analysis was possible due to sparse data). Results indicated that Shetland sheepdogs with a mucocele diagnosis were 8.2 times as likely to have reported the use of imidacloprid as controls dogs (OR 8.217, 95% CI 0.947‐71.267, P = 0.056), while for the other subset of dogs there was no indication of any association. The results of this study suggest a possible association between Shetland sheepdogs diagnosed with gallbladder mucocele formation and the use of products containing imidacloprid. Although the results should be interpreted with caution due to the large confidence interval and marginal p‐value, our results suggest further studies of imidacloprid in a larger population of Shetland sheepdogs may be warranted. HP06 A NEW COMBINATION OF BLOOD PARAMETERS FOR ACCURATE NON‐INVASIVE DIAGNOSIS OF LIVER FIBROSIS IN DOGS Alexis Lecoindre 1, Patrick Lecoindre2, Michelle Chevallier3, Aurelie Pagnon4, Marie Destro5, Veronique Miette5, Benoit Rannou1, Jean Luc Cadoré1, Marine Hugonnard1 1VetAgro Sup Veterinary Campus, Lyon, France, 2CVC veterinary clinic, Lyon, France, 3Biomnis, Lyon, France, 4Novotec, Lyon, France, 5Echosens, Paris, France The evaluation of liver fibrosis (LF) is of major importance for the management of chronic liver disease and the prediction of prognosis. Although liver biopsy is the gold standard for evaluation of fibrosis, noninvasive tests in human medicine enable the clinician to successfully stage and monitor a wide variety of liver diseases. In veterinary medicine, it would be desirable to have a simple, quick and non‐invasive way to early identify dogs with LF. The aim of this study was to develop a blood scoring system, which can help the clinicians to screen for dogs with LF and decide the need of further investigation. Forty seven adult dogs were prospectively enrolled based on persistent elevation of ALT, with exclusion of those with focal lesions on ultrasound examination. Four healthy dogs were included as controls after approval of the ethic committee. All dogs underwent liver biopsy and serum blood collection. LF was staged according to WSAVA criteria and amount of collagen was measured through morphometric analysis. A combination of several hepatic injury biomarkers and fibrogenic activity biomarkers gave a negative predictive value of moderate fibrosis between 90 and 100% meaning that this scoring system enabled to identify dogs without LF. It also distinguished dogs with significant fibrosis with an AUROC curve of 0.83 and a positive predictive value between 90 and 100%. This new blood scoring system might prove useful for the clinician to early classify patients with LF and pursue a liver biopsy as part of further investigation. HP07 THROMBOELASTOGRAPHY IN CATS WITH CHOLESTATIC LIVER DISEASE Gideon Daniel 1, Armelle De Laforcade1, Cynthia Webster1 1Cummings School of Veterinary Medicine at Tufts University, Grafton, MA, USA Despite correction of prothrombin time (PT) and activated partial thromboplastin time (PTT) with parenteral vitamin K and the presence of normal platelet counts, many cats with cholestatic hepatobiliary disease develop progressive anemia while hospitalized or after provocative procedures. We hypothesized that bleeding tendencies in these cats might be associated with decreased clot strength or fibrinolysis both which can be monitored with thromboelastography (TEG). Thus this study was conducted to compare TEG analysis in cats with cholestasis with conventional coagulation tests. Seventeen cats with serum hyperbilirubinemia > 3 times the upper limit of normal, serum alanine aminotransferase (ALT) activity > 2 times the upper limit of normal and a PCV greater than 28% were recruited. All cats were treated with 3 doses of vitamin K1 subcutaneously for 36 hrs prior to coagulation testing. The mean values for the TEG parameters (R, K, angle, MA), PT, aPTT and platelet count in cholestatic cats did not differ significantly from the reference range. The one exception was a significantly increased G (10.1 ± 5.6 dynes/s vs 6.9 ± 1.9 dynes/sec, P = 0.038). The TEG G value labeled 8/17, 6/17 and 3/17 cats as hypercoagulable, normocoagulable or hypocoagulable, respectively. Compared to normal cats, hypercoagulable cats had a significant decrease in aPTT (15.4s ± 2.1s vs 18.3s+/‐ 6.8s, P = 0.014) and K (1.23 ± 0.43s vs 1.63 ± 0.54s, P = .031) and an increase in angle (74.1 ± 5.60 vs 66.5 ± 8.70, P < 0.006). Compared to normal cats, hypocoagulable cats had significantly lower platelet counts (224 K/uL ± 42 K/uL vs 377 K/uL ± 195 K/uL, P = 0.02), prolonged K (4.1s ± 0.41s vs 1.62s ± 0.54s, P < 0.001) and decreases in angle (52.8 ± 110 vs 66.5 ± 8.60, P = 0.016). LY30 was increased in only one cat (32%, normal <20%)) that was hypocoagulable. In conclusion, by TEG analysis cholestatic cats replete with vitamin K rarely have hyperfibrinolysis, but have a variety of coagulation profiles with a tendency to be normo‐ or hypercoagulable. HP08 THROMBOELASTOGRAPHY IN DOGS WITH CHRONIC HEPATITIS Whitney Fry 1, Carrie Lester1, Nahv Etedali1, Armelle DeLaforcade1, Cynthia Webster1 1Cumming School of Veterinary Medicine at Tufts University, Grafton, MA, USA Due to the liver's role in synthesis of the majority of pro‐ and anti‐coagulants as well as most regulators of fibrinolysis, a complex rebalance of coagulation exists in dogs with chronic hepatitis (CH). On conventional plasma based coagulation testing thrombocytopenia, prolongations in prothrombin (PT) and activated partial thromboplastin time (aPTT) as well as decreases in fibrinogen have been reported in dogs with CH. Despite these alterations which should result in hypocoagulability, spontaneous bleeding is rare. In fact, CH is a risk factor for portal vein thrombosis presumably related to decreased production of anti‐coagulants or to endothelial dysfunction secondary to portal hypertension. The aim of the current study was to gain a better perspective on the state of coagulation in dogs with CH by performing thromboelastography (TEG) and comparing it with conventional coagulation testing. Seventeen dogs with biopsy proven CH and 3 dogs with clinical evidence of end stage liver disease were recruited. PT, aPTT, platelet count, PCV and kaolin activated TEG were performed according to published standards. In some dogs fibrinogen, anti‐thrombin activity (AT), protein C activity (PC), D‐dimers and vWF activity was performed. Eleven breeds with a mean age of 7.3 ± 3 yrs were enrolled. All dogs had PCV's in the normal range. When compared to reference ranges dogs with CH had longer R (5.4s ± 2s, P = 0.009) and K (3.64s ± 3.2s, P = 0.002) values, smaller angles (55.00 ± 14.50, P = 0.006) and an increase in LY30 (4.26% +/‐10.4%, P = 0.018). Fibrinogen (P = 0.03), AT activity (P = 0.0001) and PC activity (P = 0.021) were significantly decreased. TEG G value defined 8/20, 7/20 and 5/20 dogs as normocoagulable, hypercoagulable and hypocoagulable, respectively. G was positively correlated with PC activity (P = 0.688, P = 0.02) and negatively correlated with PT (‐0.459, P = 0.05) and aPTT (‐0.443, P = 0.05). Four dogs were hyperfibrinolytic with mean Ly30 of 20.9%, ± 15.2%. In conclusion, on TEG analysis dogs with CH have a variety of coagulation states. PT, aPTT and PC activity predict the state of coagulation as determined by TEG. Some dogs with CH are hyperfibrinolytic a condition that would be missed on conventional coagulation testing. HP09 THE USE OF HEPATIC NON‐INVASIVE MARKERS FOR MONITORING FIBROGENIC PROGRESSION IN 49 SCOTTISH TERRIERS WITH VACUOLAR HEPATOPATHY Alexis Lecoindre 1, Patrick Lecoindre2, Michelle Chevallier3, Hecham Azrak4, Marie Destro4, Veronique Miette4, Jean‐Luc Cadoré1 1VetAgro‐sup veterinary Campus, Lyon, France, 2CVC veterinary clinic, Lyon, France, 3Biomnis, Lyon, France, 4Echosens, Paris, France The clinical course of chronic liver diseases is significantly dependent on the progression rate of fibrosis. Liver biopsy is considered the gold standard for the staging of fibrosis. Scottish terrier (ST) have a breed‐specific syndrom associated with vacuolar hepatopathy (VH) characterized by different grades of hepatocyte vacuolation and fibrosis. The aim of this study was to evaluate the correlation between serum biomarkers and the liver fibrosis in ST and to validate the method for further use in other breeds. Forty nine ST were enrolled retrospectively on the basis of the presence of VH on histology. The biochemistry panel included ALP, ALT, AST, cholesterol, GLDH, LDH, bilirubin, basal cortisol, basal bile acids, GGT, total T4 and triglycerides. Liver biopsies were performed on the basis of ALP activity above 140U/L and the presence of diffuse hepatic parenchymal modifications on ultrasound examination. The fibrosis was evaluated using the METAVIR scoring system, in agreement with the WSAVA guidelines. Significant and severe fibrosis were predicted by logistic regression using 4 biological parameters. According to Metavir score, the prevalence of diagnostic targets in this selected population was: significant fibrosis: 40.1%, severe fibrosis: 25.7%. The AUROC curve of derived score probability was respectively 0.73 (Se = 0.80; Sp = 0.65) and 0.91 (Se = 0.89; Sp = 0.88) This study shows an interesting correlation between non‐invasive methods and liver fibrosis in ST. However no single marker is ideal to predict fibrosis and a combination of several markers might be helpful to offer a diagnostic test for liver fibrosis with acceptable sensitivity and specificity. IM01 THYMIDINE KINASE, C‐REACTIVE PROTEIN AND VITAMIN D CONCENTRATIONS IN DOGS WITH IMMUNE‐MEDIATED THROMBOCYTOPENIA AND IMMUNE‐MEDIATED HEMOLYTIC ANEMIA Megan Grobman, Hilton Outi, Hans Rindt, Carol Reinero Comparative Internal Medicine Laboratory and Department of Veterinary Medicine and Surgery, College of Veterinary Medicine, University of Missouri, Columbia, MO, USA Canine immune‐mediated thrombocytopenia (ITP) and immune‐mediated hemolytic anemia (IMHA) are associated with high morbidity and risk of mortality. Thymidine kinase (TK), an enzyme important for DNA synthesis and cell division, may correlate with uncontrolled replication or regeneration following destruction. Canine C‐reactive protein (c‐CRP) has been used as an inflammatory disease biomarker. 25‐Hydroxyvitamin D (25(OH)D) is involved in innate and adaptive immunity with hypovitaminosis D associated with increased incidence of autoimmune disease in humans. We hypothesized that plasma TK and c‐CRP would be significantly higher in dogs with poorly‐ versus well‐controlled ITP and IMHA and that these dogs would be vitamin D insufficient. Plasma was collected from client‐owned dogs (n = 24) presenting to the University of Missouri Veterinary Medical Teaching Hospital. Samples were collected for some dogs serially. Samples were collected at initial diagnosis (n = 8) and during recheck examinations (n = 26). Disease control was determined by hematocrit/packed cell volume and platelet counts. Plasma TK, c‐CRP and 25(OH)D were measured commercially (Veterinary Diagnostics Institute, CA). TK and c‐CRP were significantly increased in poorly‐ versus well‐controlled dogs (P = 0.038, P = 0.007, respectively). Poorly‐ vs. well‐controlled median concentrations for TK were 46.9 and 8.45 U/L (ref <2U/L) and for cCRP 29.8 and 1.1 mg/L (ref <2 mg/L). All but two dogs were vitamin D insufficient (defined as <100 ng/mL) with mean±SD all samples, 51 ± 27 ng/mL. TK and c‐CRP warrant further investigation as biomarkers for monitoring immune‐mediated disease control. Vitamin D insufficiency may suggest supplementation could improve therapeutic response or reduce disease risk, though additional study is needed. IM02 COMPARISON OF PROLIFERATIVE AND IMMUNOMODULATORY POTENTIAL OF ADIPOSE‐DERIVED MESENCHYMAL STEM CELLS FROM YOUNG AND OLD CATS Lara Barron, Jessica Quimby, Tracy Webb, Andrea Herndon, Polly Webb, Steve Dow Colorado State University, Fort Collins, CO, USA Adipose‐derived mesenchymal stem cells (aMSC) have recently been investigated as an immunomodulatory therapy in cats with chronic kidney disease and asthma. The influence of age on the proliferative and immunomodulatory potential of feline aMSC has not been explored, which has implications for autologous versus allogeneic therapy. The purpose of this study was to compare telomere length, ability to proliferate in culture and ability to suppress lymphocyte proliferation of aMSC generated from young and geriatric cats. Adipose tissues from six young (<5 years) and six geriatric (>10 years) cats were harvested and cryopreserved for subsequent aMSC isolation and culture. aMSC proliferation in culture was compared via determination of time until passage two (P2) and MTT proliferation assay. Immunomodulatory capacity was assessed using lymphocyte proliferation assays. Telomere length was measured using fluorescent in situ hybridization. All assays were performed on aMSC cells between P2 and P3. Adipose‐derived MSC from geriatric cats took significantly longer (P = 0.03) to reach P2 (median 11 days, range 9–22 days) compared to aMSC from young cats (median 7 days, range 6–14 days) but no difference in MTT proliferation assay performed at P3 was detected. No significant difference was found between the two groups in telomere length or ability to suppress lymphocyte proliferation. Compared to young feline aMSC, geriatric aMSC are impaired in their ability to rapidly proliferate to P2 following initial culture, presenting a concern for autologous therapy. However, replicative potential as represented by telomere length, and immunomodulatory potential were similar between the two age groups. IM03 ASSOCIATION OF A VARIANT ALLELE IN THE DOG LEUKOCYTE ANTIGEN CLASS I GENE DLA‐79 WITH MULTIPLE CANINE AUTOIMMUNE DISEASES Steven Friedenberg 1, Lhoucine Chdid1, Natasha Olby1, Julien Guillaumin2, Theresa O'Toole3, Kathryn Meurs1 1North Carolina State University College of Veterinary Medicine, Raleigh, NC, USA, 2The Ohio State University College of Veterinary Medicine, Columbus, OH, USA, 3Tufts Cummings School of Veterinary Medicine, North Grafton, MA, USA Autoimmune diseases are common and life‐threatening diseases in dogs. Many canine autoimmune diseases have strong breed predispositions and are believed to be inherited. However, the genetic mutations that cause or predict these diseases are mostly unknown. As many autoimmune diseases in humans share polymorphisms among a common set of genes, we conducted a candidate gene study to assess whether causative or predictive polymorphisms might exist in similar genes in dogs. We initially performed targeted resequencing of 14 genes in 16 dogs with immune‐mediated hemolytic anemia (IMHA). Based upon these findings, we sequenced a 2‐base‐pair allele in exon 2 of the DLA‐79 gene (DLA‐79*001:02) in 313 additional dogs. Of the 329 dogs evaluated across 55 breeds, 140 had autoimmune disease (45 IMHA, 44 atopic dermatitis, 36 immune‐mediated thrombocytopenia, 15 immune‐mediated polyarthritis (IMPA)) and 189 dogs served as controls. The frequency of the variant allele was significantly higher in dogs with autoimmune disease than in control dogs (0.23 vs. 0.08, P < 0.001), and ranged from 0.37 in dogs with IMPA to 0.15 in dogs with atopic dermatitis. The alternate allele causes two non‐synonymous substitutions resulting in p.Phe62Leu and p.Asn66Asp amino acid changes. These mutations occur in an antigen binding pocket of the homologous human protein. Variants in the DLA‐79 gene may be associated with multiple autoimmune diseases in dogs. Further studies are warranted to confirm these findings more broadly, and to determine the specific mechanism by which the identified variants may alter canine immune system function. IM04 TOWARDS THE DEVELOPMENT OF A FLOW CYTOMETRIC ASSAY FOR CANINE INTERLEUKIN‐10 Julia Wu1, Katharine Carney1, Yogesh Singh2, Bradley Cobb3, Oliver Garden 1 1Department of Clinical Sciences and Services, The Royal Veterinary College, University of London, London, UK, 2Institute of Physiology I, Eberhard‐Karls University of Tubingen, Tubingen, Germany, 3Department of Comparative Biomedical Sciences, The Royal Veterinary College, University of London, London, UK Type 1 regulatory T (Tr1) cells play a pivotal role in the maintenance of peripheral tolerance. In contrast to their extensive characterization in humans and mice, little is known about Tr1 cells in the dog, an important veterinary species and an animal model for various human diseases. Secretion of high‐level interleukin‐10 (IL‐10) is considered to be a defining hallmark of Tr1 cells. To date, a well‐validated assay for intracellular IL‐10 in the dog has not been described. Our goal was to develop a robust flow cytometric assay for canine intracellular IL‐10 as a precursor to the study of Tr1 cells in this species. We screened three commercially available antibodies (Abs): a monoclonal anti‐human IL‐10 Ab (clone JES3‐9D7) marketed for use in flow cytometry and Western blots, and both polyclonal (catalog number AF735) and monoclonal (clone 138128) anti‐canine IL‐10 Abs marketed only for use in Western blots. Cross‐reactivity of the anti‐human monoclonal Ab was predicted on the basis of 81% amino acid sequence identity of the canine and human proteins. Canine IL‐10 was cloned into two vectors, pcDNA3 and pMIG. Both constructs were used to transfect human embryonic kidney (HEK) 293T cells. Surprisingly, canine IL‐10 expressed by transfected cells failed to be detected by the anti‐human monoclonal Ab by either Western blots or flow cytometry. Both anti‐canine IL‐10 Abs detected canine IL‐10 expressed by transfected HEK 293T cells, the polyclonal Ab performing better in the setting of Western blots and the monoclonal Ab performing better in the setting of flow cytometry. In summary, two commercially available Abs have been validated for the detection of canine intracellular IL‐10. Studies examining the expression of this cytokine by both in vitro‐generated IL‐10–producing cells and the peripheral blood monoclonal cells of dogs with immune‐mediated and neoplastic diseases are ongoing. IM05 IN VITRO EVALUATION OF THE EFFECTS OF PRE‐CONDITIONING AND GENDER ON FELINE ADIPOSE‐DERIVED MESENCHYMAL STEM CELL CYTOKINE PRODUCTION Becca Timmons, Craig Webb, Tracy Webb Colorado State University, Fort Collins, CO, USA The use of mesenchymal stem cell (MSC) therapy to treat clinical disease may be optimized through selection and pre‐conditioning methods that would enhance their immunomodulatory functions. We investigated the effects of several pre‐conditioning methods and gender of the tissue donor on parameters of feline adipose‐derived MSC in vitro that may be applicable to and optimize the clinical effect of MSC application in this species. Passage 3 feline MSC were generated from cryopreserved subcutaneous adipose tissue from 6 healthy cats, three males and three females. MSC were subjected to six different pre‐conditioning agents or conditions for 24 hours: control, poly I:C, IFNγ, TGFβ, serum‐free, and hypoxia. Supernatants from the MSC were harvested after 24 hours, aliquoted, and frozen for batch analysis. Levels of IL‐6, IL‐8, IL‐10, and MCP‐1 in the supernatants were determined by ELISA. Both pretreatment and gender of donor significantly affected measured cytokine levels in the 24 hour supernatants from feline adipose‐derived MSC: IL‐6 and IL‐10 were affected by donor gender whereas all of the measured cytokines were affected by at least one of the pretreatments. Although further study is necessary, this initial study shows that specific pre‐conditioning methods hold promise for tailoring and augmenting feline adipose‐derived MSC immunomodulatory properties to treat specific diseases, and gender of donor cat may be significant in this species depending on the desired therapeutic outcome. ID01 CANINE VECTOR‐BORNE DISEASE PREVALENCE IN DOGS FROM THE SOUTHEASTERN UNITED STATES DIAGNOSED WITH IMMUNE‐MEDIATED HEMOLYTIC ANEMIA AND/OR IMMUNE‐MEDIATED THROMBOCYTOPENIA Barbara Qurollo1, Edward Breitschwerdt 1, Ramaswamy Chandrashekar2, Barbara Hegarty1, Melissa Beall2, Brendon Thatcher2, Brett Stillman2, Jiayou Liu2, Christian Leutenegger2 1Intracellular Pathogens Research Laboratory, Center for Comparative Medicine and Translational Research, College of Veterinary Medicine, North Carolina State University, Raleigh, NC, USA, 2IDEXX Laboratories, Inc., Westbrook, ME, USA Infectious pathogens linked to canine immune‐mediated hemolytic anemia (IMHA) and immune‐mediated thrombocytopenia (ITP) include viral, bacterial, protozoan or rickettsial organisms. The purpose of this study was to retrospectively screen for canine vector borne diseases (CVBD) using serology and PCR methods in IMHA/ITP cases from the Southeastern region of the United States. Archived serum from a population of dogs being tested for tick‐borne diseases (IFA for Ehrlichia canis, Rickettsia rickettsii, Babesia canis, Bartonella henselae, Bartonella vinsonii subsp berkhoffii, and SNAP® 4Dx® for E. canis, Anaplasma spp., Borrelia burgdorferi, and D. immitis) from 2008 to 2012 were further characterized by retrospective screening using SNAP® M‐A, an experimental, diagnostic ELISA assay that uses synthetic, species‐specific peptides to differentiate between exposure to E. canis, E. chaffeensis, E. ewingii, Anaplasma phagocytophilum, A. platys and Borrelia burgdorferi. Of the retrospectively screened canine sera, 56 dogs were identified with accessible clinical data, a diagnosis of IMHA and/or ITP, and corresponding stored EDTA‐anticoagulated whole blood for retrospective PCR testing. A total of 19/56 (34%) dogs were seroreactive for a CVBD pathogen. SNAP® M‐A identified an additional 7/56 (13%) dogs that were initially non‐seroreactive for all pathogens by IFA or SNAP® 4Dx®. The most prevalent antibody reactivity was to R. rickettsii antigens 6/56 (11%), followed by B. burgdorferi 5/56 (9%), Anaplasma spp. 5/56 (9%), E. ewingii 5/56 (9%), E. canis 4/56 (7%) E. chaffeensis 4/56 (7%) and B. henselae 1/56 (2%). Only 10/56 (17.8%) dogs were screened for vector‐borne pathogens by PCR at initial diagnosis; of those, 6/10 (60%) requested Babesia PCR only; 1/10 (10%) Ehrlichia PCR only; 2/10 Babesia and Ehrlichia PCR; and only 1/10 (10%) requested Babesia, Bartonella and Ehrlichia PCR. Based on PCR results done at the time of diagnosis, only 1/10 (10%) dogs was PCR positive for a vector‐borne pathogen (E. ewingii). Retrospective PCR testing, which included RT‐PCR assays to detect Anaplasma spp., Babesia spp., Bartonella spp., Canine Hemotropic Mycoplasma, Ehrlichia spp., Hepatozoon spp., Leishmania spp., Neorickettsia risticii and Rickettsia rickettsia, identified 5/56 (9%) PCR positive dogs, 1/5 (20%) Mycoplasma hemocanis and 4/5 (80%) E. ewingii. These results underscore the importance of a multimodal diagnostic approach to detect tick‐borne pathogens and support further evaluation of a potential role for E. ewingii and other CVBD in the pathogenesis of IMHA and ITP in dogs. ID02 EVALUATION OF THE FIP VIRUS REALPCR® TEST AND THE IMPACT OF SAMPLE SELECTION ON THE DETECTION OF THE VIRULENT STRAIN OF FELINE CORONAVIRUS IN FIP SUSPECT CATS Nancy Sanders 1, Christian Leutenegger2, Jane Robertson2, Marko Estrada2, Pauline Young2 1IDEXX Laboratories, Inc., Westbrook, ME, USA, 2IDEXX Laboratories, Inc., Sacramento, CA, USA Definitive diagnosis of feline infectious peritonitis (FIP) remains a challenge. Difficulties in diagnosis include lack of specific clinical signs, non‐specific laboratory abnormalities and low specificity of antibody titers. A real‐time PCR test to detect the specific virulent form of feline coronavirus has recently been developed by IDEXX Laboratories based on fusion peptide mutations in the spike gene. FIP Virus RealPCR® Test is reported as one of two biotypes: feline infectious peritonitis (FIPV) or feline enteric coronavirus (FECV). When biotyping is not possible on a specimen, it is reported as either below limit of detection or indeterminate, depending on the viral load. Fifty domestic cats initially suspected by the submitting veterinarian to have FIP were included in the study. Cats ranged in age from 2 to 195 months (2 months to 16 years). The majority of cats (41) were 2 years or younger; 6 cats were adult (3–11 years of age); and 3 were geriatric (14–16 years). Breeds represented were domestic shorthair (37), domestic longhair (5), Burmese (2), and one each of Bengal, Persian, American Shorthair, Ragamuffin, and unknown breed. The cases were graded by a boarded internal medicine specialist with a clinical scoring system ranging from 1 (confirmed not FIP) to 6 (confirmed FIP). Cases graded 1–3 were unlikely, and 4–6 were likely to have FIP. The clinical grading scale included specific information such as history, signalment, and all available laboratory diagnostic test results. Cats graded as 6 were confirmed to have FIP based on histopathology and immunohistochemical staining for coronavirus. Of the 50 cats initially suspected of having FIP, 3 were determined unlikely to have FIP based on the grading system (grade 3 or less). None of these cats obtained a FIPV biotype result. Three of 50 cats were unable to be graded given available data. The remaining 44 cats were graded likely to have FIP (grade 4 or higher). Of these 44 cats, 24 (54%) had an FIPV biotype result on one or more samples. None of the 44 cats had a FECV biotype result. Twenty of the cats had viral loads below the limit of detection. FIPV biotype positive results were obtained from 4/20 (20%) whole blood samples, 3/3 (100%) pericardial effusions, 5/6 (83%) pleural effusions, 11/23 (48%) peritoneal effusions, 1/1 (100%) aqueous humor sample, and 1/1 (100%) aspirate of an abdominal lymph node. For cats with one or more biopsies submitted, 18/18 cats (100%) had at least one biopsy sample positive for the FIPV biotype. Tissues with the highest rate of FIPV biotype were spleen, omentum and mesenteric lymph nodes. Other tissues positive for FIPV biotype included pancreas, bladder wall, eye, heart, liver, kidney, and lung. It is also noted that tissues with the most obvious gross pathology were more likely to provide a FIPV biotype result. The results of this study show that the FIP Virus RealPCR Test can be used to confirm the diagnosis of FIP in clinically suspect cats but that the sensitivity of the test appears to be highly dependent on the specimen submitted. Whole blood was the specimen least likely to provide a positive biotype result. Abdominal effusion specimens had a moderate sensitivity. Testing of pleural effusion specimens, tissue aspirates and biopsy samples (especially of spleen, omentum and mesenteric lymph nodes) provided the highest likelihood of a positive biotyping result. ID03 ANAPLASMA PHAGOCYTOPHILUM AND BORRELIA BURGDORFERI INFECTIONS IN CATS EXPOSED REPEATEDLY TO IXODES SCAPULARIS Michael Lappin 1, Ricarda Huesken2 1Colorado State University, Fort Colllins, CO, USA, 2Bayer Animal Health, Leverkusen, Germany Anaplasma phagocytophilum (AP) and Borrelia burgdorferi (BB) infections are vector‐borne and have been diagnosed in cats naturally exposed to Ixodes spp.. Laboratory diagnostic testing includes blood smears (AP), PCR (AP) and ELISA (AP and BB). In a recent study, wild‐caught I. scapularis (IS) were allowed to feed 7 days on 4 SPF cats and all four cats were tested positive for AP and BB using a commercially available kit for dogs (SNAP 4DX Plus, IDEXX Laboratories), AP morulae in neutrophils by blood smear, and AP DNA in blood (PCR). However, none of the cats developed clinical signs of disease suggesting that clinical signs of AP or BB would occur in cats only after repeated pathogen transmission through exposure to competent vectors. Two follow‐up experiments were designed to test the hypothesis of repeated exposure leading to clinical signs. In Experiment 1, 9 SPF cats were infested with 13 female and 12 male wild caught IS from Rhode Island that had approximately a 50% BB and a 15% AP infection rate, respectively. The IS were placed in chambers (approximately 7 × 10 cm) over a shaved area of the skin for 7 days to maximize potential for feeding. EDTA blood and serum samples were collected from cats prior to infestation and then every 2 weeks for 12 weeks. Each blood sample was assayed for AP DNA by PCR and each serum sample was evaluated for the presence of AP and BB antibodies by the SNAP 4DX Plus. A 6 mm skin biopsy punch was used to collect a full thickness biopsy from the shaved area from each cat 8 weeks after IS infestation for BB DNA amplification by PCR. In Experiment 2, the same nine cats were infested with 13 female and 12 male IS. The ticks were released on the fur on the flank of each sedated cat (dexmedetomidine) with counts of attached IS performed 2 days later. Blood and sera were assayed as described in Experiment 1 every 2 weeks for 12 weeks. In Experiment 1, all nine cats had attached and engorged IS and developed evidence of either BB infection alone (four cats), AP infection alone (three cat), or dual infection (one cat). In Experiment 2, all nine cats had also attached and engorged IS. A total of 4 of the 9 previously infected cats from Experiment 1 developed evidence of a new AP (two cats), repeated BB (two ats), or repeated AP infection (one cat). However, clinical signs of disease were not noted in both experiments. This study demonstrates that cats can be repeatedly infected with AP or BB without developing clinical signs of disease. Both tick infestation methods led to transmission of AP and BB by naturally infected IS. ID04 COMPARISON OF RESULTS OF 2 ASSAYS DETECTING DIROFILARIA IMMITIS ANTIGEN AND EHRLICHIA CANIS ANTIBODIES USING SAMPLES FROM HIGH AND LOW RISK REGIONS Michael Lappin 1, Rebecca Ruch Gallie1, Emily Bryan1, Camille Lehman2, Laurie Droke2, Scott Moroff3 1Colorado State University, Fort Collins, CO, USA, 2San Pedro Animal Hospital, San Pedro, Belize, 3Antech Diagnostics, Lake Success, NY, USA Dirofilaria immitis and Ehrlichia canis are important vector borne diseases in dogs. A number of assays are available to detect D. immitis antigen and E. canis antibodies in canine samples. Because different reagents and methods are used in different assays, results can be discordant. The purpose of this study was to compare the results of 2 commercially available assays using samples from dogs in low and high risk areas for the target organisms. Adult dogs (n = 20) that were screened for D. immitis antigen in San Pedro, Belize in August of 2014 were selected as those likely to be concurrently infested by Rhipicephalus sanguineous and so at high risk for E. canis exposure. Sera were also collected from dogs entering two shelters in northern Colorado from May 2014 to October 2014; this area is considered low risk for both E. canis and D. immitis. A commercially available kit (SNAP 4DX Plus; IDEXX Laboratories) was used to assess serum from dogs on site in Belize and from thawed serum in Colorado. The sera was then stored at – 20C until shipped on cold packs to a commercial laboratory (Antech Diagnostics; Lake Success, NY) for evaluation in the other assay (Accuplex 4). Agreement between the assays was determined using the kappa statistic. The results for the 2 assays for D. immitis antigen and Ehrlichia spp. antibodies using sera from the 2 regions are listed in the following Table. In the high risk area, agreement between the 2 assays was perfect for D. immitis and good for Ehrlichia spp. In the low risk area, agreement between the assays was worse than chance for Ehrlichia spp. and fair for D. immitis. When both regions were compared the agreement was good for both organisms. Prevalence Dirofilaria immitis Ehrlichia sp. Accuplex SNAP Kappa Accuplex SNAP Kappa High risk (n = 20) 7 (35%) 7 (35%) 1 10 (50%) 11 (55%) 0.9±0.1 Low risk (n = 98) 5 (5.1%) 1 (1%) 0.32±0.24 1 (1%) 4 (4.1%) −0.02±0.01 Overall (n = 118) 12 (10.2%) 8 (6.8%) 0.78±0.1 11 (9.3) 15 (12.7%) 0.78±0.1 In low risk area, the Accuplex 4 detected D. immitis antigen in 4 more dogs than the SNAP4DxPlus. Since necropsy was not available whether this finding represents the previously described increased sensitivity of the Accuplex 4 cannot be determined. In the low risk area, the SNAP4DxPlus detected Ehrlichia spp. antibodies in 4 more dogs than the Accuplex 4. The Accuplex 4 is E. canis specific whereas the SNAP4DXPlus detects antibodies against other organisms like E. ewingii and E. chaffeensis which may explain the discordant results. ID05 ATTEMPTED TRANSMISSION OF BARTONELLA HENSELAE TO CATS BY INGESTION OF INFECTED CTENOCEPHALIDES FELIS Michael Lappin, Melissa Brewer, Jennifer Hawley, Arianne Morris Colorado State University, Fort Collins, CO, USA Bartonella henselae is an important vector borne disease agent associated with clinical disease in many species including humans, cats, and dogs. Cats are a known definitive host and Ctenocephalides felis has been proven as a vector. Bartonella henselae can be grown from C. felis frass for at least 9 days. In a previous experiment, 1 of 4 research beagles was infected by B. henselae after nasal inoculation of infected C. felis frass. As cats frequently ingest C. felis and frass during grooming, the purpose of this study was assess the potential for oral B. henselae transmission after ingestion of infected C. felis. In this study, 4 purpose bred, 1 year old kittens from a barrier facility free of vectors were shown to be negative for FeLV antigen, FIV antibodies, Bartonella spp. antibodies, and Bartonella spp. DNA in blood by PCR assay. Two of the cats were infected with the CSU1 strain of B. henselae in a concurrent experiment. When used in the current experiment, the cats were shown to still be strongly positive for B. henselae DNA in blood using a previously published PCR assay targeting the 16S‐23S intergenic region. A total of 25 female and 25 male C. felis purchased from a commercial laboratory were placed into two separate flea chambers, one of which was placed on the flank of each of the two B. henselae naïve kittens and the C. felis allowed to feed for 5 days. At that time, the 10 of the C. felis from each chamber were removed from the chamber for assessment of B. henselae infection by PCR assay. The remainder of C. felis were cold stunned and were mixed together with the eggs and frass as a slurry with baby food and administered orally to one of each Bartonella spp. naïve cast. Blood and sera were then collected weekly for 12 weeks and assayed for Bartonella spp. antibodies and Bartonella spp. DNA by PCR assay. The aliquot of C. felis collected from the chamber of each B. henselae infected cat was strongly positive for B. henselae DNA. Over the course of the 12 week study, neither of the orally inoculated cats developed Bartonella spp. antibodies in serum or Bartonella spp. DNA in blood. The two cats orally inoculated with B. henselae infected C. felis and frass had no evidence of dental disease or stomatitis and did not develop evidence of infection over the time period studied with the defined assays. These results suggest that normal cats ingesting B. henselae infected C. felis are unlikely to become infected. However, further experiments should be performed to assess this route of transmission using more sensitive assays like BAPGM culture plus PCR assay and to use kittens that are teething or cats affected by oral cavity disease to provide a potential portal of entry for B. henselae. ID06 MULTIPLE DRUG RESISTANCE IN CANINE URINARY TRACT PATHOGENS Erin Burton, Leah Cohn, William Fales, Reilly Thomas, Irene Ganjam University of Missouri, Columbia, MO, USA Bacterial cystitis is common in dogs as both a primary disorder and as a complication of co‐morbid conditions. An increase in urinary isolates resistant to antimicrobial drugs has been reported in recent years. Multiple drug resistant (MDR) organisms, defined as organisms with acquired resistance to at least one member of 3 or more classes of antimicrobials, seem to be commonplace in the Veterinary Medical Diagnostic Laboratory at the University of Missouri. We undertook a descriptive retrospective study of drug resistance of urinary isolates from the MU‐VMDL from Jan.1, 2014 to Nov. 19, 2014. Urine cultures were performed using calibrated loops delivering urine to both blood agar and MacConkey agar plates for aerobic bacterial isolation. Plates were incubated under routine culture conditions. Antimicrobial susceptibility testing of resultant isolates was accomplished with the MIC technique using commercially prepared lyophilized trays. Breakpoints for determination of resistance or susceptibility were based on the standards set by the National Committee for Clinical Laboratory Standards. Antimicrobials included on routine susceptibility profiles differed slightly depending on organism (to exclude intrinsic resistance) but included aminoglycosides (amikacin and gentamicin), potentiated penicillin and carboxypenicillin (amoxicillin‐clavulanate, ticarcillin‐clavulanate), natural and synthetic penicillins and carboxypenicillin (ampicillin, penicillin, oxacillin +2% NACL, ticarcillin), 1st generation cephalosporins (cefazolin, cephalothin), 2nd generation cephalosporins (cefoxitin, ceftiofur), 3rd generation cephalosporins (cefovecin, cefpodoxime), chloramphenicol, clindamycin, doxycycline, erythromycin, fluoroquinolones (enrofloxacin, marbofloxacin), imipenem, rifampin, and trimethoprim sulfa (TMS). The 4 most common isolates (E. coli, Staphylococcus spp., Enterococcus spp., and Streptococcus spp) were evaluated. Innate antimicrobial resistance was defined for each of the four microbes of interest as follows: E. coli – clindamycin, erythromycin, penicillin; Staphylococcus spp. – aminoglycosides; Streptococcus spp. – aminoglycosides; Enterococcus spp. – aminoglycosides, all generations of cephalosporins, and TMS. A total of 786 canine urine samples were submitted for C/S, resulting in 418 bacterial isolates. Of 139 E. coli isolates, 7 were resistant to a member of 1 antimicrobial group, 94 to two groups, 8 to three groups, and 30 to >3 groups. Resistance was most likely to synthetic penicillins (122), rifampin (120) and flouroquinolones (34). Of the 70 Staphylococcus spp. isolates, 10 had resistance to 1 group, 5 to 2 groups, 7 to 3 groups, and 35 to >3 groups. Staphylococcus spp were most likely to be resistant to members of the antimicrobial groups penicillins (50), clindamycin (31) and erythromycin (30). Of the 67 Enterococcus spp. isolates, 10 had resistance to 2 groups, 26 to 3 groups, and 26 to >3 groups. Enterococcus spp were most likely to be resistant to penicillin (61), clindamycin (60) and rifampin (31). Of the 40 Streptococcus spp. isolates, 9 were resistant to members of 1 antimicrobial group, 3 to 2 groups, 3 to 3 groups, and 1 to >3 groups. Streptococcus spp were most likely to be resistant to doxycycline (7) or fluoroquinolone (6), with 3 isolates each resistant to penicillin, 2nd and 3rd gen. cephalosporins, erythromycin, and TMS. As expected, E. coli were isolated most often and 27.3% were MDR. MDR isolates were far more common among Staphylococcal isolates, (60.8%) and Enterococcus spp. isolates (85.2%). Fewer Streptococcus spp. were isolated, and they were least likely to be MDR (10.5%). While we cannot say if MDR isolates are more common now than in the past, we can state that these problematic bacteria are very commonly encountered in canine urine samples submitted to our VMDL. ID07 SPLENITIS IN DOGS: A REVIEW OF 33 CASES (2005–2013) Filippo Ferri 1, Silvia Ferro2, Edoardo Auriemma1, Massimo Castagnaro2, Luigi Michele Coppola3, Andrea Peano4, Vito Martella5, Nicola Decaro5, Eric Zini6 1Istituto Veterinario di Novara, Granozzo con Monticello, Novara, Italy, 2Department of Comparative Biomedicine and Food Sciences, University of Padova, Legnaro, Padova, Italy, 3Department of Animal Medicine, Production and Health, University of Padova, Legnaro, Padova, Italy, 4Department of Veterinary Science, University of Turin, Grugliasco, Turin, Italy, 5Department of Veterinary Medicine, University of Bari, Valenzano, Bari, Italy 6Vetsuisse Faculty, University of Zurich, Zurich, Switzerland In dogs splenitis is rarely described. The few reported cases have been associated to splenic abscesses or pseudotumor, and systemic mycosis. The aims of this study were to define the prevalence of splenitis in dogs, describe clinical findings and outcome, characterize histological patterns of splenic inflammation and investigate possible causes. Formalin‐fixed, paraffin‐embedded splenic samples of dogs with a diagnosis of splenitis were gathered from the pathology archive, between 2005–2013. New slides were obtained and stained with hematoxylin‐eosin, Gram, green‐Gram, Giemsa, periodic acid‐Schiff (PAS), Ziehl‐Neelsen. Slides were examined by two pathologists. Furthermore, an aliquot of each specimen was processed for PCR analysis to detect bacterial, fungal or protozoal DNA. Dog owners were contacted to achieve information pertaining clinical findings and outcome. Of 660 available splenic samples, 33 (5%) had a histological diagnosis of splenitis. Clinical findings and outcome were available in 19 dogs (57.6%). Presenting complaints were weakness in 16 dogs (48.5%), fever in 11 (33.3%) and abdominal effusion in 7 (21.2%). Sixteen (48.5%) of them survived to discharge following splenectomy, with or without additional abdominal surgery; the remaining three dogs died between 7 and 40 days from admission. Splenic histological findings in the 33 dogs included neutrophilic (36.4%), granulomatous (33.3%), haemorrhagic (18.3%) or lymphocytic (6%) splenitis and splenic abscess (6%). Based on clinical records or necropsy findings, 21 dogs (63.6%) had concurrent diseases along with splenitis, including 4 each with gastric dilation‐volvulus or neoplasia, 2 each with granulomatous hepatitis or pyometra, and 1 each with hepatic lobar torsion or systemic mycosis. In seven dogs (21.2%) bacteria were identified in the spleen based on histology (four dogs with Gram+ cocci, and 1 each with Gram‐ cocci or Gram+ cocci and bacilli) or bacterial culture (one dog with Pseudomonas aeruginosa). In three dogs (9.1%) protozoa were identified by PCR (three dogs with Leishmania infantum) or histology (1 of the Leishmania‐positive dogs had concurrent Hepatozoon canis) and in one dog (3%) ifae were detected with histology. PCR did not identify bacteria or fungi in any of the spleen, including those with microscopic evidence of infection. Associations between histological findings and causes were not observed. The present study shows that splenitis is an uncommon disorder in dogs. Splenitis is frequently secondary to inflammation of other organs or to systemic infections, although in numerous dogs it may not be associated with other obvious diseases. Prognosis is favourable in the majority of affected dogs. Several histological patterns of inflammation are identified in dogs with splenitis but whether they reflect different causes deserves further studies. PCR may not be a sensitive tool to detect bacteria and fungi DNA in formalin‐fixed, paraffin‐embedded splenic samples. ID08 CYNICLOMYCES GUTTULATUS FROM DOMESTIC DOGS AND LABORATORY RABBITS HAVE DISTINCT GENOTYPES Jenessa A. Winston, Adam J. Birkenheuer, Stephen Stauffer, Jody L. Gookin North Carolina State University College of Veterinary Medicine, Raleigh, NC, USA Cyniclomyces guttulatus is a commensal yeast of the rodent and rabbit gastrointestinal tract and is infrequently described in dog fecal samples. The environmental source of C. guttulatus in dogs is unclear, however it is frequently presumed that dogs diagnosed with C. guttulatus have ingested rabbit feces. This study aimed to conduct a comparative sequence analysis of a fragment of the large subunit (26S) rDNA variable D1/D2 domain of C. guttulatus from dogs, pet rabbits, and laboratory rabbits to elucidate if different genotypes are affecting each species. Voided fecal samples from 7 domestic dogs and 10 rabbits (7 New Zealand White laboratory rabbits and 3 client owned domestic pet rabbits) were obtained. Fecal DNA was extracted and standard PCR amplification of a 180 base pair (bp) fragment of 26S rDNA using custom primers was performed. Amplicons were submitted for bi‐directional DNA sequencing. Nucleotide sequence alignments were made with existing GenBank C. guttulatus sequences and sequence identity matrix analyses were conducted. Partial 26S rDNA sequences were obtained from all seven dogs and 10 rabbits. Only high quality sequences, excluding the primers, were used for analysis. For dogs, a trimmed 131 bp region of the PCR product was compared across all seven dogs and two additional C. guttulatus dog sequences deposited in GenBank. Polymorphisms among all dog sequences ranged from 0 to 6 nucleotide substitutions resulting in a range of similarity from 95.4 to 100%. For rabbits, a trimmed 66 bp region of the PCR product was compared. The seven New Zealand White laboratory rabbit sequences shared 98.4–100% sequence identity and were similar to a previously reported Brazilian New Zealand White rabbit genotype deposited in Genbank. The 3 pet rabbits and an additional pet rabbit sequence in Genbank shared a homology of only 96.9% compared to laboratory rabbits. Alignment of all data for partial C. guttulatus 26S rDNA sequences obtained from the seven dogs and 10 rabbits examined in this study, in addition to sequences in Genbank, demonstrated 9 nucleotide positions wherein polymorphism between dog and rabbit C. guttulatus genotypes was identified. Among these, 4 nucleotide positions revealed a conserved difference between dog genotypes and laboratory rabbit genotypes. Pet rabbit genotypes were largely identical to that observed in dogs. The pet rabbits included in the study did not all share a history of exposure to dogs nor did the dogs in the study share a history of exposure to pet rabbits. Results of these studies demonstrate that partial 26S rDNA genotypes of C. guttulatus amplified from fecal DNA of dogs are similar to those observed in domestic pet rabbits but are genetically distinct to sequences amplified from fecal DNA of laboratory rabbits. As such these studies identify a significant genetic diversity of C. guttulatus between different populations of rabbits and suggest the possibility of transmission of C. guttulatus genotypes between dogs and domestic pet rabbits. ID09 COMPARISON OF 2 COMMERCIALLY AVAILABLE PCR ASSAYS FOR THE AMPLIFICATION OF EHRLICHIA SPP. DNA FROM BLOOD OF NATURALLY EXPOSED DOGS IN OKLAHOMA Madeline Anna 1, Jennifer Hawley1, Melissa Brewer1, Scott Moroff2, Michael Lappin1 1Colorado State University, Fort Collins, CO, USA, 2Antech Diagnostics, Lake Success, NY, USA Ehrlichia canis, E. chaffeensis, and E. ewingii are vector borne pathogens of dogs and people. Amplification of specific DNA is commonly used in the diagnosis of Ehrlichia spp., PCR assays for these agents are commonly available commercially, and positive PCR assay results often occur prior to detection of specific antibodies against Ehrlichia spp.. The purpose of this study was to compare the results of two different commercially available (USA) PCR assays for the amplification of Ehrlichia spp. DNA using samples from naturally exposed dogs in an endemic area. Dogs (n = 72) being sterilized at a low cost clinic in Checotah, OK during July 2014 were selected as those likely to be exposed to Ambloymma americanum and so at high risk for E. ewingii and E. chaffeensis exposure. Rhipicephalus sanguineus also is found in the region and so E. canis infection was also possible. A total of 1 mL of blood in EDTA collection tubes was stored in a refrigerator overnight, and the next day were sent on cold packs to the first testing laboratory via overnight shipping. On arrival, the samples were split into 2 aliquots with 1 aliquot being frozen at ‐80C and 1 aliquot being prepared for evaluation in a commercially available conventional Ehrlichia spp. PCR assay (www.dlab.colostate.edu). The frozen EDTA blood was shipped to the second PCR laboratory on dry ice for evaluation in a commercially available real time multiplex Ehrlichia spp. PCR assay (www.antech.com). Results from samples with sequence confirmation were compared using the kappa statistic. In the first laboratory, DNA of E. canis (one dog), E. ewingii (six dogs), and E. chaffeensis (one dog) were amplified from some dogs (11.1%). In the second laboratory, DNA of E. ewingii was amplified from nine dogs (12.5%). The agreement between the assays at the genus level was good (kappa = 0.667 [SE 0.139]; 95% CI = 0.395–0.938) with an overall percentage agreement of 93.1%. The dog with DNA most homologous with E. chaffeensis in the first laboratory showed DNA most homologous with E. ewingii in the second laboratory. Ehrlichia ewingii infection transmitted by A. americanum is common in Oklahoma. The agreement between the 2 laboratories was good even though some discordant results were detected with the second laboratory appearing to have a higher sensitivity for E. ewingii DNA. However, 100% concordance was not expected because the samples were not handled identically and the assays use different technologies. The sequence homology between E. ewingii and E. chaffeensis is very high and so a different result from two different sequencing laboratories can occur. Alternately, dog may have had dual infection. ID10 INFECTION OF BABESIA CONRADAE IN HUNTING GREYHOUNDS FROM OKLAHOMA Jennifer E. Thomas 1, Ramaswamy Chandrashekar2, Christian Leutnegger2, Gene Yost3, Mason V. Reichard1 1Center for Veterinary Health Sciences, Oklahoma State University, Stillwater, OK, USA, 2Idexx Laboratories, Inc., Westbrook, ME, USA, 3Crescent Veterinary Hospital Inc., Crescent, OK, USA Babesia conradae is a small, intraerythrocytic piroplasm found in dogs from southern California. In domestic dogs, B. conradae infection causes lethargy and hemolytic anemia. In March 2014 a greyhound used to hunt coyotes in Oklahoma presented with lethargy and pale mucous membranes. Evaluation of stained blood film revealed small piroplasms in red blood cells. Subsequent DNA extraction of whole blood, PCR amplification using piroplasm‐specific primers, and DNA sequencing confirmed infection by B. conradae. The index case dog was euthanized by the owner due to deteriorating condition. Follow‐up evaluation of the majority of dogs in the owner's kennel revealed that 3 of 6 (50%) greyhounds and 0 of 4 (0%) walker hounds were PCR positive for infection with B. conradae. No dogs were known to have travel histories west of Oklahoma. Two of the greyhounds subclinically infected with B. conradae were treated with an 8 day course of azithromycin (10 mg/kg [4.5 mg/lb], PO, q 24 hour) plus atovaquone (13.3 mg/kg [6 mg/lb], PO, q 8 hour). The third B. conradae‐infected dog was sold and lost to follow‐up. Approximately 1 month after completion of azithromycin and atovaquone treatment, both dogs were PCR negative for infection with B. conradae. To the authors' knowledge, this is the first report of B. conradae in dogs outside of California. ID11 CROSS‐SPECIES TRANSMISSION OF CANINE PARVOVIRUS 2 TO HEALTHY SHELTER KITTENS Christian Leutenegger 1, Hongwei Liu2, Niels Pedersen2 1IDEXX Laboratories, Inc., West Sacramento, CA, USA, 2Center for Companion Animal Health, University of California, Davis, Davis, CA, USA A recent study of shelter cats in the UK provided strong evidence that healthy cats may shed strains of canine parvovirus type 2 (CPV2). The purpose of this study was to confirm these findings among healthy cats in shelters and in privately owned sick cats in the USA. We developed and validated a CPV2 real‐time PCR assay to discriminate CPV2a, CPV2b, and CPV2c from feline panleukopenia virus (FPV). The specificity of the typing assays was confirmed by sequencing. The typing assays were then used to test CPV2 strains obtained from puppies with naturally acquired parvovirus infection and high fecal viral loads. All samples could be typed successfully and confirmed again by sequencing. After confirming the specificity of the strain typing assays in puppies, the same assays were used to test feces from healthy shelter kittens. In the group of the healthy shelter cats 58% shed parvovirus in their feces. Low levels of parvovirus shedding can be explained by chronic infection, gastrointestinal passage of environmental parvovirus or presence of vaccine virus. But 21% of these were shedding high parvovirus load indicating infection. Within the group of high shedders, 6% of kittens tested positive for canine parvovirus strain 2a. The importance of cross‐species transmission of CPV2 in shelters and multi‐animal environments has yet to be determined. However, the implication of this study that healthy cats may be a source of CPV2 infection for susceptible dogs, suggests that appropriate husbandry practices be adopted to reduce this risk. In addition, a current investigation is now applying the typing assays to cats with panleukopenia to determine the importance of CPV2 strains as a cause of infectious panleukopenia. ID12 SEROLOGIC AND MOLECULAR SURVEILLANCE TO PREDICT PREVALENCE OF CANINE VECTOR‐BORNE DISEASES ON THE ISLAND OF ST. KITTS Ramaswamy Chandrashekar1, Jennifer Ketzis2, Linda Shell2, Phyllis Tyrrell1, Melissa Beall1, Christian Leuteneggar1, Diana Scorpio 2 1IDEXX Laboratories, Westbrook, ME, USA, 2Ross University, Basseterre, Saint Kitts and Nevis The island of St. Kitts in the West Indies has a large feral and semi‐feral dog population. Semi‐feral dogs present a unique opportunity for study since they are mostly outdoors but are comparatively better taken care of compared to feral dogs. Care tends to range from minimal to adequate, though many owners use the Ministry of Agriculture Veterinary Services for their preventative medicine rather than a veterinary hospital setting. Historically, dogs on the island are endemically infected and infested with internal and ectoparasites, which bring with them numerous vector‐bone diseases. Due to the high prevalence of the tick Rhipicephalus sanguineus, it is not unusual to see prevalence of Ehrlichia canis immunity up towards 40–50%. Recently, a group of veterinarians and students from Ross University School of Veterinary Medicine conducted an island‐wide “field” study of semi‐feral dogs, which was more advantageous than feral dogs when gathering pet information during interactions with owners. Over a period of 3 months, veterinarians and students visited all parishes on the island of St. Kitts, thus covering a diverse geographic area with a wide diversity in ecosystem landscape. One hundred and eleven dogs were identified and visited, and were given a complete physical exam, then blood and feces were collected. Demographic data was also collected, including breed, age, housing, other animals in household, evidence of ticks or fleas, deworming and vaccine administration, and other important data. The primary focus of our sample analysis for this study was blood, which was processed immediately and separated into whole blood, serum, and plasma. Ticks and fleas were also collected when seen on dogs or in the immediate environment. Blood was analyzed with the SNAP® 4Dx test (IDEXX), followed by the Tick/Vector Comprehensive RealPCR® Panel detection of 19 different vector‐borne pathogens from the genera Anaplasma, Ehrlichia, Babesia, Hepatozoon, Leishmania, Bartonella, Neorickettsia, Rickettsia, and Mycoplasma. The data collected during this study has revealed that a great majority of dogs are not only exposed but infected with many vector‐borne organisms, including E. canis, Anaplasma platys, Babesia spp (mostly B. gibsonii), and Mycoplasma hemocanis. Many dogs were co‐infected with more than 1 pathogen. In most cases, dogs did not exhibit any clinical signs except for pale mucous membranes, low body weight, and icterus (this one dog was positive for four pathogens). Dogs are also heavily infested with Ancylostoma (hookworm) and therefore clinical evidence of anemia was frequently confounded with so many red cell‐destroying pathogens potentially acting in a synergistic manner with each other. The most significant finding from this study is the high rate of infection with M. hemocanis (17%) as diagnosed by qPCR. Mycoplasma infection is not a common differential in island dogs due to the well‐known high level exposures (ELISA) and infections (qPCR) with E. canis (34% and 11%, respectively). The importance of Mycoplasma infections will require further study; however, the data implicates the importance of including an exhaustive list of differentials when suspecting a vector‐borne infection. Island geography, ecosystems, tick burdens, and dog demographics may aid veterinarians in the diagnosis of clinical disease with these pathogens. ID13 SURVEILLANCE OF FERAL CATS IN CENTRAL IOWA FOR SUBCLINICAL INFECTION WITH CYTAUXZOON FELIS J. E. Slovak 1, Y. Nagamori2, J. E. Thomas2, M. V. Reichard2 1Washington State University, Pullman, WA, USA, 2Oklahoma State University, Stillwater, OK, USA Cytauxzoonosis is a deadly tick‐borne disease of domestic and wild felids. Cytauxzoon felis is transmitted by Amblyomma americanum and Dermacentor variabilis. Bobcats (Felis rufus), the wild reservoir host, largely remain clinically unaffected. Although C. felis infection of domestic cats can be fatal, documented subclinical C. felis‐infections have been found and may provide an additional source of infection to naïve cats. Many factors, such as habitat for bobcats, exposure of outdoor domestic felids, presence of the tick vector, and lack of tick control contribute to C. felis infections. The traditional distribution of C. felis has been the south‐central and southeastern United States. However, C. felis has recently been detected in bobcats and domestic cats in non‐enzootic areas. Iowa is considered a non‐enzootic state, although there has been one known confirmed case. Its proximity to known enzootic states (Missouri, Arkansas, Kansas), a growing bobcat population, and the presence of the tick vector, put cats in Iowa at risk of becoming widely infected. Whole blood samples (n = 292) were obtained from cats that presented to the Iowa State University feral cat trap‐neuter‐release program from August 2012–April 2014. Cats were of variable health status. Caretakers of the cats signed a waiver for blood (1 mL) to be drawn for minimal laboratory and infectious disease screening, including PCR for C. felis. All samples were negative for C. felis. Since C. felis is an emerging infectious disease, continued surveillance is prudent. As the range of A. americanum continues to expand, veterinarians in Iowa and other states that border C. felis enzootic areas must remain vigilant for prompt diagnosis and treatment of patients. ID14 A NOVEL PORE‐FORMING TOXIN IN TYPE A CLOSTRIDIUM PERFRINGENS ASSOCIATED WITH FATAL CANINE HEMORRHAGIC GASTROENTERITIS AND NEONATAL FOAL NECROTIZING ENTEROCOLITIS Iman Mehdizadeh, Valeria Parreira, John Prescott, Vivian Nicholson, Kaitlyn Oliphant University of Guelph, Guelph, ON, Canada A role for type A Clostridium perfringens in acute hemorrhagic and necrotizing gastroenteritis in dogs and in necrotizing enterocolitis in foals has long been suspected but not confirmed and characterized. The supernatant of isolates made from a dog and from a foal that died from these diseases was highly cytotoxic for an equine ovarian (EO) cell line. Partial genome sequencing of the dog isolate revealed three novel putative toxin genes related to the pore‐forming Leukocidin/HemolysinSuperfamily. These were designated netE, netF and netG. Both netE and netF were located on one large conjugative plasmid, and netG was located with the cpe enterotoxin gene on a second large conjugative plasmid. netE and netF were found in all EO‐cytotoxic isolates, as was cpe, but netG was less consistently present. Mutation and complementation of the three genes showed that only netF was associated with the cytotoxicity for EO cells. Equine antisera prepared against recombinant Net proteins showed that only antiserum to NetF neutralized the cytotoxicity of supernatants of these strains. Although netE and netG were not associated with cytotoxicity, immunoblotting with specific antisera prepared against the proteins showed NetE and NetG to be expressed in vitro. There was a highly significant association between the presence of netF with type A strains isolated from cases of canine hemorrhagic gastroenteritis compared to type A isolates from dogs with undifferentiated diarrheal illness (P < 0.0000001) and from neonatal foals with necrotizing enteritis compared to foals with undifferentiated enterocolitis (P < 0.0002). Pulsed‐field gel electrophoresis (PFGE) showed that netF‐positive isolates belonged to a clonal population, and that some canine and equine isolates were genetically indistinguishable. The identification of NetF as a novel toxin strongly associated with canine hemorrhagic gastroenteritis and with necrotizing enteritis in neonatal foals opens the way to improving diagnosis, treatment and control of these important and serious diseases. ID15 EFFECT OF PARENTERAL ADMINISTRATION OF MODIFIED LIVE OR INACTIVATED FELINE HERPESVIRUS 1, CALICIVIRUS, AND PANLEUKOPENIA VACCINES ON CLINICAL SIGNS IN A DAY 7 FELINE HERPESVIRUS 1 CHALLENGE MODEL Stacie Summers, Rebecca Ruch‐Gallie, Jennifer Hawley, Michael Lappin Colorado State University, Fort Collins, CO, USA A previous study showed that >50% of cats housed in an animal shelter acquired feline herpesvirus‐1 (FHV‐1) infection within 7 days of entry. Thus, use of vaccines to induce rapid protection against FHV‐1 is indicated. In a previous study, cats administered one SQ dose of an inactivated FHV‐1, feline calicivirus (FCV), and panleukopenia virus (FPV) vaccine (FVRCP) developed protective antibody titers against FHV‐1 significantly faster than cats administered a modified live FVRCP vaccine (MLV). However, a FHV‐1 challenge was not performed. The objective of this study was to determine the effect of one dose of an inactivated FVRCP vaccine or one dose of a MLV FVRCP vaccine on the clinical signs of FHV‐1 in SPF kittens challenged with FHV‐1 7 days after vaccination. A total of 24 male and female 5‐month‐old purpose bred kittens from a FHV‐1 naïve barrier facility that were shown by PCR assay to be negative for nucleic acids of Mycoplasma spp, FHV‐1, and FCV from materials collected from the oropharynx and seronegative for FHV‐1 antibodies were purchased and randomly divided into three groups of eight kittens (four males, four females). Group 1 kittens were maintained as unvaccinated controls, Group 2 kittens were administered 1 dose of the inactivated FVRCP vaccine SQ on Day 0 (Ultra™ Fel‐O‐Vax®, Boehringer‐Ingelheim Vetmedica, Inc.), and Group 3 kittens were administered 1 dose of the MLV FVRCP vaccine SQ on Day 0 (PureVax; Merial LTD). The kittens were housed and handled separately to avoid cross‐contamination of the MLV vaccine virus among groups for the duration of the study. A standardized clinical score system was applied to each Group of kittens by 2 masked observers for 30 minutes each day of the study. After the clinical scores were collected on Day 7, all 24 kittens were administered 105.3 TCID50 of a USDA challenge strain of FHV‐1 (0.1 mL per nostril and 0.3 mL into the oropharynx). Group clinical scores were compared among study periods using Kruskal Wallis with pair‐wise comparison and significance defined as P < 0.05. In the 21 days (Days 8–28) after FHV‐1 challenge, both groups of vaccinated cats were less likely to be clinically ill (indicated by lower cumulative clinical scores) than control cats (P < 0.001). There was no statistical difference in total clinical score between the two vaccinated groups (P = 0.97). Parenteral administration of either an inactivated or modified live FVRCP vaccine can decrease clinical signs of illness due to FHV‐1 on a Day 7 challenge when compared to controls. Use of either vaccine product is indicated in cats at risk of acute exposure to FHV‐1. ID16 EFECT OF TREATMENT ON BORRELIA BURGDORFERI AND ANAPLASMA PHAGOCYTOPHILUM TEST RESULTS Ramaswamy Chandrashekar 1, Melissa Beall1, Brendon Thatcher1, Jiayou Liu1, Brett Stillman1, Phyllis Tyrrell1, Michael Lappin2 1IDEXX Laboratories, Westbrook, ME, USA, 2Colorado State University, Fort Collins, CO, USA Anaplasma phagocytophilum (AP) and Borrelia burgdorferi (BB) share Ixodes spp. ticks as vectors and are both associated with clinical illness in some infected dogs and cats. The purpose of this study was to evaluate canine antibody responses to a genus and species‐specific AP peptide (APH‐1 and APH4, respectively) and to BB C6 peptide before and after doxycycline treatment. A total of eight young adult, mixed sex, laboratory‐reared research dogs were used with IACUC approval. The dogs were shown to be negative for antibodies against AP and BB using a commercially available kit (SNAP® 4Dx®, IDEXX Laboratories). Dogs were infected by exposure to wild‐caught Ixodes scapularis for 1 week. Blood was collected from all dogs prior to tick attachment and then on weeks 1–11 and weeks 31–43 after tick attachment. Blood was assayed for AP DNA by PCR. Sera was separated from clotted blood and assessed for AP and BB antibodies by a multi‐analyte, research SNAP (SNAP MA), AP IFA, AP peptide quantitative ELISAs, C6 quantitative ELISA (Quant C6, IDEXX Laboratories) and the commercial kit. On week 31, doxycycline was administered at 10 m/kg, PO, daily for 28 days. Paired t‐tests were used to assess changes in optical density on SNAP MA as an approximation of antibody levels. AP DNA was amplified from blood of all dogs and all had AP antibodies in serum by multi‐analyte SNAP® research ELISA. Antibodies to AP were detected prior to those detected by the commercially available kit for 6 of 8 dogs. Antibodies to BB were detected only in 6 of 8 dogs and were first detected between 3–6 weeks post‐tick attachment, with 4/6 seropositive by day 28. All dogs were seropositive for AP from 31 to 43 weeks by SNAP MA, AP IFA, APH‐1 and APH‐4 ELISA with modest declines in seroreactivity post‐treatment. In the five dogs that were BB seropositive on week 31, C6 antibody levels declined (by week 43) from 63 to 92% with a mean decline of 81% relative to pre‐treatment values. Optical density readings for AP and BB on SNAP MA demonstrated significant decreases between 31–43 weeks however the magnitude of the difference was significantly greater for BB than AP (P < 0.05). The results suggest that antibodies against the APH‐4 peptide may be useful for early detection of AP infection. Doxycycline administration using this protocol decreases C6 antibody levels to a greater extent than antibodies measured by AP peptide ELISA. ID17 PREVALENCE OF DIROFILARIA IMMITIS IN NON‐ENDEMIC NORTHERN COLORADO ANIMAL SHELTERS Emily Kirkpatrick, Rebecca Ruch‐Gallie, Michael Lappin Colorado State University, Fort Collins, CO, USA Dirofilaria immitis is endemic to certain areas of the United States and causes a detrimental disease (heartworm disease) that is relatively costly to treat both financially and with regards to animal welfare. Though the Colorado climate is not conducive to year‐round replication of mosquito species that carry the organism, microclimates exist wherever humans inhabit thus potentially allowing year‐round transmission of D. immitis and propagation of strains from dogs entering the Colorado shelter system. Animal shelters on the northern Colorado Front Range frequently transport dogs from other states including those with a high risk of D. immitis infection. The hypothesis for this study is that shelter dogs have a higher prevalence than the general pet dog population. The purpose of this study was to determine the prevalence of D. immitis antigen in the serum of dogs entering animal shelters in Northern Colorado. Samples were collected from dogs entering three shelters in northern Colorado from May to October 2014. The study area was restricted by the foothills of the Rocky Mountains to the west and the high plains to the east (Larimer and Weld counties). Only dogs >6 months of age were included in this study. Blood or sera from each dog were evaluated for D. immitis antigen by one of three differently commercially available kits (Heska SoloStep®, IDEXX SNAP® 4Dx® Plus, Abaxis VetScan®) following the manufacturer's instructions. Positive results for D. immitis antigen were confirmed with a second test differing from the original test used, though the confirmatory test varied among samples. A total of five samples from the 340 shelter dogs were positive (1.5%) for D. immitis antigen. Three of the positive dogs came from areas considered low‐risk (Northern Colorado and South Dakota). The other two positive dogs came from Arkansas, a known high‐risk state. The D. immitis prevalence rate of 1.5% in the shelter dogs described here is three times greater than the currently reported prevalence rate in Colorado dog populations and 2–7 times greater than the rate reported for the participating counties (capcvet.org). Results of this study indicate D. immitis testing should be considered as a receiving procedure in northern Colorado animal shelters for both in‐area and out‐of‐area intakes. Results also emphasize the importance of the use of D. immitis preventives in the general population. ID18 PREVALENCE OF NOVEL HELICOBACTER STRAINS IN JAPAN Sanae Kubota, Koichi Ohno, Kenjiro Fukushima, Hideyuki Kanemoto, Yuko Goto‐Koshino, Ko Nakashima, Kazuyuki Uchida, Tsutomu Sekizaki, Hajime Tsujimoto The University of Tokyo, Bunkyo‐ku, Tokyo, Japan Epidemiological and pathological studies were performed on the Helicobacter spp. in canine stomachs in Japan to investigate strain‐specific pathogenicity. Endoscopic gastric biopsies from 144 dogs were evaluated for the presence of Helicobacter spp. by using genus‐ and species‐specific (H. felis [HF], H. bizzozeronii [HB], H. heilmannii sensus stricto [HH s.s.], and H. pylori [HP]) PCR. To investigate unclassified‐Helicobacter spp., partial urease and HSP60 genes were amplified with U430F/U1735 and H60F/H60R primers, respectively, and the sequences were compared with those in the NCBI/GenBank. All Helicobacter spp. with valid published names were included for phylogenetic analysis by using CLUSTAL X2, and phylogenetic trees were constructed by using the neighbor‐joining method. The PCR results indicated that 51 of 144 dogs were infected with Helicobacter spp.: HH s.s. (19), HB (13), HF (2), HP (1), and unclassified‐Helicobacter spp. (22). The urease and HSP60 gene sequences of the unclassified‐Helicobacter spp. showed respective similarities of 88–100%/89–99% with each other and 90–92%/86–89% with the HH s.s. ASB1 strain, which was the most closely related species. Sequencing of the partial urease and HSP60 genes demonstrated that the unclassified‐Helicobacter spp. constitute separate taxa from the other Helicobacter spp. The gastric pathological diagnoses of canines with unclassified‐Helicobacter spp. were mild (11), moderate (8), and severe (1) chronic gastritis, adenocarcinoma (1), and high‐grade lymphoma (1). Our polyphasic taxonomic study results suggest that novel Helicobacter strains prevail in Japan. To ascertain the pathogenicity of these strains, further studies, including in vitro cultivation, must be conducted. ID19 PREVALENCE OF GIARDIA DUODENALIS AND CRYPTOSPORIDIUM SPP. AMONGST SENIOR VETERINARY STUDENTS AND THEIR DOGS AND CATS Hanaa Thigeel, Andrea Valeria Scorza, Francisco Olea‐Popelka, Michael Lappin Colorado State University, Fort Collins, CO, USA Giardia spp. and Cryptosporidium spp. are capable of zoonotic transmission and few data are available assessing the genotypes of these organisms in people and their pets. The primary objectives of this study were to identify and to molecularly characterize Giardia duodenalis and Cryptosporidium spp. isolates obtained from feces of senior veterinary students, their dogs and their cats. All senior veterinary students were provided information concerning this anonymous, elective study by hard copy and email. The students were requested to complete a survey designed to assess risk factors for enteric zoonoses and to supply a personal fecal sample as well as a fecal sample from one dog and one cat if present in the household. Feces were analyzed using microscopic examination for parasite eggs, cysts and oocysts after using Sheather's sugar centrifugation. For detection of Cryptosporidium spp. oocysts and Giardia spp. cysts a commercially available FA (MERIFLUOR® Cryptosporidium/Giardia, Meridian Biosciences) was used. DNA extracted from each fecal sample was assayed for G. duodenalis using PCR assays for the glutamate dehydrogenase, triosephosphate isomerase, and ß‐giardin genes. DNA extracted from each human fecal sample was assayed for Cryptosporidium spp. using PCR assays for the 18SrRNA and gp60 genes whereas dog and cat DNA extracts were evaluated by PCR assays for the 18SrRNA and heat shock protein 70 genes. To date, 23 senior veterinary students have supplied a personal fecal sample; 15 concurrently supplied a dog fecal sample and 13 concurrently supplied a cat fecal sample. Giardia spp. and Cryptosporidium spp. (Crypto) were the only parasites detected in the human, dog, and cat samples tested to date. The distribution of results by species and assay are listed in the Table. Species Giardia FF Giardia FA Crypto FA Giardia PCR Crypto PCR Any Giardia Any Crypto Any agent Human (n = 23) 0 (0%) 1 (4.3%) 1 (4.3%) 0 (0%) 3 (13.0%) 1 (4.3%) 4 (17.3%) 4 (17.3%) Dog (n = 15) 1 (6.6%) 2 (13.3%) 0 (0%) 2 (13.3%) 2 (13.3%) 3 (20.0%) 2 (13.3%) 3 (20.0%) Cat (n = 13) 0 (0%) 0 (0%) 0 (0%) 1 (7.6%) 0 (0) 1 (7.6%) 0 (0%) 1 (7.6%) Of the 4 positive people, 1 owned a negative dog, 1 owned a negative dog and cat, 1 owned a negative dog, and 1 owned a dog positive for both Giardia and Cryptosporidium and a negative cat (the owner was Cryptosporidium spp. positive). The other dogs and cats positive for either organism were owned by people that were negative for both organisms. None of the humans or cats had dual infections whereas two dogs were infected with both protozoans. DNA sequencing has been completed in two dogs that were positive for G. duodenalis Assemblage D using tpi and gdh genes and one dog that was positive to Cryptosporidium spp. using the 18SrRNA gene. The sequence analysis for the other PCR positive samples failed due to low DNA concentration. The results of this study show that both protozoans are common in humans, dogs, and cats in the region. The Giardia positive dogs that were successfully sequenced harbored a host‐adapted assemblage and the owners were negative which suggests that zoonotic transfer did not occur in these families. Additional samples will be required to further evaluate the risk for zoonotic transfer of G. duodenalis and Cryptosporidium spp. in this ongoing study. ID20 DETECTION OF FELINE IMMUNODEFICIENCY VIRUS ANTIBODIES IN SERUM OF VACCINATED CATS USING A COMMERCIALLY AVAILABLE KIT Michael Lappin Colorado State University, Fort Collins, CO, USA A vaccine against feline immunodeficiency virus (FIV; Fel‐O‐Vax FIV®, Boehringer‐Ingelheim) is available in some countries and has been shown to protect against some clades of the virus. This virus does not induce high levels of viremia and so the primary serum diagnostic tests detects antibodies to FIV. However, the FIV vaccine readily induces antibodies that are indistinguishable from those induced by natural exposure in some commercially available kits and the western blot immunoassay. New FIV antibody tests are being introduced around the world; the purpose of this study was to evaluate whether a kit (Witness®FeLV‐FIV, Zoetis), marketed in some countries, detects antibodies induced by vaccination. In this study, two experiments were performed using 19 purpose bred kittens that were seronegative for FeLV antigen and FIV antibodies (SNAP®FeLV/FIV Combo, IDEXX Laboratories). Each kitten was inoculated SQ with the inactivated FIV vaccine on Weeks 0 and 4; sera were collected and stored at ‐80C multiple times after vaccination. The kittens were boosted on either Week 35 (11 cats) or 27 (eight cats) and sera were collected 1, 2, or 3 weeks after inoculation and stored at – 80C. The samples were thawed and assayed for FIV antibodies and FeLV antigen using the commercially available kit (Witness®FeLV‐FIV, Zoetis) following the manufacturer's recommendations. None of the 19 kittens was positive for FeLV antigen or FIV antibodies prior to vaccination. However, after the first two vaccinations, FIV antibodies were detected in multiple samples.   Week 0 Week 5 Week 6 Week 7 Week 8 Week 9 Total Tested 19 19 11 2 4 8 54 Positive (%) 0 (0) 14 (73.7) 2 (18.2) 2 (100) 4 (100) 4 (50) 26 (48.1) None of the 11 cats tested Week 34 after the first vaccination was positive in the test. However after the booster vaccination on Week 34 or Week 27, FIV antibodies were detected in serum collected Week 1, 2 or 3 after the booster in 7 of 32 samples; 5 of the 19 cats accounted for these samples. Vaccination against FIV with this product induces antibodies that are detected transiently by the commercially available kit. This assay cannot be used alone to accurately distinguish a cat with naturally occurring FIV infection from a cat that has been vaccinated against FIV. ID21 GENE EXPRESSION OF INTERLEUKINES IN THE SPLEEN OF CATS NATURALLY INFECTED BY LEISHMANIA INFANTUM CHAGASI Juliana Vides, Karina Hirata, Marcia Laurenti, Ludmila Sobrinho, Mary Marcondes Sao Paulo State University, Aracatuba, Sao Paulo, Brazil In an attempt to understand the immune response of cats infected by Leishmania infantum chagasi, the aim of this study was to evaluate the level of expression of genes encoding interleukins IL‐4, IL‐10, IL‐12p40 and IFN‐γ in spleen samples of naturally infected cats, by quantitative reverse transcription PCR (qRT‐PCR). Disease diagnosis was based on the finding of amastigote forms of the parasite by cytological examination and confirmed by DNA amplification of L. infantum chagasi in bone marrow tissue. Three groups of cats were evaluated; the first group composed by six symptomatic naturally infected cats, the second composed by six asymptomatic naturally infected cats and the third composed of six clinically healthy and uninfected cats. Prior to cats inclusion in the study owners were informed of the necessary interventions and their consent was obtained for each cat. The study was approved by the Committee for Ethics in Animal Research, of São Paulo State University. Total RNA was extracted from spleen and converted into cDNA. Cytokines gene expression was measured and final quantitation was evaluated by 2‐ΔΔCT method. An endogenous “housekeeping” gene (beta‐actin) was quantified and used to normalize the results. The mRNA expression of IL‐4 was reduced in both groups in comparison to healthy cats, with gene expression of symptomatic and asymptomatic cats supressed 2.52 and 2.22 times, respectively. The mRNA expression of IL‐10 gene was reduced in all naturally infected cats. The reduction was slightly higher in symptomatic (supressed 2.40 times) than in asymptomatic (supressed 2.15 times) cats when compared to the control group. The mRNA expression of IL‐12 was reduced in both groups in comparison to healthy cats, with gene expression of asymptomatic and symptomatic cats supressed 2.52 and 2.30 times, respectively, in comparison to healthy cats. The mRNA expression of IFN‐γ gene was reduced in all naturally infected cats, regardless of the presence or absence of symptoms, when compared to the control group. The reduction was more significant in asymptomatic (supressed 1.15 times) than in symptomatic (supressed 0.57 times) cats. The results of this study suggest that there is reduced expression, and eventually production, of the four cytokines in the spleen of cats naturally infected by L. infantum chagasi, regardless of the clinical status of animals, and that spleen cells of the studied cats did not to elicit a specific immune response to Leishmania infection. ID22 CANINE AND LIVESTOCK HUSBANDRY PRACTICES AMONG OHIO FARMERS AND POTENTIAL FOR ZOONOTIC DISEASE TRANSMISSION Nancy E. Moran, Jason Stull, Teresa Mills The Ohio State University, Columbus, OH, USA Dogs are often on livestock farms, where they serve important management and companion roles. However, they may also be involved in disease transmission between livestock and people. The increase in canine raw diets, canine companions as family members and an increasing population of immune‐compromised people make this understudied disease pathway particularly important. The objective of this study was to quantify and qualify dog ownership among livestock producers, their husbandry and biosecurity practices, and concern for zoonotic (animal to human) disease potential. A 10‐minute questionnaire was developed. Household and individual‐level data were collected – demographics, number of livestock owned, dog ownership and husbandry practices, and attachment level to and preventive veterinary care for the dogs. It was mailed to 2000 Ohio livestock owners randomly selected from a farming magazine mailing list. Of respondents who had both dogs and livestock, a dog fecal sampling kit was mailed. Returned kits were tested for Salmonella spp. and extended‐spectrum beta‐lactamase‐producing organisms using standard enrichment and selective media. The survey response rate was 33%, of which 223 (73%) were livestock dog‐owners (LDO). Approximately 55% of LDO households included at least one high‐risk individual (i.e., <5 years, ≥ 65 years, diagnosed with an immunocompromising condition). Dog‐owning respondents frequently reported higher‐risk husbandry practices: 54% had outdoor‐only dogs, 50% never leashed or fenced their dogs, 75% rarely/never pick up their dogs’ feces. Dog‐livestock contact was often reported (73%), of which 70% indicated high‐risk management practices (e.g., access to sick pens, new arrivals). Raw meat or animal products were fed in 24% of the households. Many respondents had little/no concern of the dog transmitting a disease to livestock or family members. Households with high‐risk members reported similar husbandry, biosecurity, and concern levels (all P > 0.05). Fecal sample testing is on‐going. High‐risk practices were regularly reported and the level of zoonotic disease concern was low. This indicates a need for improved education and outreach for the livestock dog owning community and the healthcare providers in the area, particularly those working with children, the elderly and those with immunocompromising conditons. ID23 EVALUATION OF SELECTED CANINE VECTOR BORNE PATHOGENS IN BLOOD AND PERICARDIAL SAMPLES FROM DOGS WITH PERICARDIAL EFFUSION Maria‐Dolores Tabar 1, Rebeca Movilla2, Lorena Serrano3, Laura Altet3, Olga Francino4, Xavier Roura2 1Hospital Veterinario San Vicente, San Vicente del Raspeig, San Vicente del Raspeig, Alicante, Spain, 2Hospital Clinic Veterinari, Universidad Autonoma Barcelona, Barcelona, Spain, 3Vetgenomics, Barcelona, Spain, 4Servei Veterinari Genetica Molecular, Universidad Autonoma de Barcelona, Barcelona, Spain In the last years several vector borne diseases have been sporadically associated with pericardial effusion in human and veterinary medicine, namely bartonellosis, leishmaniosis, babesiosis and Lyme disease. The aim of this work is to evaluate selected canine vector borne pathogens nucleic acid in samples from dogs with pericardial effusion. Archived EDTA‐blood and pericardial samples from dogs with pericardial samples were retrospectively included. A quantitative PCR was performed for Leishmania infantum and conventional PCR were performed for Ehrlichia/Anaplasma spp., Hepatozoon canis, Babesia spp., Theileria annae, Ricketssia spp. and Bartonella spp. Partial sequencing of PCR products from positive samples for Ehrlichia/Anaplasma spp and Babesia spp was used to confirm species specificity 59 EDTA‐blood samples and 49 pericardial effusion samples from 59 dogs with pericardial effusion were analysed. The study included 19 females and 40 males, ranging in age from 7 to 14 years. Twenty dogs presented idiopathic pericardial effusion, 34 neoplastic effusion and in five dogs it was unknown. Canine vector borne pathogens were found in 14 dogs (23.7%). Positive dogs reveal DNA of Leishmania infantum in blood and pericardial effusion (n = 1) or only in blood (n = 4), A. platys in blood (n = 1, coinfected with Leishmania infantum), Babesia canis canis in blood (n = 4), Babesia canis canis in pericardial effusion (n = 1), Babesia equi in pericardial effusion (n = 1), Babesia gibsoni in blood (n = 1) and Hepatozoon canis in blood (n = 1). There was no relationship between ethiology of the pericardial effusion and presence of CVBD. Most studies performed to ascertain the ethiopathogenesis of pericardial effusion have focused on bacterial or viral pathogens showing that they are a rare finding in dogs although they are relevant for human pericarditis. Moreover previous research suggested an immune response dominated by humoral effector mechanisms but don′t support a primary immune‐mediated pathogenesis for idiopathic pericarditis in dogs. This study adds new information about the potential role of canine vector borne diseases (CVBD) in canine pericardial effusion. The most frequent pathogens detected were babesiosis and leishmaniosis and corroborated previous literature findings. In dogs infected with leishmaniosis with heavy parasitaemia minimal pericardial effusion can be present that usually resolved spontaneously. In infectious diseases such as babesiosis, heart may suffer from the same pathological processes described in other organs, such as inflammation and hypoxia, causing pericardial effusion. However in this study CBVD were not found more often in dogs with pericardial effusion compared with neoplastic effusion, therefore a cause‐effect relationship can not be stablished with those results. Interestingly some infectious disease such as bartonellosis are suggested to have oncogenic potential and more research is needed to clarify the possible role of infectious diseases in dogs with neoplastic pericardial effusion. It is worth to note that the sensitivity of the PCR protocol used in the present investigation may not have been high enough to detect some pathogens such as Bartonella spp. Moreover Borrelia spp was not included as blood sample is not the ideal sample to detect this pathogen. Another limitation was the retrospective nature of the study, with no information about previous treatments received that may have altered the results of PCR. To conclude molecular biology investigation seems to provide an additional tool in the diagnostic of pericardial effusion with unknown etiology. CVBD warrant more consideration and further research must be performed in dogs with pericardial effusion. ID24 MOLECULAR DETECTION OF SELECTED VECTOR‐BORNE PATHOGENS IN DOGS WITH AND WITHOUT TICK OR FLEA INFESTATION Sahatchai Tangtrongsup 1, Saruda Tiwananthagorn1, Jarunee Loythong1, Duanghatai Sripakdee1, Michael R. Lappin2 1Chiang Mai University, Muang, Chiang Mai, Thailand, 2Colorado State University, Fort Collins, CO, USA The current study was conducted to molecularly investigate the presence of selected vector‐borne pathogens, Anaplasma spp., Ehrlichia spp., Rickettsia spp., Babesia spp., Bartonella spp., and hematotropic Mycoplasma spp., in dogs with or without tick or flea infestation in Chiang Mai, Thailand. In total, 55 dogs were included in this study (28 dogs with and 27 dogs without tick or flea infestation) regardless of their health status. Whole blood and serum samples were collected for complete blood count and blood chemistry profile. Twenty tick and nine flea pool samples were collected from 28 dogs with tick or flea infestation and morphologically identified. DNA samples were extracted from whole blood, ticks and fleas. Polymerase chain reactions (PCR) targeting to each pathogen were performed. DNA sequencing was performed on all positive DNA fragments for genetic information. Association of vector‐borne pathogen infections and age, sex, presence of anemia, thrombocytopenia, evidence of tick or flea infestation, use of tick or flea control products, and allowing of free roaming were assessed using Fisher's exact test. The overall prevalence of at least one vector‐borne pathogen detected in dogs, ticks and fleas were 32.7% (18/55), 40% (8/20) and 66.7% (6/9), respectively. All ticks and fleas were identified as Rhipicephalus sanguineus and Ctenocephalides felis. The most prevalent pathogen in dogs was E. canis (20%), followed by Babesia spp. (12.7%), Bartonella spp. (3.6%) and hematotropic Mycoplasma spp. (1.8%). The most prevalent canine pathogen in R. sanguineus (n = 20) was Bartonella spp. (25%), followed by E. canis (10%), hematotropic Mycoplasma spp. (2%), and Anaplasma phagocytophilum (0.5%). Of nine pool samples from fleas, the most prevalent pathogen detected was Rickettsia felis (66.7%), followed by Bartonella spp. (22.2%). Of 18 PCR positive samples from dogs, 15 (27.3%) and 3 (16.7%) samples were infected with one and two pathogens, respectively. Of 8 PCR positive samples from ticks, 62.5%, 25%, and 12.5% carried one, two and three pathogens, respectively. Of 6 PCR positive samples from fleas, 66.7% and 33.3% carried one and two pathogens, respectively. Genetic information revealed the presence of E. canis, A. phagocytophilum, B. canis vogeli, B. gibsoni, M. haemofelis, ‘Candidatus M. haemominutum’, ‘Candidatus M. haematoparvum’, B. henselae, B. clarridgeiae, and R. felis in dogs or vectors evaluated in this study. Allowing free roaming (P < 0.001) and presence of thrombocytopenia (P = 0.018) were significantly associated with vector‐borne pathogen infection. Other factors were not significantly associated with the infection. The findings suggest that infection of vector‐borne pathogens in dogs in Chiang Mai, Thailand is common. Dogs can be a potential reservoir for vector‐borne infection in cats. Effective tick and flea control in this population is suggested. Thrombocytopenic dogs in the region should be evaluated for vector borne infectious agents. Multiple pathogen detection of vector‐borne pathogens in Thailand may be beneficial as some pathogens are sensitive to different medication. ID25 PERFORMANCE OF THE ANIGEN RAPID CANIV‐4 TEST KIT USING CHARACTERIZED CANINE SAMPLES Brendon Thatcher 1, Melissa Beall1, Jiayou Liu1, Richard Goldstein2, Ramaswamy Chandrashekar1 1Idexx Laboratories, Westbrook, ME, USA, 2Animal Medical Center, New York, NY, USA The Anigen Rapid CaniV‐4 Test Kit is a chromatographic immunoassay for the qualitative detection of Dirofilaria immitis antigen, antibody against A. phagocytophilum / A. platys, antibody against B. burgdorferi and antibody against E. canis in canine serum, plasma or whole blood. In this report, we compared the performance of the Anigen CaniV‐4 test for detection of heartworm antigen in sera from dogs having necropsy‐confirmed infections and for detection of antibodies to A. phagocytophilum, A. platys, B. burgdorferi and E. canis with indirect immunoflourescence assay‐characterized canine samples. In all a total of 220 characterized samples were evaluated. Cani‐V4 Test Strip Sensitivity Compared to Reference Method D. immitis antigen 97.6% A. phagocytophilum/A. platys antibody 73.3% B. burgdorferi antibody 10.0% E. canis antibody 80.0% In summary, the Anigen CaniV‐4 test had very poor sensitivity for B. burgdorferi detecting only 3 of 32 IFA positive samples. The diagnostic sensitivity for other three tick‐borne diseases ranged from 73 to 80%. Specificity was >90% for all four assays. In addition, the test protocol for Anigen CaniV‐4 was complicated by two different sample volume requirements; two drops for heartworm analyte and 10 μL for the tick borne disease markers. ID26 PREVALENCE OF CYCLOSPORA CAYETAENSIS AND OTHER ENTERIC PARASITES IN SHELTER CATS WITH DIARRHEA IN COLORADO, USA Mandy van Koppen 1, Valeria Scprza2, Paul Overgaauw1, Denise Hoek‐van Deursen3, Elena Contreras2, Michael Lappin2 1Utrecht University, Utrecht, The Netherlands, 2Colorado State University, Fort Collins, CO, USA, 3National Institue of Public Health and Environment, Bilthoven, The Netherlands Diarrhea is commonly found in shelter cats and can be caused by different factors such as stress, dietary changes and enteric parasites. Protozoans are commonly detected and most often include Cryptosporidium spp., Giardia spp., and Tritrichomonas foetus (blagburni). Cyclospora cayetanensis is another enteric protozoan that can infect people but little is known about the potential for zoonotic transfer. Thus, the objectives of this study were to optimize a C. cayetanensis PCR assay and apply it to feces of cats with diarrhea in a shelter in Colorado, USA. The optimization of a published SYBR Green based quantitative PCR (qPCR) of a fragment of the internal transcribed spacer 2 (ITS‐2) gene of C. cayetanensis followed by melting curve analysis were performed as published. To determine the linear range of the reaction, a 10‐fold serial dilution of the synthetized oligonucleotide (AF301386) was performed and regression analysis was performed on the resulting standard curve using the instrument software. Intra and interassay variability were determined by running five replicates of the dilution series in three different runs. Specificity for C. cayentanensis was determined by including genomic DNA from G. duodenalis, C. parvum, and Toxoplasma gondii in duplicate along with C. cayentanesis DNA. Optimal annealing temperature was determined by performing an annealing gradient ranging from 53°C to 60°C. Fecal samples from a total of 60 cats estimated to be between 16 weeks and 3 years of age with acute diarrhea of at least 2 days duration that was recognized while housed in a shelter in North Central Colorado were evaluated. Feces were analyzed using microscopic examination for parasite eggs, cysts and oocysts after using Sheather's sugar centrifugation. For detection of Cryptosporidium spp. oocysts and Giardia spp. cysts, a commercial IFA was used (MERIFLUOR® Cryptosporidium/Giardia, Meridian Biosciences). DNA extraction was performed using methods previously described. PCR assays targeting the 18S rRNA and heat shock protein‐70 genes for Cryptosporidium spp. and the glutamate dehydrogenase, b‐giardin and triose phosphate isomerase genes for Giardia spp. were used to speciate positive samples and the extracted DNA was also evaluated in the optimized C. cayetanensis PCR assay. The lowest reliable level of detection for C. cayetanensis DNA was 1fM of DNA/reaction and the assay was linear over a range of four dilutions with an r 2 value of 0.926 and a reaction efficiency of 102%. The melting curve showed that the Tm values for the C. cayetanensis oligonucleotide ranged between 81.1°C and 81.8°C. Amplification of G. duodenalis, C. parvum, and T. gondii DNA was not observed. Optimal annealing temperature was 57°C. Enteric parasites were detected in 8 of 60 cats (13.3%) and included Cryptosporidium spp. (6.7%), Giardia spp. (3.3%) and I. felis (3.3%). DNA of C. cayetanensis was not amplified from any sample. For those samples with adequate DNA for sequencing, the 18S rRNA assay typed only to Cryptosporidium spp. for four isolates and two isolates typed as C. felis. The b‐giardin and the glutamate dehydrogenase genes typed one Giardia spp. as Assemblage F. The parasites detected in this study were similar to those in other similar studies with only feline host adapted strains of Cryptosporidium and Giardia were detected. Since all cats were negative for C. cayetanensis, it is not likely the diarrhea was caused by this agent, but more research needs to be done to investigate whether C. cayetanensis has the potential to infect cats. NU01 A NEW METHOD FOR MEASURING GLOMERULAR FITLRATION RATE (GFR) IN CATS WITH COMPARISON TO TWO STANDARD METHODS Jewel Milo 1, Carl Wesolowski2, Michal J. Wesolowski2, James Montgomery1, Carling McCrea1, Surajith N. Wanasundra2, Jin Huang1, Elisabeth Snead1 1Western College of Veterinary Medicine, University of Saskatchewan, Saskatoon, SK, Canada, 2Royal Univerisity Hospital, University of Saskatchewan, Saskatoon, SK, Canada Kidney disease is one of the most common causes of morbidity and mortality in cats. Early detection is the key to slowing disease progression. GFR is the most accurate means of determining renal function. However, bolus intravenous injection techniques used to measure GFR vary in accuracy and precision. A new, more accurate and precise method of GFR measurement using GFR error adaptive Tikhonov regularization of gamma variates (Tk‐GV) for fitting to 99mTc‐DTPA human data was applied for the first time in veterinary patients. This was compared to standard exogenous creatinine plasma clearance using non‐compartmental area under the curve (Cr‐CL) and iohexol computerized tomographic Patlak plotting (CT‐Pp) in seven laboratory raised cats deemed healthy by evaluation of routine blood work, urinalysis and urine culture. Each cat had their GFR measured twice by each method 7–10 days apart to evaluate for accuracy and precision of GFR‐values. The median of all GFR‐values for Tk‐GV, Cr‐CL and CT‐Pp were 2.64, 3.14‐ and 2.83‐mL/min‐kg, respectively, with similar interquartile ranges (IQRs, 1.12‐ to 1.22‐mL/min‐kg). However, the precision of GFR/kg‐values as measured between studies varied markedly for Tk‐GV, Cr‐CL and CT‐Pp with IQRs of 0.56‐, 1.16‐ and 2.97‐mL/min‐kg, respectively. Using Wilcoxon signed‐rank testing, the differences between the first and second studies for all cats for Tk‐GV, Cr‐CL and CT‐Pp had P‐values of 0.04, 0.8 and 1.0 such that only the Tk‐GV method detected a borderline significant decrease of median GFR of 0.42 mL/min‐kg between the two serial studies. All three methods tested provided GFR measurements within the reported reference range for cats (1.5–3.5 mL/kg/min). However, only the recently developed Tk‐GV method provided stable and precise results. This allowed detection of a difference between sequential studies suggesting superior utility for detecting interval changes in kidney function in healthy cats. NU02 PERIOPERATIVE COMPLICATIONS, MORTALITY, AND LONG‐TERM OUTCOME OF URETERAL SURGERY AND URETERAL STENTING IN CATS: 117 CASES Chloe Wormser, Dana Clarke, Lillian Aronson University of Pennsylvania School of Veterinary Medicine, Philadelphia, PA, USA Due to the heightened prevalence in the number of cats diagnosed with obstructive ureterolithiasis in recent years, surgery on the feline upper urinary tract has become more commonplace. Several techniques have been described for treatment of ureteral pathology in cats, including ureterotomy, neoureterocystostomy, ureteral stenting, and subcutaneous ureteral bypass. However, there is currently a paucity of veterinary literature documenting the success of these procedures. The objective of our study was to report on the perioperative complications and mortality rate in cats undergoing ureterotomy, neoureterocystostomy, and surgical ureteral stent placement. We also describe the long‐term success of intervention, both with regards to the prevalence of persistent clinical signs of urinary tract dysfunction and the prevalence of urinary tract re‐obstruction. Medical records from the University of Pennsylvania Veterinary Teaching Hospital of all cats that underwent ureteral surgery (ureterotomy, neoureterocystostomy) or open ureteral stent placement from 2006 to 2014 were reviewed. Signalment, clinical signs, clinicopathologic test results, and imaging findings were recorded. Surgical procedure performed, perioperative complications, and perioperative mortality were described. Long‐term follow up was obtained via medical record review and client interview. Prevalence of chronic lower urinary tract signs, urinary tract infection, and ureteral re‐obstruction were documented. One hundred and seventeen cats were included in the study. The majority (89 cats, 76%) were Domestic Short Hair cats. The median age at the time of surgery was 8 years (range 1–17 years), and the median weight was 4.4 kg (range 2–7.5 kg). Cases were evenly distributed between the sexes, with 57 (49%) females and 60 (51%) males. Ureterotomy was performed in 72 (62%) cats and neoureterocystostomy in 12 (10%) cats. Forty‐two cats (37%) had a ureteral stent placed. Perioperative complications related to the urinary tract were identified in 10 cases (8.5%) including uroabdomen (eight cats) and stent migration (two cats). The perioperative mortality rate was 8.5%, with the majority of deaths (7/10) attributable to progressive renal failure. The median duration of hospitalization was 6 days (range 2–19 days). Long‐term follow up was available for 87 cats (74%). Of those, 48 (55%) were alive, and 39 (45%) were dead. Of the cats that were dead at follow up, death occurred a median of 450 days post‐operatively (range 15–2332 days). Death was attributed to progressive chronic renal failure in 25 cats (64%), ureteral re‐obstruction in seven cats (18%), and disease processes unrelated to the urinary tract in five cats (13%). An unknown cause of death was assigned to the remaining 2 (5.1%) deceased cats. For the cats that were alive at follow up, the median interval between surgery and follow up was 543 days (range 65–5000 days). Of the cats for which long‐term follow up was available, 19 (22%) suffered from ureteral re‐obstruction, 17 (20%) had chronic lower urinary tract signs, 12 (14%) had urinary tract infections, and 2 (2%) had stent encrustation. The results of this study suggest that the perioperative mortality rate of ureteral surgery is lower than that previously reported. Importantly, death was most commonly due to progressive renal failure and not surgical complications. Chronic lower urinary tract signs and urinary tract infection were common, particularly in cats with ureteral stents placed. The high incidence of re‐obstruction warrants appropriate client counseling and judicious follow up. NU03 EFFECT OF DIALYSATE SODIUM CONCENTRATION ON SERUM SODIUM CONCENTRATION DURING INTERMITTENT HEMODIALYSIS IN DOGS Jonathan Foster 1, Kenneth Drobatz1, Larry Cowgill2 1University of Pennsylvania School of Veterinary Medicine, Philadelphia, PA, USA, 2University of California‐Davis Veterinary Medical Teaching Hospital, Davis, CA, USA Derangements of sodium and water balance are clinically significant consequences of kidney failure. During hemodialysis, sodium will equilibrate between the patient and the dialysate, providing an opportunity for changes to occur in the patient's plasma sodium concentration (PNa). It is not currently known how a dialysate‐patient sodium gradient (D‐PNa) may affect the PNa. The purpose of this study was to compare dialysate sodium concentration (DNa) to the pre‐ and post‐dialysis sodium concentrations in dogs undergoing intermittent hemodialysis to identify a parameter that can be used to better prescribe DNa to correct alterations in plasma sodium. Medical records from two institutions were evaluated retrospectively from 10/2009 to 4/2014 to identify dogs that underwent intermittent hemodialysis utilizing a Gambro Phoenix platform. Dogs were excluded if different chemistry methodologies were used to measure pre‐ and post‐dialysis sodium concentrations, if concurrent intravenous fluids were administered, or if dialysate sodium profiling was performed. Linear regression was utilized to determine the correlations between dialysate and patient sodium concentrations. Two hundred and eleven treatments performed on 40 dogs were found to fit the inclusion criteria. Statistically significant correlations were found between the D‐PNa and change in post dialysis PNa, the DNa and the final PNa, and the starting D‐PNa to final D‐PNa. The initial D‐PNa had the strongest correlation to the observed change in patient sodium; the predicted change in PNa is predicted by 0.73 * D‐PNa – 5.83 (regression coefficient = 0.73, 95%CI: 0.67–0.79, r 2 = 0.72, P < 0.0001). The DNa was moderately correlated with the final PNa (regression coefficient = 0.76, 95%CI: 0.66–0.86, r 2 = 0.50, P < 0.0001). The initial D‐PNa had the weakest correlation to the final D‐PNa (regression coefficient = 0.27, 95%CI: 0.21–0.33, r 2 = 0.27, P < 0.0001). A 1 mmol D‐PNa gradient should produce 0.73 mmol change in PNa over the course of the dialysis treatment. These results suggest the initial D‐PNa had the best correlation with the final PNa. It has potential to be used in the prescription of DNa to correct disorders in pre‐dialysis sodium concentration and prediction of the final PNa. NU04 URINARY ACTIVE TGF‐β1 IN FELINE CHRONIC KIDNEY DISEASE Jack Lawson, Harriet Syme, Caroline Wheeler‐Jones, Jonathan Elliott Royal Veterinary College, London, UK Chronic kidney disease (CKD) is a common condition of aging cats, with the predominant pathological diagnosis being tubulointerstitial inflammation and fibrosis. The cytokine TGF‐β1 has been widely implicated in the development and progression of renal fibrosis in experimental models of kidney disease, and total TGF‐β1 has been demonstrated to be elevated in the urine of cats with CKD. The aims of this study were: firstly, to validate a method for the quantification of active TGF‐β1 in feline urine; secondly, to assess the association between urinary active TGF‐β1 and (a) the development of azotemic CKD in a cross‐sectional study and (b) the deterioration of renal function in a longitudinal study. The records from 2 first opinion practices were searched for cats >9 years of age with and without renal azotemia. Renal azotemia was defined as a plasma creatinine concentration >2.0 mg/dL with a concurrent urine specific gravity (USG) <1.035, or demonstration of azotemia on 2 consecutive occasions (at least 14 days apart) without evidence of a pre‐renal cause. Cats were excluded if concurrent hyperthyroidism or other significant systemic disease were suspected. Active TGF‐β1 concentrations were measured in feline urine using an active TGF‐β1 ELISA developed for use with human samples, validated by intra‐ and inter‐assay variability, assessment of dilutional parallelism and spike and recovery of exogenous TGF‐β1. For the purposes of the cross‐sectional study, cases were classified based upon baseline status and subsequent renal function, resulting in three groups: cats with renal azotemia (Group 1), non‐azotemic cats which remained non‐azotemic for 24 months (Group 2) and non‐azotemic cats which developed azotemia within 24 months (Group 3). Comparisons between groups were made using the Kruskal‐Wallis test. For the purposes of the longitudinal study cases with three consecutive urine samples leading up to a deterioration of renal function were classified into two groups: non‐azotemic cats which developed azotemia after ~12 months and azotemic cats with progression of CKD after ~12 months. Progression was defined as a repeatable >25% increase in plasma creatinine from baseline. Comparisons between visits within these two groups were made using the Friedman test with post‐hoc Wilcoxon sign rank tests where appropriate. Results are reported as median [range] or mean (± S.D) as appropriate. Intra‐assay and inter‐assay CV of urinary active TGF‐β1 measurements were 2.3–7.5% (n = 4) and 4.7–20.9% (n = 4) respectively, dilutional parallelism indicated a mean recovery of 91.6% ± 17.5% (n = 4) and spiking with exogenous TGF‐β1 resulted in mean recovery of 115.5% ± 12.9% (n = 9). Cross sectional analysis revealed no significant difference in urinary active TGF‐β1:creatinine ratio (UTGF:C) between groups with differing renal function (Group 1: 17.6 pg/mg [0–39.0] n = 20, Group 2: 14.5 pg/mg [0–81.2] n = 23, Group 3: 21.2 pg/mg [0–185.0] n = 24). In the longitudinal analysis, non‐azotaemic cats which developed CKD (n = 6) demonstrated a significant (P = 0.028) increase in UTGF:C approximately 6 months prior to development of azotaemia, which remained elevated (P = 0.046) at the point of diagnosis (~12 m prior: 8.4 pg/mg [0–16.0], ~6 m prior: 22.2 pg/mg [10.8–34.0], CKD diagnosis: 24.6 pg/mg [6.3–63.3]) whilst in cats with progressive CKD (n = 6) the UTGF:C did not alter significantly over time (~12 m prior: 23.9 pg/mg [2.0–61.9], ~6 m prior: 30.7 pg/mg [9.5–85.0], progression of CKD: 42.7 pg/mg [23.2–68.0]). In conclusion, no significant difference was found between cats with differing degrees of renal function in the cross sectional study but UTGF:C did increase prior to documented deterioration of renal function in healthy cats which developed CKD in the longitudinal analysis. This suggests urinary TGF‐ β1 followed longitudinally in an individual cat may indicate the development of CKD in cats, and further investigation into the role of this mediator is warranted. NU05 PLATELET FUNCTION IN DOGS WITH NATURALLY OCCURRING CHRONIC KIDNEY DISEASE Alicia Dudley, Julie Byron, Mary Jo Burkhard, Emma Warry The Ohio State University, Columbus, OH, USA Bleeding is a complication of chronic kidney disease (CKD) in humans, partially caused by platelet dysfunction and abnormal platelet‐vessel wall interactions. Dogs with experimentally induced CKD have prolonged buccal mucosal bleeding times, but coagulation and platelet function have not been evaluated. We compared platelet function between healthy dogs and dogs with CKD using the PFA‐100® and thromboelastography (TEG), and platelet GPIb, GPIIb‐IIIa, and p‐selectin expression. Eleven dogs with naturally occurring IRIS stage IV CKD and 10 control dogs were enrolled. Complete blood count, serum biochemistry, urinalysis, urine protein:creatinine ratio, prothrombin time, partial thromboplastin time, fibrinogen concentration, and antithrombin activity were performed. PFA‐100® closure times (CT) using collagen and epinephrine (Col+EPI) and Col+ADP agonists were measured. Reaction time (R), clot formation time (K), α‐angle (α), maximal amplitude (MA) and global clot strength (G) TEG variables were analyzed. Platelet GPIb, GPIIb‐IIIa and p‐selectin expression was assessed by flow cytometry. CKD dogs had prolonged Col+ADP CT (P = 0.01). No difference in Col+Epi CT was found. TEG MA (P < 0.01) and G (P = 0.01) were increased and K‐time (P = 0.035) was decreased in CKD dogs. CKD dogs had increased platelet GPIIb‐IIIa and p‐selectin mean fluorescence intensity compared with controls (P = 0.01 and 0.03). There was no difference in any remaining variables. No correlation was found between platelet receptor expression and other variables. Dogs with CKD may have platelet dysfunction, but are hypercoagulable based on elevated TEG values. Platelet receptor expression may not explain changes in platelet function and coagulation NU06 EVALUATION OF EFFECTIVENESS AND POSSIBILITY OF NAFAMOSTAT MESILATE AS AN ALTERNATIVE ANTICOAGULANT DURING INTERMITTENT HEMODIALYSIS ON HEALTHY BEAGLE DOGS Ji Hye Hwang 1, Joon Hyuk Choi1, Hwa Young Youn2, Kun Ho Song1, Jong Bok Lee3, Kyoung Won Seo1 1Chungnam National University, Daejeon, Korea, 2Seoul National University, Seoul, Korea, 3OraeOrae Animal Hospital, Seoul, Korea Nafamostat Mesilate (NM) is a protease inhibitor with a short half‐life used as an anticoagulant during the hemodialysis on patients with bleeding‐risk. The aim of this study is evaluating NM as an alternative anticoagulant agent in intermittent hemodialysis (IHD). Eighteen healthy beagle dogs were used. In group 1 (n = 6), NM was administered at the dose of 0.5 mg/kg/h during 5 hours of IHD. In group 2 (n = 6), NM was administered at the lower dose of 0.25 mg/kg/h during IHD. In group 3 (n = 6), which is the control group, unfractionated heparin was administered during IHD. Evaluation parameters included residual blood clot in blood chamber and dialyzer arterial sides, hemoglobin, hematocrit, platelets, prothrombin time (PT), activated partial thromboplastin time (aPTT) and activated clotting time (ACT). Both group 1 and group 2 successfully completed IHD without serious coagulation on extracorporeal circulation. Statistical significance was not identified at the clot formation in the residual blood clot of blood chamber and dialyzer arterial sides in group 1, 2 and 3 (group 1 and 3, P = 1.000, group 2 and 3, P = 1.000). In group 1 and group 2, statistically significant change in PT (P = 0.476 and 0.597; respectively), aPTT (P = 0.983 and 0.977; respectively), and ACT (P = 0.282 and 0.401; respectively) was not observed. In group 3, statistically significant elevation of ACT was observed (P = 0.000). The result of study suggests that it is possible to use low‐dose NM (0.25 mg/kg/h) as an alternative anticoagulant on patients with bleeding‐risk during IHD. NU07 MODIFYING URINE GLYCOSAMINOGLYCAN CONCENTRATIONS BY INTRAMUSCULAR INJECTION OF POLYSULFATED GLYCOSAMINOGLYCANS IN DOGS Michael Wood University of Wisconsin‐Madison, Madison, WI, USA An intact urothelial glycosaminoglycan (GAG) layer creates a hydrophilic barrier capable of preventing the adherence of uropathogenic bacteria to the bladder wall. A defective barrier may predispose patients to recurrent urinary tract infections (UTIs). In human studies, the intravesicular instillation of exogenous GAG replacers reduced UTI recurrence from 4.5 to 0.4 infections per year. In veterinary medicine bladder instillation therapy is impractical as it requires regular catheterization and urine retention. Intramuscularly administered GAGs are primarily excreted through the urinary tract and may provide an alternative method of urinary GAG administration. The aim of the described study was to determine the magnitude and duration of urinary GAG excretion after the intramuscular injection of a polysulfated glycosaminoglycan (PSGAG). For this study 12 client owned dogs received a single intramuscular 5 mg/kg injection of PSGAG (n = 8) or no treatment (n = 4). Urine samples were collected prior to PSGAG injection (day 0) and each day afterward for 6 days (day 1–6) for both the treatment and control groups. Urine GAGs were quantified using a spectrophotometric dimethylmethylene blue assay. Variable urine output, and hence GAG dilution, was accounted for by dividing the resultant optical densities by urine creatinine (uGAG:uCrea). Ratios for days 1–6 were then compared to baseline day 0 values to calculate the fold change in GAG excretion each dog. After intramuscular injection with PSGAG the median uGAG:uCrea increase was 2.5× on day 1, 1.8× on day 2 and 1.1× on day 3. These values were not significant when compared to control dog uGAG:uCrea that ranged from 0.7–1.1× baseline GAG concentrations. These data indicate that increases in urinary GAG concentrations after an intramuscular injection of PSGAG are unlikely to achieve the concentrations used in intravesicular instillations. Instead, the PSGAGs are excreted at a low level over the period of 3 days. The therapeutic benefit of low level GAG infusions for the prevention of UTI is unknown at this time. NU08 EFFECTS OF GRADUAL INCREASE OF DIETARY SODIUM CHLORIDE ON URINARY PARAMETERS AND RELATIVE SUPERSATURATION IN HEALTHY CATS Yann Queau, Laurence Le Verger, Alexandre Feugier, Vincent Biourge Royal Canin Research Center, Aimargues, France Sodium chloride (NaCl) is used in diets to promote water intake and urine dilution in dogs and cats at risk for urolithiasis. All but one studies in both species show that dietary Na≥ 2.6 g/1000 kcal decreases relative supersaturation (RSS) of calcium oxalate (CaOx), but the threshold is unclear. The objective was to determine the dose response relationship between dietary NaCl and CaOx RSS in cats. Thirteen cats (seven females) aged 3.8–5.6 years were fed sequentially four dry diets for 7 days, followed by 3 days of individual urine collection. Diets (A,B,C,D) were identical but NaCl was added to obtain sodium of 0.7, 1.7, 2.4 and 3.3 g/1000 kcal respectively. Volume, pH and concentrations of ten ions were measured in the individual urine pools. CaOx RSS was calculated using SUPERSAT. A linear mixed model assessed the effect of diet (fixed effect in 4 levels), including cat as a random term. Data are expressed as Least Square Means±SE as residuals of the model were normally distributed. Caloric intakes were similar between diets (P = 0.83). In the table, superscripts indicate significant differences between diets (Scheffe adjustment for test multiplicity). Diet A B C D P‐value Urine volume (mL/kg/day) 10.9 ± 0.69a 13.7 ± 0.69b 15.5 ± 0.69b 19.6 ± 0.69c <0.001 Urine Ca (mmol/L) 0.66 ± 0.05a 0.58 ± 0.05a 0.48 ± 0.05b 0.41 ± 0.05b <0.001 Urine Ox (mmol/L) 1.99 ± 0.08a 1.67 ± 0.08ab 1.55 ± 0.08b 1.02 ± 0.08c <0.001 RSS CaOx 3.39 ± 0.25a 2.80 ± 0.25b 2.41 ± 0.25b 1.64 ± 0.25c <0.001 Increasing dietary levels of NaCl was effective at decreasing urinary mineral concentrations and CaOx RSS in healthy cats. A longer‐term study in a population of stone‐forming animals is warranted. NU09 SERUM MAGNESIUM CONCENTRATION IN DOGS WITH NATURALLY OCCURRING CHRONIC KIDNEY DISEASE Cinthia Martorelli 1, Marcia Kogika1, Luciano Giovaninni1, Douglas Caragelasco1, Fernanda Chacar1, Ana Carolina Pinto1, Carla Lorigados1, Lucia Andrade2 1School of Veterinary Medicine and Animal Science, University of Sao Paulo, Sao Paulo, Brazil, 2School of Medicine – University of Sao Paulo, Sao Paulo, Brazil The kidneys play an important role in magnesium homeostasis and the renal handling of magnesium is highly adaptable, and when renal function declines significantly, increase in serum magnesium is detected. However, hypomagnesemia is also common in chronic kidney disease (CKD) in human, and it has been associated with the development of vascular calcification and the poor clinical outcomes. Most of studies of serum magnesium concentrations in veterinary medicine were related to a few diseases such as feline and canine diabetes mellitus, canine parvoviral enteritis, and protein‐losing enteropathy in dogs, but so far, there is no data in dogs with CKD. The goal of this study was to evaluate the serum total magnesium in dogs with chronic kidney disease in stages 2 and 3 (IRIS). Sixteen CKD dogs were included in the study (stage 2, n = 6; stage 3, n = 10), and they were followed‐up for two to 12 months, and the measurements of serum total magnesium (kit MG3880, Randox) were performed monthly. Serum total magnesium was also measured in 10 clinically normal dogs (mean ± SD; 2.24 ± 0.31 mg/dL); the normal range considered for this study was from 1.93 to 2.55 mg/dL. Mean of serum magnesium concentrations in CKD dogs stage 2 and stage 3 were 2.32 mg/dL (SEM ± 0.10) and 2.40 mg/dL (± 0.07), respectively. Hypermagnesemia (>2.55 mg/dL) was detected in all dogs in stage 2 (6 out of 6) and in 8 out of 10 (80%) dogs in stage 3, and it was noticed at some point of the follow‐up, and usually it was associated with the highest level of serum creatinine. Hypomagnesemia (<1.93 mg/dL) was detected in 5 out of 6 dogs (83%) in stage 2 and in all dogs (100%) in stage 3 at some point, and usually it was detected in the beginning of the follow‐up. No correlation was observed between serum magnesium and serum urea, creatinine, ionized calcium, total calcium, bicarbonate and phosphorus. In conclusion, magnesium may be considered as an indicator or a marker of progression of CKD in dogs, since the impairment of renal function may result in hypermagnesemia, as it has been observed in humans with CKD. In other hand, it has been well proven that in humans and rats, hypomagnesemia is associated with faster deterioration of kidney function and vascular calcification. The role of magnesium in dogs with CKD has not been investigated yet, and its clinical importance could be underestimated. Thus, additional studies are needed for the better understanding of magnesium in the pathophysiology of mineral metabolism in dogs with chronic kidney disease. NU10 A COMPARISON OF PERIOPERATIVE OUTCOME IN CATS UNDERGOING URETEROTOMY OR URETERAL STENTING FOR THE TREATMENT OF BENIGN URETERAL OBSTRUCTIONS William Culp 1, Carrie Palm1, Christine Hsueh1, Philipp Mayhew1, Geraldine Hunt1, Eric Johnson1, Kenneth Drobatz2 1University of California‐Davis, Davis, CA, USA, 2University of Pennsylvania, Philadelphia, PA, USA To prospectively evaluate the perioperative outcome of cats with benign ureteral obstructions undergoing ureteral stenting with a secondary goal of comparing to similar cases undergoing ureterotomies. Cohort study Sixty‐two client‐owned cats that underwent ureteral stenting (n = 26) or ureterotomy (36) for treatment of benign ureteral obstruction. Cats diagnosed with benign ureteral obstruction were treated with ureteral stenting. A historical group of cats with benign ureteral obstruction treated with ureterotomy was identified for comparison. Data was recorded prospectively (ureteral stent cases) or collected retrospectively from the medical records (ureterotomy cases) and variables were statistically compared. Cats treated with ureteral stent(s) had a significantly greater decrease in blood urea nitrogen (BUN) and creatinine 1‐day postoperatively and at discharge, as compared to cats that underwent ureterotomy. Cats developing abdominal effusion postoperatively were significantly less likely to be discharged from the hospital. Cats in the ureteral stent group were significantly more likely to develop an abdominal effusion when a ureterotomy was performed in conjunction with stent placement. Cats with ureteral stents were significantly more likely to have resolution of azotemia in‐hospital, and cats with resolution of azotemia while hospitalized were significantly more likely to be discharged from the hospital. Cats undergoing ureteral stent placement demonstrated early postoperative improvement in azotemia and a greater chance of resolution of azotemia, as compared to cats undergoing ureterotomy only. The early postoperative results of ureteral stenting are encouraging. NU11 SIGNALMENT AND DIAGNOSIS OF 205 NEUTERED FEMALE DOGS WITH URINARY INCONTINENCE IN THE UNITED STATES Julie Byron 1, Katherine Taylor1, Gary Phillips1, Madeleine Stahl2 1The Ohio State University, Columbus, OH, USA, 2Merck Animal Health, Madison, NJ, USA Urethral sphincter mechanism incompetence (USMI) is the most common cause of acquired urinary incontinence in neutered female dogs. Several factors may impact the risk of USMI including weight, body condition score (BCS), and age at neuter. The objectives of this study were to describe a population of spayed female dogs with USMI and investigate their initial presentation, diagnosis, and treatment. Information was collected on 205 dogs with USMI (UI) diagnosed between January 2009 and December 2012, and 197 continent neutered female dogs (C). Neither age nor BCS differed between groups. UI dogs weighed more than C dogs (P = 0.001). There was no significant difference in age at neuter between UI dogs and C dogs, however, for every 3 months delay in neutering during the first year of life, USMI risk declined (P = 0.038). Median time from neuter to UI signs was 3.43 years, and from first signs to veterinary consult was 21 days. 85.7% of dogs were diagnosed with USMI at initial presentation. Phenylpropanolamine (PPA) was prescribed in 77.45%, diethylstilbestrol in 19.61%, and combination therapy in 2.94% of dogs. This data supports previous studies showing heavier dogs have increased risk for USMI and that its onset occurs within a few years of neuter. Risk association with age at neuter remains controversial. Owners appear likely to seek veterinary care relatively soon after the onset of clinical signs, and in most cases diagnosis of USMI is prompt. Treatment choices may reflect the differences in availability of PPA and estrogenic compounds during the period studied. NU12 CONSUMPTION OF VIYO RECUPERATION RENAL – ADVANCED FORMULA IN CATS WITH STABLE CHRONIC KIDNEY DISEASE AND ITS EFFECTS ON BIOCHEMICAL PARAMETERS Crystal Cooley, Jessica Quimby, Laura Martin, Michael Lappin Colorado State University, Fort Collins, CO, USA Viyo is a nutritional supplement that may increase appetite and water consumption. This pilot study evaluated effects of Viyo Recuperation™ – Renal (VRR) on biochemical parameters in cats with chronic kidney disease (CKD) and compared palatability in CKD cats with normal appetites versus CKD cats with decreased appetites. In this pilot study, a total of 12 cats with stable CKD (creatinine concentration >2 mg/dL and <10 mg/dL) were enrolled in two groups; 6 CKD cats with normal appetite and 6 CKD cats with decreased appetites. CBC, serum biochemistry, urinalysis, T4 and blood pressure were measured prior to enrollment. 30 mL of VRR was offered prior to meals daily for 14 days, and amount voluntarily consumed in 24 hours was measured. Owners scored appetite and water consumption daily. Body weight and biochemical profiles were assessed on day 0, 7, and 14. Wilcoxon test was used to compare results pre and post treatment. Prior to starting the trial, the median amount of food consumed by cats in the normal appetite group was 100% (range 100–100%) and the median amount of food consumed by cats in the decreased appetite group was 50% (range 25–100%), but there was no significant difference in biochemical parameters between the groups. Of the CKD cats with a normal appetite, 5 of 6 consumed >70% of the total 14 days volume of VRR and a statistically significant decrease in serum phosphorus (P = 0.03) was detected. There were no significant differences in other biochemical parameters or body weight. In CKD cats with decreased appetites, 3 of 6 cats consumed >70% of the total 14 days volume of VRR. There were no obvious differences in measured biochemical parameters between the cats that consumed <70% or >70% of the total 14 days volume of VRR, however the numbers of cases entered to date is too small for statistical comparisons. Overall, VRR was palatable to cats with CKD and decreased serum phosphorus levels in CKD cats with normal appetites. Cats with CKD and a decreased appetite were less likely to voluntarily consume VRR and data collection is continuing to determine overall effects on appetite and biochemical findings in a greater number of cats. NU13 EFFECT OF BLOOD CONTAMINATION IN CANINE URINE ON THE PERFORMANCE OF A COMMERCIAL IMMUNOASSAY FOR THE ACUTE KIDNEY INJURY (AKI) MARKER: URINARY CLUSTERIN (UCLUS) Jack Quinn, Mirolee Zieba, Murthy Yerramilli IDEXX Laboratories Inc, Westbrook, ME, USA Clusterin or Apolipoprotein J is a 75–80 kDa disulphide linked heterodimeric protein. Clusterin has been shown to be up‐regulated during proximal and distal tubular injury. uClus is part of the FDA and ICH renal toxicity biomarker panels for drug development. Clusterin changes are more profound and significant in urine compared to serum during kidney injury events. A commercial EIA kit from Biovendor is available for the quantification of canine clusterin in both serum and urine. A recent study validated the biomarker using this kit in dogs with leishmaniasis. In healthy dogs serum concentrations of clusterin are 1000‐fold higher (60–70 μg/mL) than the urinary concentrations (<100 ng/mL). It has been reported that uClus is an AKI biomarker which increases during and, as a result of, kidney injury in dogs. Contamination of canine urine samples with blood is a commonly observed occurrence due to infection, trauma, neoplasia and inflammation, and accidental contamination during catherization and cystocentisis. Hence it is important to ensure that the quantification of uClus is not impacted by contamination of serum clusterin from the blood. Failure to do so could result in false positives. A negative canine urine sample was value assigned using the kit and it was then spiked with various amounts of negative canine serum (0.002% to 10% v/v). The resulting mixtures were analyzed using the kit and the results obtained are shown in the table below: Sample Observed [uClus] ng/mL Neat Negative Urine 13 Urine + 10% Serum 4869 Urine + 5% Serum 2587 Urine + 2% Serum 1142 Urine + 1% Serum 623 Urine + 0.2% Serum 113 Urine + 0.01% Serum 62 Urine + 0.002% Serum 23 These results show that the present assay is impacted by even minute amounts of blood which are not visible to the naked eye. This means that the patient samples that have any hint of blood contamination have to be evaluated very carefully since the possibility of false positives leading to false clinical diagnoses is increased. NU14 EFFECT OF HEMATURIA ON URINE PROTEIN:CREATININE RATIO IN CATS Aida I. Vientós‐Plotts 1, Ellen N. Behrend1, Dennis Chew2, Elizabeth G. Welles1, Hollie P. Lee1 1Auburn University College of Veterinary Medicine, Auburn, AL, USA, 2The Ohio State University College of Veterinary Medicine, Columbus, OH, USA Measurement of urine protein:creatinine ratios (UPCR) is important in the detection, evaluation, and monitoring of renal disease. As no studies, to our knowledge, have evaluated the effects of hematuria on UPCR in feline urine, our purpose was to do so. Consecutive feline urine samples that were yellow and had <5 RBC and WBC per hpf were used. Samples were pooled to a final volume ≥ 7 mL (n = 30 pools). Feline blood (1:20 dilution of blood in urine) was added to an aliquot of each pool. Serial dilutions (1:2) were performed with the remainder of each pool. The undiluted pool and the 1:20, 1:40, 1:80, 1:160, 1:320, 1:1280 and 1:5280 dilutions were evaluated in a blinded fashion. Urine sample color was recorded (light yellow, dark yellow, light pink, dark pink, or red). Pools were centrifuged and supernatants collected and stored at −20C. Protein (colorimetric assay) and creatinine concentrations were measured via automated clinical chemistry analyzer. A signed rank test was used to compare UPCR between the undiluted samples and each color. Also, results for the light and dark yellow were averaged and called “yellow” and for the light and dark pink and red were averaged and called “colored”. Data were compared between undiluted, “yellow” and “colored” using a repeated measures ANOVA on ranks; post hoc comparisons were made via the Tukey test. A Chi‐square was used to compare categorizations (normal, borderline or high) between colors. Significance was set at the P < 0.05 level. Color (Average RBC/HPF) Normal (UPCR <0.2) Borderline (UPCR 0.2–0.4) High (UPCR >0.4) Undiluted (<5) 9 12 9 Yellow (285) Light Yellow (138) 7 14 9 Dark Yellow (432) 1 11 18 Colored (1105) Light Pink (604) 0 10 20 Dark Pink (957) 0 0 30 Dark Red (1747) 0 0 30 UPCR were significantly increased (P < 0.001) in “colored” and “yellow” samples compared to the undiluted samples and in all individual colors compared to undiluted. Proteinuria categorizations differed significantly (P < 0.001) between undiluted and all colors except light yellow (see table). Proteinuria categorizations differed significantly between undiluted and “colored” but not between undiluted and “yellow”. Previous studies using canine urine suggested that hematuria should only be considered as a differential diagnosis for proteinuria if the sample is visibly hematuric. In contrast, our data suggest that the presence of hematuria significantly increases UPCR in feline urine samples that are not visibly hematuric. Thus, the presence of blood may have a greater effect on UPCR in cats than in dogs, and sediment examination results must be taken into account when interpreting UPCR. NU15 ACUTE EVALUATION OF URETHRAL THERMOPLASTY IN NORMAL DOGS Carrie Palm 1, Brian Hardy2, Dana Clarke3, William Culp1, Simon Thomas4, Jeffrey Solomon2 1University of California at Davis School of Veterinary Medicine, Davis, CA, USA, 2Infiniti Medical, Menlo Park, CA, USA, 3University of Pennsylvania School of Veterinary Medicine, Philadelphia, PA, USA, 4Palo Alto Health Sciences, Danville, CA, USA Urinary incontinence affects up to 20% of spayed female dogs. Many dogs do not respond to current treatment regimens and remain incontinent. In women, a minimally invasive procedure, low temperature urethral thermoplasty (LTUT), restores continence by reducing the compliance of the periurethral tissue through the controlled application of heat. The goals of this study were to determine the feasibility of LTUT in female dogs, and to evaluate for acute complications following treatment. Additionally, treated canine urethral tissue was examined for histologic changes previously associated with successful LTUT. Three female purpose bred mongrel dogs (18–22 kg) were treated with LTUT. The thermoplasty probe was positioned in the proximal urethra via fluoroscopic guidance. Heating of the periurethral tissue was performed in three different planes for a total of 48–72 treatment sites. Within 24 hours of treatment, the bladder and urethra were removed en bloc, distended with saline to investigate for leakage at treatment sites and evaluated grossly for serosal lesions. The bladder and urethra were then fixed, and evaluated histologically. Urethral thermoplasty was performed successfully in each case without complication. No leakage or serosal lesions were noted on gross examination. Changes typical of thermal therapy consisting of focal congestion/edema of connective tissue and muscle were noted histologically. Urethral thermoplasty is feasible and safe in the acute setting and induced the acute histopathologic changes required for successful treatment. Long term‐studies in research dogs and eventually clinical patients are necessary to determine the efficacy of LTUT for treatment of canine incontinence. NU16 EFFECTS OF HEMATURIA ON DIPSTICK EVALUATION IN DOGS AND CATS Aida I. Vientós‐Plotts 1, Ellen N. Behrend1, Dennis Chew2, Elizabeth G. Welles1, Hollie P. Lee1 1Auburn University College of Veterinary Medicine, Auburn, AL, USA, 2The Ohio State University College of Veterinary Medicine, Columbus, OH, USA Dry reagent strips (dipsticks) are commonly used as part of a routine urinalysis. To our knowledge no studies have evaluated the effects of blood on the dipsticks. As the results depend on the color of the reagent pad, discolored urine could potentially affect the final readings. Therefore, the purpose of this study was to determine the effect of hematuria on manual urine dipstick evaluation in dogs and cats. Consecutive urine samples from dogs and cats submitted for urinalysis that were yellow and had <5 RBC and WBC per hpf on sediment exam were used. Sample supernatants were pooled to obtain a final volume of ≥ 7 mL; 30 pools were evaluated from each species. Canine or feline whole blood was added to an aliquot of each pool to obtain a 1:20 dilution of blood in urine. Serial dilutions (1:2) were performed with the remainder of each pool. The undiluted pool and the 1:20, 1:40, 1:80, 1:160, 1:320, 1:1280 and 1:5280 dilutions were evaluated (Siemens MultiStix 10 SG Reagent strips) in a randomized blinded fashion per manufacturer instructions and were visually inspected. The greatest dilution resulted in urinary sediment with 80–100 RBC/hpf. The color of the urine sample was recorded (light yellow, dark yellow, light pink, dark pink, or red). Specific gravity (SG) was measured with a handheld refractometer (Reichert TS 400). For statistical comparisons, the results for the light and dark yellow were averaged and called “yellow.” and the results for the light and dark pink and red were averaged and called “colored.” A repeated measures ANOVA on ranks was used to compare the readings for each analyte between undiluted, “yellow” and “colored.” Post hoc comparisons were made with the Tukey test. Significance was set at the P < 0.05 level. In canine samples, bilirubin, ketones, pH, protein and SG were not significantly different between the undiluted and yellow samples, but the colored samples were significantly different from both the undiluted and yellow samples. Despite the color of the sample, the presence of any blood resulted in significantly increased readings for blood; once blood was added most readings were 3+. In feline samples, ketones, pH, protein and SG were not significantly different between the undiluted and yellow samples; however, the colored was significantly different from both the undiluted and yellow samples. Despite the color of the sample, the presence of blood resulted in significantly increased readings for blood and for bilirubin. Understanding the effect of hematuria on the urine dipstick is necessary for accurate interpretation of results. Glucose readings were not affected regardless of the amount of blood in a sample. Ketone readings were significantly higher in colored urine in dogs and cats. Bilirubin was significantly increased in cat urine with addition of any RBC even if the sample remained visibly yellow; in dog urine, bilirubin was not significantly affected until the sample was visibly pink. The dipstick protein reading is not significantly elevated in either species until the sample is visibly pink. Any degree of hematuria, despite the color of the sample, affects the blood reagent pad for blood in dogs and cats. NU17 FELINE SERUM CYSTATIN C DOES NOT DIFFERENTIATE HEALTHY CATS FROM THOSE WITH CHRONIC KIDNEY DISEASE (CKD) Giosi Farace, Daniel Patch, Murthy Yerramilli IDEXX Laboratories Inc, Westbrook, ME, USA Cystatin C is a 13.3 kDa protein that is freely filtered by the kidney. In human medicine it has been shown to be a superior marker for CKD compared to serum creatinine. A small number of studies have investigated the potential of this marker in dogs and cats with mixed results. The goal of this current study was to investigate if serum cystatin C could differentiate healthy animals from cats with CKD; especially those with early disease. A total of 65 feline serum samples were obtained and analyzed for serum creatinine and serum cystatin C. Serum creatinine was determined using the Olympus assay while serum cystatin C was determined using a human commercial plate assay. The commercial assay was validated for cats using a feline recombinant cystatin C. The samples were grouped according to IRIS staging with healthy cats being those with creatinine <1.8 mg/dL; early disease being 1.8–2.8 mg/dL and late disease being >2.8 mg/dL and the serum cystatin C quantiles determined. The range of serum cystatin C for each of three groups was determined. The healthy group ranged from 899 to 1627 ng/mL; the early disease group ranged from 832 to 1949 ng/mL and the late disease group ranged from 1202 to 2475 ng/mL. This data suggests that cats with early disease will be hard to differentiate from either the healthy group or the late stage disease group. Therefore in this study serum cystatin C does not appear to be better than serum creatinine at classifying kidney disease. These findings are in agreement with the work published by Martin et al. (Martin et al. Revue de Medicine Veterinaire, Vol. 153 p. 305–310, 2002) who recommended that cystatin C should not be used to diagnose kidney failure in cats. NU18 PHARMACOKINETIC STUDY OF MEROPENEM IN HEALTHY BEAGLE DOGS ON INTERMITTENT HEMODIALYSIS Hye Young Kim 1, Seok Yeong Byun1, Jong Woo Jeong1, Jong Bok Lee2, Tae Sung Koo1, Kyoung Won Seo1 1Chungnam National University, Deajeon, Korea, 2OraeOrae Animal Hospital, Seoul, Korea Meropenem, a second carbapenem antimicrobial agent with broad spectrum activity, is used in treating sepsis and resistant‐bacterial infections in veterinary medicine. The objective of this study is identifying the pharmacokinetics of meropenem in dogs receiving intermittent hemodialysis (IHD) and determining proper dose for the patients with renal failure receiving IHD. Five healthy beagle dogs were given a single IV dose of 24 mg/kg, of meropenem, and received IHD. IHD initiated at the end of infusion of meropenem, and FX paed was used as artificial kidney. Blood flow rate was 40 mL/min. Dialysate flow rate and ultrafiltration rate were 300 mL/min and 40 mL/h respectively. Blood samples were collected for 24 hours from jugular vein, extracorporeal arterial and venous line. Urine samples and dialysate were collected. Concentrations of meropenem were measured using HPLC/MS/MS determination. Pharmacokinetic data and simulation of dosing regimen were analyzed by WinNolin 4.1 program. Mean peak plasma concentration was 116 ± 37 mg at 15 minute. Mean total body clearance was 347 ± 117 mL/h/kg and mean steady state volume of distribution was 223 ± 67 L/kg. Approximately 69% of administered drug was excreted by urine in unchanged form. Dialysis clearance was 75 ± 23 mL/h/kg and extraction ratio of dialysis was 0.46 ± 0.09. Half‐life of meropenem in dogs with renal failure was prolonged by 2‐fold compared to the dogs with renal failure receiving IHD. In conclusion, 21% meropenem was removed from the dogs through dialysis. Based on the simulated data, additional dose of 24 mg/kg of meropenem after dialysis is suggested. NU19 HIGH THROUGHTPUT IMUUNOASSAY THAT COREELATES TO GOLD STANDARD LIQUID CHROMATOGRAPHY‐MASS SPECTOMETRY (LC‐MS) ASSAY FOR THE CHRONIC KIDNEY DISEASE (CKD) MARKER SYMMETRICALDIMETHYL ARGININE (SDMA) Daniel Patch, Edward Obare, Hongzhi Xie, Giosi Farace, Mahalakshmi Yerramilli, Julie Cross, Troy Goddu, Jason Lee, Murthy Yerramilli IDEXX Laboratories Inc, Westbrook, ME, USA We have recently shown that serum symmetric dimethylarginine (SDMA) is a biomarker for canine and feline CKD, which is earlier than serum creatinine, using a highly sensitive and specific liquid chromatography‐mass spectrometry LC‐MS analytical method which was validated per FDA and CLIA guidelines. This report describes a high throughput immunoassay that has been developed and correlated to the gold standard LC/MS assay. As part of the method validation, performance characteristics including sensitivity, precision, recovery, accuracy, linearity, interferences, carryover, stability, and robustness were investigated. All performance metrics were within established NCSLI guidance. Lower Limit of Quantitation (LLOQ) was found to be 3.8 μg/dL; A five point standard curve ranging from 5 to 100 μg/dL was established with a linearity of R 2 > 0.99; Intra and inter day CV were found to be <6.5% and no significant interference from hemolysis, lipemia or icterus was observed and related compounds such as arginine, monomethylarginine (MMA) and asymmetric dimethylarginine (ADMA) had no significant impact on performance. Correlation to the LC‐MS method was R 2 = 0.98 for dogs and cats using a standard linear fit. In conclusion, we have developed and validated a high throughput immunoassay that correlates to the LC‐MS and accurately quantifies SDMA in biological samples from dogs and cats. NU20 RETROPERITONEAL DELIVERY OF MESENCHYMAL STEM CELLS FOR THE TREATMENT OF FELINE CHRONIC KIDNEY DISEASE Jessica Quimby, Tracy Webb, Elissa Randall, Angela Marolf, Steve Dow Colorado State University, Fort Collins, CO, USA Feline chronic kidney disease (CKD) is characterized by chronic tubulointerstitial nephritis, and inflammation contributes to the progression of renal fibrosis. Mesenchymal stem cell (MSC) therapy may reduce intra‐renal inflammation and fibrosis following systemic delivery and has demonstrated modest effects in feline CKD following intravenous delivery. However, the effects of perinephric MSC delivery have not been evaluated previously. The purpose of the present study was to evaluate the clinical effects of retroperitoneal delivery of MSC in cats with CKD. Six stable CKD cats (3 IRIS Stage II, 3 IRIS Stage III) with no concurrent disease were treated with allogeneic MSC cultured from cryopreserved adipose tissues of SPF research cats. 1 × 107 MSC were injected into the left retroperitoneal space via ultrasound guidance (week 2). CBC, serum biochemistry, urinalysis and UPC were performed at 0, 2, 4, 6, and 10 weeks. Glomerular filtration (GFR) via nuclear scintigraphy was performed at 0 and 6 weeks. Renal function parameters and GFR were analysed with repeated measures ANOVA and paired t‐test respectively. Retroperitoneal delivery of MSC did not elicit adverse effects; no significant change in weight, serum biochemistry parameters, GFR (global, left or right), UPC, or PCV was seen in cats treated with MSC except transient mild increase in serum creatinine at 6 weeks in comparison to 2 weeks. We concluded that ultrasound‐guided perinephric MSC in CKD cats was feasible and appeared safe. Long‐term follow up and larger study numbers will be necessary to determine whether there is a significant beneficial effect on slowing progression of disease. NU21 SAFETY EVALUATION OF THE LIVE BIOTHERATPEUTIC PRODUCT, ASYMPTOMATIC BACTERIURIA E. COLI 2‐12, IN HEALTHY DOGS Jodi Westropp 1, Jane Sykes1, Mary Thompson2, David Klumpp3, Anthony Schaeffer3 1School of Veterinary Medicine, University of California Davis, Davis CA, USA, 2School of Veterinary Science, University of Queensland, Qld, Australia, 3Department of Urology, Northwestern University Feinberg School of Medicine, Chicago, IL, USA Recurrent urinary tract infections (UTIs) are a problem in dogs and women. They are often treated with numerous antimicrobials, which can be associated with bacterial resistance. Asymptomatic bacteriuria E. coli strain 2‐12 (ASB 2‐12) has been investigated as an alternative therapy and has shown significant analgesic properties and reduced pathogenic bacterial reservoirs in experimental mouse models. The aim of this study was to investigate the safety of ASB 2‐12 in six healthy female beagle dogs. Dogs were sedated with 10ug/kg of dexmedetomidine and 0.3 mg/kg butorphanol IM and then a 6‐Fr catheter was inserted into the urinary bladder; 1010 CFU/mL of ASB 2‐12 strain was then instilled (diluted into 10 mls of sterile saline). The dogs were monitored for clinical signs and urine collected by cystocentesis on days 1, 3, and 7‐post instillation. On day 7, the bladder was infused again with ASB 2‐12 and urine collected again on days 1, 3, and 5‐post instillation. Positive aerobic bacterial cultures (E. coli spp.) were obtained from four dogs 1 day after instillation of ASB 2‐12; only one dog had a positive culture 3 days after instillation. After the second infusion, two dogs had positive cultures. All pathogens had the same antimicrobial susceptibility pattern and MALDI‐TOF spectra. Pulse field gel electrophoresis confirmed the pathogens we isolated from beagles by cystocentesis were ASB 2‐12. No dogs exhibited lower urinary tract signs or adverse events. This protocol appeared to be easy and safe and can be investigated for dogs with clinical, recurrent UTI. NM01 THE EFFECTS OF A MEDIUM CHAIN TRIGLYCERIDES DIET TRIAL ON DOGS WITH IDIOPATHIC EPILEPSY Tsz Hong Law 1, Pan Yuanlong2, Brian Zanghi2, Elizabeth Want3, Holger Volk1 1Royal Veterinary College, London, UK, 2Nestlé Purina Research, St Louis, MO, USA, 3Imperial College London, London, UK Epilepsy is the most common chronic neurological disorder in both humans and dogs and despite appropriate antiepileptic drug treatment, approximately one third of humans and dogs with epilepsy continue experiencing seizures. This emphasises the importance of new treatment strategies to improve the welfare of epileptic patients. A 6‐month prospective, randomised, double‐blinded, placebo controlled crossover dietary trial was designed to compare a medium chain triglyceride ketogenic diet (MCTD) to a standardised placebo diet in chronically treated dogs with idiopathic epilepsy, which had ≥3 seizure(s)/3 months prior enrolment. Dogs were fed either MCTD or placebo diet for 3 months followed by a subsequent respective switch of diet for a further 3 months. Seizure frequency, clinical and laboratory data were collected prospectively and evaluated for 21 dogs completing the study. Seizure frequency was significantly lower when dogs were on MCTD (2.31/month, 0–9.89/month) in comparison to placebo diet (2.67/month, 0.33–22.92/month, P = 0.0195); three dogs achieved seizure freedom, seven dogs had ≥50% reduction in seizure frequency, 5 had an overall reduction in seizures and 6 showed no response. Number of seizure days were also significantly less when dogs were on the MCTD (1.63/month, 0–7.58/month) in comparison to placebo diet (1.69/month, 0.33–13.82/month, P = 0.0216). There were no significant changes in serum concentrations of glucose (P = 0.9027), phenobarbital (P = 0.4223), potassium bromide (P = 0.4037) and weight (P = 0.2997) between groups. In conclusion the data shown are in accordance with other studies, showing antiepileptic properties of a MCTD, and provides evidence for the implications of the MCTD as a therapeutic option for treatment in epilepsy. NM02 EVALUATION OF SERUM FATTY ACIDS IN CATS WITH CHRONIC KIDNEY DISEASE Laura Tonkin, Nolie Parnell Purdue University, West Lafayette, IN, USA Nutritional modification of diets for cats with chronic kidney disease (CKD) has been shown to significantly improve survival time. Therapeutic diets for the CKD cat have been modified by restricting both protein and phosphorus while supplementing omega‐3 fatty acids, particularly eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). Supplementation with dietary omega‐3 fatty acids has been shown to be beneficial in dogs but direct evidence in the cat is lacking. Omega‐3 fatty acid deficiency is documented in humans with CKD but omega‐3 fatty acid status is unknown in cats with CKD. The purpose of this study was to evaluate serum fatty acids in cats with CKD. Cats with IRIS stage 2–4 CKD were recruited from multiple clinics for this study and were assigned to one of two groups based on their dietary history for the previous month: Group R cats were fed a prescription renal diet and group M cats were fed an adult maintenance diet. Healthy, age‐matched cats fed an adult maintenance diet served as controls (group C). Serum fatty acid profiles were measured during a fasted state in all cats and serum fatty acid levels were expressed as a percentage of the total fatty acids. There were two hypotheses: 1) cats with CKD on an adult maintenance diet would have lower measured omega‐3 fatty acid concentrations than healthy control cats and 2) there would be no difference in serum omega‐3 fatty acid concentrations between cats with CKD fed a prescription renal diet and healthy controls fed an adult maintenance diet. Parametric data were analysed using a one‐way ANOVA by a commercial statistical software program. Non‐parametric data were analysed using the Kruskal‐Wallis test. A total of 23 cats with CKD were enrolled in this study including 13 cats in group M and 10 cats in group R along with 12 controls. There is no significant difference in age between the three groups (P = 0.847). The mean percentage arachidonic acid is 11.5% in group R, 12.2% in group M, and 12.2% in group C (P = 0.922). The mean percentage EPA is 1.5% in group R, 4.4% in group M, and 2.9% in group C (P = 0.089). The mean percentage DHA is 4.7% in group R, 3.9% in group M, and 4.5% in group C (P = 0.978). There is no significant difference between the three groups for any of the fatty acids analysed. Cats with CKD in this study do not appear to have alterations in fasted serum fatty acid levels compared to age‐matched healthy controls. In particular, there is not a difference in the serum omega‐3 fatty acids, EPA and DHA. Feeding a therapeutic veterinary diet formulated for renal disease in this study does not appear to result in significant differences in serum EPA and DHA levels compared to maintenance commercial diets. NM03 EFFECTS OF PROTEIN: CARBOHYDRATE RATIOS ON WEIGHT LOSS IN OVERWEIGHT CATS Yuanlong Pan, Hui Xu, Sandeep Bhatnagar Nestle Purina Research, St. Louis, MO, USA Up to 50% of cats in developed countries between 5 and 10 years of age are either overweight or obese. There is a need for effective weight management diets to promote weight loss in overweight cats. The objective of this study was to compare the effects of two diets with comparable levels of dietary fat but differing in protein‐carbohydrate ratios of 4.44 (HPC) and 1.03 (LPC) on weight loss, fat loss, and body composition in overweight cats. Thirty overweight cats were divided into two equal groups based on measured maintenance energy requirements (MER), body weight, and % body fat, then randomly assigned to one of the two diets. Cats in both groups were fed 75% of their MER for 6 months. Daily food intake, weekly body weight, and monthly quantitative magnetic resonance for body composition were performed during the trial. After 6 months, there was no significant difference in loss of body weight between two groups, but % body fat loss was doubled in the HPC cats (−7.60 ± 0.75%) compared with the LPC cats (−3.58 ± 0.95%). The mean % Body fat of the HPC cats at the end of the study (20.40 ± 1.39%) was significantly (P = 0.002) lower than that at the baseline (27.86 ± 1.39%), while the mean % body fat of the LPC cats at the end of study (24.12 ± 1.60%) was not significantly (P = 0.102) different from baseline (28.11 ± 1.60%). The results show that the HPC diet was more effective than the LPC diet in promoting the loss of body fat. NM04 PREDICTORS OF CORE TEMPERATURE DURING INTERVAL EXERCISE IN DOGS Patrick Robbins 1, Brian Zanghi2, Meghan Ramos1, Cynthia Otto1 1University of Pennsylvania, Philadelphia, PA, USA, 2Nestle Purina Research, St. Louis, MO, USA To evaluate the impact of environmental and dog‐specific factors influencing core body temperature (CBT) following exercise, exercise conditioned dogs (n = 12; 7M/5F; age 7–42 months), owned and in training as detection dogs through the University of Pennsylvania, were evaluated on 6 separate occasions over 3 weeks. Exercise training included 5‐minute searching, 5‐minute agility and 10‐minute ball retrieve, with 5‐minute rest between bouts. Exercise period was 30‐minute or until one designated trainer determined that the dog was not performing adequately (e.g. panting excessively, seeking shade or reluctance to retrieve). Pre‐ and postexercise heart rate, CBT (ingested temperature monitor; CoreTemp), and venous blood for lactate (Lactate Scout) and blood gas (ISTAT) were collected. A backward stepwise regression was used to evaluate factors predicting CBT. Combining all exercise occasions, median exercise time was 27.5 minutes (range 20–0 minute), median outdoor temperature was 84°F (range 71–94°F), and median humidity was 48% (range 36–76%). Mean postexercise CBT was 105.9°F (range 102.4–108.8°F). No dog showed signs of heat stress requiring veterinary intervention. Regression analysis determined that the dependent variable peak CBT could be predicted from a linear combination of the independent variables: average duration of exercise (P = 0.019), peak heart rate (beats/min) (P = 0.029), postexercise blood pH (P = 0.027), lactate (P = 0.011), and pCO2 (P = 0.052). Model not significantly influenced by individual dog, dog‐age, outdoor temperature, humidity, change from baseline for body weight, venous pO2, or hematocrit. This study suggests that the influence of temperature and humidity are less important in peak body temperature than the metabolic parameters associated with exertion. NM05 SAFETY OF THE GHRELIN AGONIST, CAPROMORELIN, ADMINISTERED DAILY TO BEAGLE DOGS FOR 1 YEAR Bill Zollers1, Linda Rhodes 1, Lesley Rausch‐Derra1, Glenda Armintrout1, Margie Bell2 1Aratana Therapeutics, Inc., Kansas City, KS, USA, 2Clindata Services, Fort Collins, CO, USA Capromorelin, a ghrelin agonist, is under development for use in dogs for appetite stimulation. Capromorelin, dosed orally, has been shown to stimulate food intake, weight gain, increases in serum growth hormone (GH) and insulin‐like growth factor‐1 (IGF‐1) in laboratory dogs (Zollers et al. 2014) and to increase owner appetite assessment scores and body weight in a clinical field study of dogs with inappetence (Zollers and Rhodes, 2014). This study evaluated the safety of capromorelin in dogs. Thirty‐two Beagle dogs were randomly assigned to four groups (n = 4/sex/group) and received 0, 0.39, 9.2 or 52.4 mg/kg capromorelin (dose calculated per the tartrate salt) once daily by oral gavage for 12 consecutive months. Safety parameters were evaluated including clinical signs, food consumption, body weight, clinical pathology, urinalysis, ophthalmology, electrocardiogram and vital signs. Capromorelin, GH and IGF‐1 were measured in plasma samples collected at various timepoints during the study. At the end of the dosing period, dogs were euthanized and necropsied. Selected organs were weighed and tissue samples collected and processed for microscopic examination. Capromorelin at daily doses up to 52.4 mg/kg (approximately 17.5X the active ingredient concentration in the proposed clinical dose for dogs) for 12 months resulted in minimal toxicity with no negative effects on food consumption, body weight, ophthalmic exams, vital signs or gross pathology. Clinical signs related to capromorelin were limited to salivation and loose stool noted sporadically during the study. Occasional episodes of emesis were observed and considered unrelated to treatment. The higher dose treatments were associated with an increased incidence of reddening/swollen paws. One dog in the 52.4 mg/kg group died when capromorelin was accidentally delivered into the respiratory tract during gavage as confirmed by necropsy. Electrocardiogram data indicated slight increases in the PRQ interval in the 52.4 mg/kg and 9.2 mg/kg groups 1–2 hours following dosing. However, no histological lesions were observed in the heart. In general, clinical pathology and urinalysis parameters were within normal ranges or lacked a consistent dose/time relationship. However, there were slight decreases in red blood cells, hemoglobin and hematocrit that did not appear to be clinically significant in the 52.4 mg/kg group. Cholesterol, HDL and alkaline phosphatase serum levels were statistically significantly increased in the 52.4 mg/kg group compared to placebo and individual animal values tended to be at the high end or slightly above the normal reference range. Increased absolute liver weights in dogs treated with the 52.4 mg/kg dose of capromorelin were noted. A slight increase in hepatocellular cytoplasmic vacuolation was seen in all capromorelin treated groups. Capromorelin plasma levels increased with increasing dose, were similar on Days 90, 181 and 349 indicating no accumulation of drug and there were no gender‐related differences. GH plasma levels increased modestly as expected on Days 1, 170 and 351 following capromorelin treatment. The GH response was controlled by a physiological negative feedback so that the magnitude of the GH response to capromorelin treatment lessened over time. As expected, IGF‐1 plasma levels increased following capromorelin treatment and increased levels were sustained over time as evidenced in plasma collected on Days ‐1, 1, 7, 14, 21, 28, 62, 121, 170 and 351. The results of this study demonstrated that capromorelin was well‐tolerated in dogs dosed up to 52.4 mg/kg for 12 months. Further, this study indicates an expected wide safety margin for capromorelin as the high dose is approximately 17.5X the proposed clinical dose. NM06 PALATABILITY AND CLINICAL EFFECTS OF AN ORAL RECUPERATION FLUID DURING THE RECOVERY OF DOGS WITH PARVOVIRAL ENTERITIS Reut Tenne, Elena Contreras, David Twedt, Michael Lappin, Lauren Sullivan Colorado State University, Fort Collins, CO, USA Dogs infected by canine parvovirus (CPV) can develop severe enteritis that requires supportive care until voluntary food and water consumption returns. An oral recuperation fluid (ORF, Viyo Recuperation™) containing essential nutrients and prebiotics designed for restoration and support of the gastrointestinal and immune system, may assist in the overall recovery from CPV. The hypotheses of the study were that naturally infected CPV dogs would prefer the ORF to water and that dogs consuming the ORF would have a more rapid return to voluntary appetite and improved caloric intake during the initial recovery period compared to dogs consuming water. A total of 28 dogs with clinical signs consistent with CPV but no prior treatments were enrolled after detecting CPV antigen in feces. Dogs were excluded if voluntary food or water intake was present at the time of study enrollment. At hospital admission (Day 0), dogs were randomized to either an ORF or water group. The designated fluid was offered q12h throughout hospitalization, followed by offering the opposite fluid 1 hour later if the designated fluid was refused. A standardized treatment protocol including intravenous fluids, cefoxitin (22 mg/kg IV q6h), and maropitant (1 mg/kg IV q24h) was used for all dogs. Beginning on Day 2, all dogs were offered a gastrointestinal diet q8h. Recorded parameters included fluid preference (ORF vs water vs neither), days until return of appetite, and percentage of prescribed resting energy requirement (%RER) consumed within the first 24 hours of appetite return. A higher percentage of CPV dogs preferred the ORF as their designated fluid [6/15 (40%)] when compared to those offered water as their designated fluid [4/13 (31%), P = 0.71]. For those dogs that refused their designated fluid, 5/9 (56%) of the ORF dogs ultimately preferred water and 2/9 (22%) of the water dogs preferred the ORF (P = 0.33). Dogs that consumed the ORF demonstrated a more rapid return to voluntary appetite [median 1.5 days (range 1–3), P = 0.01], compared to those that consumed water [median 4.25 days (range 1.5–5.5)] or neither fluid [median 2 days (range 1.5–5.5)]. Additionally, those dogs consuming the ORF ingested a higher %RER [median 100% (range 61–100%), P = 0.03], compared to those that preferred water [median 19% (range 9–100%)] or neither fluid [median 37% (range 3–100%)]. This study suggests that some CPV dogs will voluntarily consume this ORF during the recovery phase of their illness, and that consumption of the ORF may foster a more rapid return of voluntary appetite as well as improved caloric intake. Additional studies are warranted to further assess additional effects of this ORF during recovery from CPV. NM07 EFFECT OF SINGLE‐DOSE TREATMENT WITH SYNTHETIC GHRELIN ON CANINE ANOREXIC PATIENTS: A PRELIMINARY STUDY Daigo Azakami1, Eri Onozawa 1, Katsumi Ishioka1, Masahiro Miyabe1, Takeshi Hanada2, Yujiro Hayashi2, Noboru Murakami3, Toshinori Sako1 1Nippon Veterinary and Life Science University, Musashino, Tokyo, Japan, 2Asubio Pharma Co, Ltd, Kobe, Hyogo, Japan, 3University of Miyazaki, Miyazaki, Miyazaki, Japan Ghrelin, a peptide hormone produced in the stomach, is known to have physiological and pharmacological actions on stimulating appetite in humans. In veterinary medicine, physiological effects of ghrelin in healthy dogs have been established; however, clinical application of ghrelin has not ever been reported. Here, to investigate the potential clinical use of ghrelin in dogs, we performed a preliminary study of single‐dose treatment with synthetic canine ghrelin on clinically anorexic dogs. Ten client‐owned dogs which had admitted to Veterinary Medical Teaching Hospital of Nippon Veterinary and Life Science University were used. All dogs were confirmed poor appetite by veterinarians during hospitalization. Degree of appetite was scored by the amount of food intake. Appetite scores were evaluated after serving meals, then we administered single‐dose of synthesized canine ghrelin (5 μg/kg intravenously or 50 μg/kg subcutaneously) followed by re‐evaluation of appetite scores. All dogs showed appetite score 0 before ghrelin administration. Single‐dose ghrelin treatment was effective in eight cases (80%), in which six dogs improved their appetite scores from 0 to 1, and 2 dogs from 0 to 2. No appetite change was observed in remaining two dogs. Adverse effects and behavioral changes were not observed in all treated dogs. In conclusion, the appetite‐stimulatory effect of single‐dose treatment with ghrelin seen in this preliminary study suggests that ghrelin could be effective in the treatment of canine anorexic patients. NM08 SAFETY OF THE GHRELIN AGONIST, CAPROMORELIN, ADMINISTERED DAILY TO CATS FOR 91 DAYS AT AN ORAL DOSE OF 6 MG/KG Bill Zollers, Julie Allen, Chelsey Kennedy, Linda Rhodes Aratana Therapeutics, Inc., Kansas City, KS, USA Capromorelin, a ghrelin agonist, is under development for use in cats. This controlled, non‐GLP, study evaluated the safety of capromorelin in laboratory cats. Two treatment groups were included, following an 11‐day acclimation phase, with 4 adult cats receiving an oral placebo solution (Group 1) and 8 adult cats receiving an oral capromorelin solution at 6 mg/kg (Group 2) for 91 consecutive days. All cats were weighed on Days −11, −1, 1, 14, 30, 59, 75 and 91. Physical examinations were performed on Days −11, 1, 30, 59 and 91 and animals were observed at least once daily throughout the study period. On Days 1 through 91, animals were administered a flavored oral solution of placebo or capromorelin once daily via syringe. Hematology, serum chemistry (including fructosamine) and urinalyses were conducted on Days ‐5 (acclimation) and Days 30, 59 and 91, approximately 1 hour following dose administration. On Days 1, 30, 59 and 91, blood samples were collected prior to daily dose administration and 8 hours post‐dose and serum IGF‐1 and GH levels measured. Beginning 1 hour post‐dose, all cats were offered food during a 6‐hour period. Variables of interest included weight gain/loss, food consumption, clinical pathology (hematology, chemistry, and urinalysis), and serum analysis of IGF‐1 and GH. During the study, no abnormalities were detected upon physical examination and the most common clinical observations were ptyalism, lip smacking, emesis, and head‐shaking. These findings were seen immediately post‐dose and usually resolved within five minutes. No other abnormal general health or clinical observations were noted. Cats treated with capromorelin gained weight whereas placebo control cats maintained a consistent body weight; this weight gain trend was expected based on previous pilot study data. The capromorelin treated cats trended toward increased food consumption compared to placebo treated cats. Certainly, treatment with capromorelin at a dose of 6 mg/kg did not negatively affect food consumption in the cat. In general, the statistically significant differences for most clinical pathology safety parameters were not clinically relevant. Serum glucose levels were increased in the capromorelin group compared to the placebo group and there was an increasing trend over time; however, no glucose values exceeded normal limits. Additionally, there were no capromorelin‐related serum fructosamine level trends and no fructosamine values exceeded normal limits in treated cats. Finally, urine glucose levels were negative except trace readings in one capromorelin treated cat and one placebo treated cat. BUN and creatinine levels were similar for cats in both treatment groups and all values were within the normal reference ranges. Since renal disease is prevalent in cats, the lack of change in the renal parameters might suggest that capromorelin may be used in these patients. This will need to be confirmed in safety studies in these patients. Serum GH levels increased in cats treated with capromorelin compared to placebo controls as expected. The increase of GH was pronounced on Day 1 of capromorelin treatment; however, this increase was attenuated as daily capromorelin treatment progressed, indicating intact negative feedback of IGF‐1 on GH secretion. In contrast, the increase of IGF‐1 became sustained as daily treatment of capromorelin continued and this peaked at Day 30 and decreased further over time. In conclusion, capromorelin was well‐tolerated in cats dosed orally at 6 mg/kg for 91 consecutive days. Further, this study indicates an expected adequate safety margin for capromorelin as the dose used was 3× the proposed clinical dose. NM09 CAPROMORELIN, AN ORALLY ACTIVE GHRELIN AGONIST, CAUSED SUSTAINED INCREASES IN IGF‐1, INCREASED FOOD INTAKE AND BODY WEIGHT IN CATS Bill Zollers, Julie Allen, Chelsey Kennedy, Linda Rhodes Aratana Therapeutics, Inc., Kansas City, KS, USA Capromorelin is a ghrelin agonist that, when dosed orally in laboratory dogs, has been shown to stimulate food intake, weight gain, growth hormone (GH) and insulin‐like growth factor‐1 (IGF‐1) (Zollers et al. 2014). This randomized, placebo‐controlled study was performed to determine if capromorelin would also increase food consumption, promote weight gain and increase serum IGF‐1 levels in laboratory cats. Thirty‐two cats (16 neutered males, 16 females) were divided into four treatment groups including four animals of each sex per group. The cats were acclimated to the study environment for 10 days prior to study start. All cats were orally dosed with placebo (Group 1) or capromorelin (30 mg/mL oral solution) at 1 mg/kg (Group 2), 2 mg/kg (Group 3) or 3 mg/kg (Group 4), for 21 days, with dosing starting on Day 1. Physical examinations were performed on Day −10 and general health observations made daily. Body weights were evaluated on Days –10, −8, −1, 1, 8, 15 and 22 days. Cats were fed approximately 1 hour after dosing; 300 g of commercial dry cat food was provided for 5 hours and then removed and the amount of food consumed (g) calculated per cat. On Days 1, 14 and 21, blood samples were collected prior to daily dose administration and 8 hours post‐dose and serum processed for measurement of IGF‐1 levels. On Day 12, two study animals (one male each from Groups 1 and 4) were removed for reasons unrelated to the study and were not included in the data analysis. All treatment groups were observed to have increased mean food consumption from baseline [average of Days −3, −2 and −1] to study period [average of Day 1 to Day 21]. The placebo cats (Group 1) had a mean food intake increase of 10.83% over baseline, while the three capromorelin groups (Groups 2, 3 and 4) had mean food intake increases over baseline of 25.32%, 45.67% and 29.59%, respectively with only Group 3 showing a statistically significant (P < 0.01) increase in food consumption when compared to the placebo group. Mean body weights for the capromorelin treatment groups increased during the 21‐day exposure period. Groups 2, 3 and 4 had increases of 5.41%, 6.61% and 3.92% respectively whereas the placebo group lost a small amount of weight (minus 1.11%). Statistically significant differences from placebo (P < 0.05) in mean percent body weight change were observed for Group 3 at all 3 time points measured with increases of 3.23%, 5.97% and 6.61% on Days 8, 15 and 22, respectively. Additionally, Group 2 had mean percent body weight changes that were statistically significantly increased (P < 0.05) when compared to placebo at Day 15 (4.24%) and Day 22 (5.41%) but not at Day 8. The interaction of treatment by sex was not statistically significant (P > 0.05) for either food consumption or weight gain, indicating that the effect of treatment was similar in male and female cats. In Group 1, IGF‐1 levels remained at baseline levels throughout the study. On Day 1, group mean serum IGF‐1 levels increased from 0 to 8‐hours post‐dose by 46.69%, 29.38% and 36.77% for Groups 2, 3 and 4, respectively. On Day 14, IGF‐1 levels were sustained at a higher level at hour 0. Therefore, the group mean increase in serum IGF‐1 levels was lessened from 0 to 8‐hours post‐dose (17.27%, 9.00% and 18.56% for Groups 2, 3 and 4, respectively). On Day 21, the trend of a smaller IGF‐1 response continued as group mean IGF‐1 levels increased from 0 to 8‐hours post‐dose by 9.81%, 2.68% and 5.17% for Groups 2, 3 and 4, respectively. The smaller increases of IGF‐1 levels following capromorelin treatment on Day 14 and Day 21 were due to the fact that by Day 14 there was a sustained elevation in IGF‐1 resulting from repeated daily capromorelin treatment. In conclusion, capromorelin increased food intake, promoted weight gain and caused sustained increases in IGF‐1 in laboratory cats. NM10 EFFECT OF ENTEROCOCCUS FAECIUM STRAIN SF68 ON THE GASTROINTESTINAL CLINICAL SIGNS OF CATS ADMINISTERED AMOXICILLIN‐CLAVULANATE Camille Torres‐Henderson, Stacie Summers, Amber Caress, Francisco Olea‐Popelka, Michael Lappin Colorado State University, Fort Collins, CO, USA Amoxicillin‐clavulanate is frequently prescribed by veterinarians for a wide variety of feline conditions including bite wound abscesses and upper respiratory infections. As the drug has a broad spectrum against many bacteria, including anaerobes, diarrhea, vomiting, and loss of appetite can occur when it is prescribed to cats. One objective of the study was to quantify the frequency of gastrointestinal signs induced by amoxicillin‐clavulanate. The second objective was to determine if administration of Enterococcus faecium strain SF68 (SF68) could ameliorate these side effects as this probiotic has been shown to be effective in the management of diarrhea in cats. In this double blinded, placebo‐controlled study, a total of 34 young adult, healthy, mixed sex, purpose bred cats were randomized into two rooms by body condition. The cats were caged so that clinical findings, including appetite, could be monitored. Cats were monitored daily for appetite, attitude, hydration, vomiting and consistency of feces (FS) using the following scale; 7 = watery puddles; 6 = texture but no shape; 5 = moist piles; 4 = moist log shape; 3 = normal. After a 10 day equilibration period, cats that developed a fecal score of >4 more than twice were blocked from the study as those predisposed to stress diarrhea. All cats were administered amoxicillin‐clavulanate (Clavamox®; Zoetis) at 62.5 mg/cat PO twice a day for 7 days. Two hours prior to the antibiotic administration, 1/4 can of canned food mixed with either 5 × 108 E. faecium SF68 (equivalent to the commercial product; 13 cats) or the palatability enhancer (14 cats) used in the commercial product. The palatability enhancer, SF68, and clinical scoring were continued for an additional 5 days after stopping the antibiotic. While vomiting was not noted during the equilibration period, during the antibiotic administration period, 7 SF68 cats and 6 placebo cats vomited at least once. However, the total number of vomiting episodes during the antibiotic administration period were not significantly different between the groups and vomiting resolved in all cats after stopping the antibiotic. During the antibiotic period, while a stool with a FS of ≥5 was detected at least once in 11 SF68 cats and 12 placebo cats, a FS of ≥ 6 was detected in only 8 SF68 cats (61.5%) in contrast to 12 placebo cats (85.7%) and a FS of 7 was only detected in 3 placebo cats (21.4%). When percentages of samples with different FS were compared between the groups in the antibiotic administration period, it was shown that cats in the placebo group (61 of 95 samples; (68.4%) had a greater number (P = 0.006) of samples with a FS ≥5 than the SF68 group (42 of 96 samples; 43.8%) and samples with a FS = 7 were only detected in the placebo group (7 of 95 samples; P = 0.007). Percentages of cats with samples with FS ≥5 decreased in both groups in the post‐antibiotic period (SF68 = 15 of 40 samples [37.5%]; Placebo cats = 23 of 42 samples [54.7%]) but the difference between groups was not significant (P = 0.1283). Overall, attitude and hydration was normal for all cats over the course of the study. From this pilot study we concluded that amoxicillin‐clavulanate can cause gastrointestinal symptoms. Diarrhea episodes (FS ≥ 5), were more likely to be in the placebo group, suggesting the concurrent administration of SF68 may decrease the severity of amoxicillin‐clavulanate diarrhea in some cats. NM11 WEIGHT LOSS PROGRAM ASSOCIATED WITH VIBRATING PLATFORM IN OBESE DOGS AND OVERWEIGHT Mariana Bail Afonso1, Bruna Duarte Pacheco1, Gisele Vieira Sechi1, Rayane Sol Amaral Silva Sgarbossa1, Sthefany Buba Lucina1, Andressa Thaís Roncoski1, Stefan Renê Schwanz2, Claudia Turra Pimpão1, Bianca Hasselmann Cottar1, Mariane Miranda Gonçalves1, Ana Paula Sarraff Lopes1, Carolina Zaghi Cavalcante 1 1PUCPR, Curitiba/Paraná, Brazil, 2PhysicalPet, Curitiba/Paraná, Brazil Obesity is defined as excessive accumulation of adipose tissue in the body, enough to affect the physiological functions and predispose various diseases. The obesity treatment is based on reducing calorie intake and increasing energy consumption. The vibrating platform has been investigated as a biophysical type of exercises which provides an alternative intervention for obesity in people. This study aimed to evaluate the efficiency of vibrating platform for weight control in obese and overweight dogs and the effects on the reduction of body condition and laboratory findings. Participants were 8 obese and overweight dogs divided into two groups (group 1‐ animals submitted to a hypocaloric diet and group 2 – animals submitted to a hypocaloric diet and two weekly workouts in vibrating platform). The study lasted 2 months and were performed weighing, morphometric and laboratory assessments and the measurement of subcutaneous fat deposition by ultrasound. There was a reduction in body weight, morphometric and the subcutaneous fat measurements of all dogs, with significant differences between weights and also when comparing the groups. Regarding the laboratory abnormalities, there were observed reduction in serum cholesterol level in both groups. The vibration platform is an important tool for treating obesity, since the use of this method allowed the reduction of weight, body condition and promoted beneficial laboratory results. In veterinary medicine several studies are needed with a larger number of animals to confirm the efficacy of the vibratory platform in obese dogs. OT01 EVALUATION OF SERUM ALPHA1‐PROTEINASE INHIBITOR CONCENTRATIONS IN DOGS WITH SYSTEMIC INFLAMMATORY RESPONSE SYNDROME OR SEPSIS Romy Heilmann 1, Niels Grützner2, James Barr1, Brittany Thames1, Jan Suchodolski1, Jörg Steiner1 1Texas A&M University, College Station, TX, USA, 2Vetsuisse University Bern, Bern, BE, Switzerland Alpha1‐proteinase inhibitor, which is mainly synthesized by hepatocytes, is a major serum proteinase inhibitor that protects the body from the detrimental effects of neutrophil proteinases. In contrast to humans, canine α1‐PI (cα1‐PI) has not convincingly been demonstrated to be an acute‐phase protein. This may be due to its species‐specific physiologic and pathophysiologic role. However, serum cα1‐PI concentrations were increased in a group of dogs with pancreatitis, a disease process that is often associated with an activation of the systemic inflammatory cascade. Dogs with protein‐losing diseases on the other hand were shown to have decreased serum cα1‐PI concentrations. Evaluation of cα1‐PI in dogs with systemic inflammatory response syndrome (SIRS) would further our understanding of its pathophysiologic role. Therefore, this study investigated serum cα1‐PI concentrations in dogs with SIRS or sepsis. Serum samples from dogs with non‐infectious SIRS (n = 8) or sepsis (n = 11), collected on the day of enrollment (day 1) and on the following days (i.e., days 2 and 3), and from a group of healthy control dogs (n = 50) were used. These samples were surplus material from a previous study. The criteria by Hauptmann et al. (1997) were used for a diagnosis of SIRS, and the source of infection was confirmed in septic dogs via histologic, microbiologic, and/or cytological examination. Concentrations of cα1‐PI were measured in all samples using an established in‐house RIA, and were tested among the groups of dogs using non‐parametric 2‐group/ multiple‐group comparisons. Serum cα1‐PI concentrations were significantly lower in dogs with SIRS (P = 0.0006) or sepsis (P = 0.0002) than in the healthy control dogs. While on day 1 serum cα1‐PI concentrations did not differ between dogs with SIRS and those with sepsis (P = 0.6201), dogs with sepsis had significantly lower serum cα1‐PI concentrations on days 2 (P = 0.0297) and 3 (P = 0.0360) than dogs with non‐infectious SIRS. Serum cα1‐PI concentrations were not significantly different between survivors (n = 13) and non‐survivors (n = 6; P > 0.05). In this study, serum cα1‐PI concentrations were significantly lower in dogs with SIRS or sepsis than in healthy control dogs, and septic patients had lower serum cα1‐PI concentrations on days 2 and 3 than dogs with non‐infectious SIRS. These results would argue for a role of cα1‐PI as a negative acute‐phase protein in dogs, and may suggest that the protein‐proteinase inhibitor balance is altered in dogs with systemic inflammation. Further studies to evaluate the role of cα1‐PI in the acute‐phase response as well as its utility as a biomarker for the diagnosis and differentiation of dogs with SIRS or sepsis are warranted. OT02 IN VITRO DETERMINATION OF THE UTILITY OF THERAPEUTIC PLASMA EXCHANGE (TPE) IN THE TREATMENT OF IBUPROFEN TOXICITY Stuart Walton 1, Kirk Ryan1, Mark Acierno1, Jennifer Davis2 1Louisiana State University, Baton Rouge, LA, USA, 2North Carolina State University, Raleigh, NC, USA In dogs, ibuprofen intoxication is common, occurring through either administration by owners or dietary indiscretion. Intoxication can result in acute gastrointestinal signs and kidney injury. Historically, it has been treated with supportive care. No published studies have reported use of TPE to reduce serum ibuprofen concentrations. We hypothesized that TPE could be used as an adjunctive or alternative treatment for ibuprofen toxicity. A measured amount of injectable ibuprofen was apportioned in 500 mL of canine whole blood (blood volume equivalent to a 5 kg dog) and incubated for 30 minutes at 38.5 °C. TPE was performed over 2 hours to achieve filtration of 1.5 plasma volumes. Heparinized blood (2 mL) and effluent (1 mL) samples were collected at T = 0, 0.5, 1, and 2 hours. All samples were centrifuged (500 g for 5 minute) and the serum and RBC fractions were separated and collected. Ibuprofen concentrations were determined by high performance liquid chromatography. Percentage removal and sieving coefficients were calculated to assess ibuprofen clearance from blood. A sharp rise in sieving coefficient was noted within 30 minutes of TPE with a plateau attained at 30 minutes (median sieving coefficient = 0.95782; range = 0.804–1.1). Calculated median serum ibuprofen concentration reduction was 31.8% (range 30.2% – 48.2%). Calculated median RBC ibuprofen concentration reduction was 26.7% (range 6.5–36.0%). Serum ibuprofen concentration reduction was less than expected (65 – 70%), which we attribute to sequestration of drug into red blood cells. Nonetheless, the results suggest TPE may be used to rapidly reduce ibuprofen concentration following acute toxic ingestion. OT03 PROGNOSTIC INDICATORS IN DOGS WITH SEPTIC PERITONITIS: 54 CASES (2004–2014) Jeffrey Mitchell 1, Gwendolyn Levine1, Micah Bishop2, James Barr1, Medora Pashmakova1 1Texas A&M Veterinary Teaching Hospital, College Station, TX, USA, 2Animal Specialty Hospital of Florida, Naples, FL, USA Septic peritonitis (SP) remains a significant cause of mortality in dogs. Reported survival rates are variable and few reliable prognostic indicators exist, many of which are extrapolated from human literature. The aim of this study was to identify prognostic factors on admission in dogs with SP. Medical records of all dogs hospitalized for SP between 2004 and 2014 were reviewed. Inclusion criteria required a CBC, serum chemistry, and NOVA at admission. Diagnosis of SP was based on intracellular bacteria and/or positive culture of abdominal fluid. APPLE scores were calculated for all patients. Data was checked for normality using a Shapiro‐Wilk and D'Agostino & Pearson omnibus tests. Parametric and non‐parametric data were compared using an unpaired t‐test and a Mann‐Whitney test, respectively. Fisher's exact test and Odds ratios were constructed to compare proportions between variables with categorical data. Alpha was set at 0.05 Fifty‐four dogs met inclusion criteria and had a survival rate of 64.8%. Gastrointestinal origin of SP comprised 62.9% of cases. E.coli and Enterococcus species comprised 60% of bacterial isolates. There was no difference in the median circulating neutrophil count (12,193 vs. 7140/μL; P = 0.19), total WBC count (16.9 vs. 13.8 × 109/L; P = 0.20), time to surgery (8.0 vs. 6.0 hour; P = 0.30), and APPLE (full) score (35.0 vs 37.5; P = 0.06) between survivors and non‐survivors. Exact testing revealed no association between the presence of Dohle bodies (P = 0.53), toxic changes (P = 0.24), band neutrophils (P = 0.08), ionized hypocalcemia (<1.23 mmol/L; P = 0.72), or ionized hypomagnesemia (<0.38 mmol/L; (P = 0.26). Appropriate initial antibiotic therapy was not associated with survival (P = 0.24); however, dogs that received appropriate antibiotics were 2.36 times more likely to survive. Initial antibiotic choice was only appropriate in 39% of cases. The mean APPLE (fast) score was the only factor associated with death (24.5 vs 29.3; P < 0.01). In this population of dogs with SP, tests commonly associated with severe disease, timing of surgery, and appropriateness of initial antibiotic choice were not predictive of case outcome. Interestingly, initial antibiotic choice was only correct in a minority of cases. Prospective studies incorporating antibiotic therapy that addresses the most common isolates in dogs with SP are warranted to determine if a survival benefit exists. OT04 DOES DO NOT RESUSCITATE (DNR) ALWAYS MEAN DNR? EXPLORING DNR ORDERS IN VETERINARY MEDICINE Keriann Cabral 1, Elizabeth Rozanski1, Gareth Buckley2, Howard Cabral3 1Tufts University, North Grafton, MA, USA, 2University of Florida, Gainesville, FL, USA, 3Boston College, Boston, MA, USA Cardiopulmonary resuscitation is often performed in hospitalized patients, typically based upon the prior establishment of a “CPR” code based upon a conversation between the owner and a veterinary team member. Animals may be placed on a do‐not‐resuscitate (DNR) code status. However, in human medicine there often exists confusion as to what a DNR order does and does not mean. The purpose of this study was to assess whether differences exist among populations of veterinarians, veterinary technicians, and veterinary students regarding responses to clinical scenarios involving do‐not‐resuscitate (DNR) orders. A hyperlink to the online survey was distributed via email to variety of veterinary listserves. The survey consisted of questions regarding respondent characteristics, followed by five case examples where interpretation of the DNR order could be questioned, including four hypothetical cases where a patient with a pre‐existing DNR suffered an unexpected cardiopulmonary arrest and one case where a terminally ill dog's owner requested CPR. Data were evaluated using the chi‐square or Fisher's exact test, as appropriate with a P‐value of <0.05 considered significant. Seven hundred and thirty‐three individuals, including 533 veterinarians, 66 veterinary technicians, 134 veterinary students responded. The majority of the veterinarians were either ACVIM or ACVECC diplomates. Significant differences in individual questions were found based on gender, frequency of CPR events in practice, years of experience, and geographic location. Men, veterinarians that have been in practice longer, and respondents who participate in multiple CPR events per year were more likely to perform CPR in the face of a pre‐existing DNR order where an extenuating circumstance existed. CPR codes are not routinely suspended in pets undergoing anesthesia and code status may not be re‐evaluated during hospitalization, even in the face of changing patient status. This study supports that there is the potential for confusion surrounding CPR and DNR in veterinary patients. The definition of “DNR” is not universal, and may result in different clinical interpretations. The veterinary profession would benefit from more education and discussion regarding this topic OT05 IMPACT OF STATISTICAL METHODS ON ANALYTE REFERENCE INTERVAL ACCURACY Kevin Le Boedec University of Illinois College of Veterinary Medicine, IL, USA The aim of this study was to evaluate the impact of the choice of statistical methods on reference interval accuracy. Gaussian, lognormal, asymmetric, and bimodal analyte distributions were studied. Gaussian distribution was derived from normal canine glycemia distribution, and the other distributions were derived from serum ALT distribution in normal dogs. For each distribution, a population of 10 000 000 values was simulated using the statistical software R, and samples of n = 250, n = 120, n = 60, and n = 30 values were randomly extracted 50 times from the population. Reference intervals obtained by six different statistical methods were compared to the 2.5% and 97.5% quantile values of the populations. Although statistically significant differences in accuracy were detected among statistical methods, most of them were not considered clinically relevant. The only relevant inaccuracies were found when parametric or robust methods were used on untransformed data with non‐Gaussian population distributions. Specificity of the Shapiro‐Wilk test to predict population Gaussianity declined as the sample size decreased (92%, 78%, 67%, and 45% for n = 250, n = 120, n = 60, and n = 30 respectively). Apart from parametric and robust methods on untransformed data, the choice of the statistical method has minimal clinically‐relevant impact on reference interval accuracy. Parametric and robust methods on untransformed data are only appropriate when an analyte population distribution is Gaussian, which cannot always be ascertained from the sample distribution. Consequently, their use should be discouraged. During reference interval construction, accuracy likely relies more on non‐statistical steps such as choice of which animals to include, rather than on the statistical analysis. OT06 ACUTE PHASE REACTION AND CARDIAC BIOMARKERS IN FELINE NON‐CARDIAC DISEASES Inigo Sanz, Paul Wotton, Jorge Prieto‐Ramos, Valentina Palermo, David Eckersall, Tim D. H. Parkin, Anne T. French Glasgow University, Glasgow, UK There is increasing awareness that myocardial injury occurs in primary non‐cardiac diseases and often goes unrecognized. Myocardial injury is associated with increased morbidity and death in both human and canine patients with non‐cardiac diseases. This prospective study assessed the presence of myocardial damage and characterized the acute phase response in a population of cats with systemic disease. Two specific cardiac biomarkers, high sensitivity cardiac troponin I (hs‐cTnI) and amino terminal prohormone brain natriuretic peptide (NTpro‐BNP) and four acute phase proteins (APP), C‐reactive protein (CRP),α₁‐acid glycoprotein (AGP), serum amyloid A (SAA) and haptoglobin (Hp) were measured in 57 cats with systemic disease. The population was divided in four groups: 11 hyperthyroid (group H), 17 neoplastic (group N), 17 inflammatory (group I) and 12 miscellaneous conditions (group M). Significant differences between groups for NT‐proBNP, hs‐cTnI, AGP and CRP were found. Group H had significantly higher concentrations of NTpro‐BNP than any of the other groups. Group H had significantly higher hs‐cTnI than groups N and I but not group M. Group H had significantly lower AGP concentrations than groups I and N and lower CRP than group I. In multivariate analysis both hs‐cTnI and NT‐proBNP were associated with azotaemia but APP were not. Anaemia was associated with SAA and AGP but not with hs‐cTnI, NT‐proBNP, SAA or Hp. Cats with non‐cardiac diseases experience myocardial damage, which does not appear to be related to systemic inflammation. AGP appears to be the most sensitive APP in these cats. OT07 IDIOPATHIC IMMUNE‐MEDIATED POLYARTHRITIS IN DOGS: 67 CASES (2006–2014) Katherine Nadolny, Leah Cohn University of Missouri, Columbia, MO, USA To describe presentation and treatment of dogs with immune‐mediated polyarthritis (IMPA), the University of Missouri computerized records were searched. Only dogs with neutrophilic joint fluid were included, while dogs with infection, breed‐related arthropathy, systemic lupus, or erosive disease were excluded. Most of the 67 identified dogs (87%) were <9 years of age (mean 5.8, median 6). Neither sex predominated (37 female, 30 male). Cases included small dogs <10 kg (22), medium 10–25 kg (21), and large dogs >25 kg (24). Few (n = 9) were diagnosed in winter. Owners recognized lameness in 46/67 dogs; 1 (n = 11), 2 (24), 3 (4), or all 4 (7) limbs were affected, with hindlimbs most affected (P = 0.04). There was no correlation between severity of joint inflammation, temperature, peripheral neutrophilia, or hospitalization rate. Most had uncomplicated IMPA (type I; n = 51), while type II reactive IMPA (10), enteropathic type III (4), and malignancy‐associated type IV (2) IMPA were uncommon. Hospitalization was more likely (P = 0.005) for dogs with type II‐IV. Treatment included prednisone in 64/65; azathioprine (23), mycophenolate mofetil (10), cyclosporine (8), and leflunomide (7) were also used. Tetracyclines were given to 62. Outcome information was available for 41 dogs; 71% achieved clinical remission, 68% cytological remission, and 36% relapsed. Remission was less likely in dogs with type II‐IV than type I IMPA (P = 0.008). Dogs of any size may develop IMPA, usually in warm months. Owners often but not always recognized lameness. A good response to therapy was common, but remission was less likely with type II‐IV IMPA. OT08 EVALUATION OF THE EFFECTS OF TIME, TEMPERATURE, AND STORAGE CONDITIONS ON IN VITRO LACTATE CONCENTRATIONS Colin Reich, Jonathan Bach, Julie Walker University of Wisconsin, School of Veterinary Medicine, Madison, WI, USA The purpose of the study was to determine the effects of various storage conditions on canine lactate concentrations. Prospectively 12 mL blood was collected form 30 healthy dogs. The blood was promptly divided into storage aliquots that were each used once. Storage conditions consisted of a gold standard (anaerobic chilled), and three comparator groups (aerobic chilled, anaerobic room temperature (RT), aerobic RT). Time points of analysis were T0, T15‐25, T30‐40, T60‐70 and T120‐130 minute in the gold standard group; and T15‐25, T30‐40, T60‐70 and T120‐130 minute in the comparator groups. Nonparametric statistics were used. To assess clinical significance all values were then compared to the gold standard by calculating absolute and percent differences versus T0. A mean change >20% from T0 values was considered to be clinically unacceptable. At all time points, there were statistically significant differences in lactate concentrations within the anaerobic chilled samples (gold standard group). The aerobic chilled samples did not have statistically significant change in lactate when compared to anaerobic chilled at the same time points. The anaerobic RT samples had statistically significant increases at all time points after 15–25 minute. The aerobic RT samples had statistically significant changes in lactate at all time points. T0 T15‐25 T30‐40 T60‐70 T120‐130 Anaerobic chilled (gold standard) 1.0 (0.7–2.3) 1.1 (0.6–2.5)# 1.2 (0.8–2.6)# 1.2 (0.9–2.6)# 1.2 (0.7–2.8)# Aerobic chilled n/a 1.2 (0.9–2.6) 1.3 (0.9–2.7) 1.2 (1.0–2.8) 1.4 (1.1–3.0) Anaerobic RT n/a 1.3 (0.9–2.8) 1.5 (1.1–2.9)* 1.8 (0.7–3.4)* 2.4 (2.0–4.3)* Aerobic RT n/a 1.3 (1.1–2.8)* 1.6 (1.1–2.9)* 2.0 (0.4–3.5)* 2.6 (2.1–4.7)* Results are median & range; lactate values in mmol/L. #Statistically significant versus T0. *Statistically significant versus anaerobic chilled (gold standard) at same time point. The percent change from the initial T0 lactate was then calculated at all time points & storage conditions. T0 T15‐25 T30‐40 T60‐70 T120‐130 Anaerobic chilled (gold standard)   8.4 ± 12.3% 12.1 ± 11.5% 19.5 ± 12.5% 25.3 ± 17.9% Aerobic chilled n/a 16.0 ± 8.6% 23.7 ± 14.2% 27.2 ± 11.9% 37.0 ± 15.5% Anaerobic RT n/a 28.1 ± 10.7% 44.9 ± 16.9% 73.9 ± 28.0% 134.5 ± 43.2% Aerobic RT n/a 36.3 ± 14.6% 54.0 ± 21.7% 86.4 ± 39.7% 161.4 ± 58.3% Results support analyzing lactate samples immediately. Findings indicate that time, temperature and storage conditions all cause significant changes in lactate. If lactate analysis is delayed, anaerobic chilled samples should be analyzed within 70 minutes, and aerobic chilled samples should be analyzed within 25 minutes. Aerobic RT & anaerobic RT sample storage should be avoided. Clinicians should be aware that pre‐analytical factors including storage time and sampling handling may cause changes in lactate values. OT09 ACUTE PHASE PROTEIN CONCENTRATIONS IN CATS WITH URINARY TRACT OBSTRUCTION Elizabeth Schmidt 1, Milica Filipovic2, Jelena Francuski2, Nenad Andric2, Luciano Barbosa1, Mary Waterston3, David Eckersall3 1São Paulo State University, Botucatu/SP, Brazil, 2University of Belgrade, Belgrade, Serbia, 3University of Glasgow, Glasgow/UK, UK The aim of this study was to determine if cats with urinary obstruction have an inflammatory acute phase response. For that, the concentrations of haptoglobin (HP) and serum amyloid A (SAA) were determined in cats with clinically urinary tract obstruction at the time of diagnosis. Thirty‐four cats were included in the study. The animals were between 6 months to 15 years old. Cats were assigned to groups according to clinical signs, hemogram and urinalysis. 16 cats showed clinical signs of urinary tract obstruction and 18 cats were clinically healthy control cats. The HP serum concentrations were measured via haemoglobin binding assay modified for automation on an ABX Pentra Analyser (Horiba Medical, Montpellier, France). SAA concentrations were determined using LAT Eiken Chemical Co. (Japan) modified for automation on ABX Pentra Analyser. The HP concentrations in the control group ranged from 2.02 to 4.12 g/L (median 3.08 g/L) and SAA concentrations in this group were all <5 mg/L. In the clinically obstructed cats the HP concentrations ranged from 2.5 to 6.04 g/L (median 4.04 g/L), SAA concentrations ranged from <5 to 228.5 mg/L (median <5 mg/L), white blood cells (WBC) ranged from 7100 to 24,200/μL (median 12,195/μL) and neutrophils ranged from 5760 to 18,030/μL (median 11,320/μL). All these parameters were significantly higher (P < 0.05) when compared to the clinically healthy cats. The increases seen in HP and SAA values were accompanied by increases in WBC and neutrophils in the clinically obstructed cats demonstrating an acute phase response in cats with urinary tract obstruction. However, there was no correlation between HP and WBC indicating that these analytes respond with differing dynamics to urinary obstruction and that measurement of acute phase protein could add value to the diagnostic regime for this condition in cats. OT10 MONOCLONAL ANTIBODIES SPECIFIC TO SERUM AMILOID A FROM DIFFERENT SPECIES Karina Seferian 1, Valentina Podoprigora1, Stanislav Kozlovsky1, Olga Kolosova1, Alexander Kogan2, Alexey Katrukha1 1HyTest OY, Turku, Finland, 2School of Biology, Moscow State University, Moscow, Russia Serum amyloid A (SAA) is a major acute phase protein in many species. SAA has been reported to be a sensitive biomarker of inflammation in dogs, cats, and horses. The high level of identity between SAA from different species allows measurement of canine, feline and equine SAA using the same immunoassay. The aim of this study was to select antibodies with cross‐reactivity for canine, feline and equine SAA from the panel of murine monoclonal antibodies (mAbs) developed against either human or canine SAA. The study also included development of recombinant species‐specific calibrators for SAA immunoassays. Recombinant canine, feline and equine SAA were expressed in E. coli. Cross‐reactivity of antibodies (41 different mAbs) was tested with recombinant and endogenous SAAs. Among tested antibodies, 6 mAbs recognized recombinant and endogenous SAAs from all three species in direct ELISA and Western blotting studies. These 6 mAbs were further evaluated in sandwich ELISA. In sandwich ELISA strong non‐specific binding of SAAs to the polystyrene surface (Costar 96‐well plates) was observed. The same problem has earlier been reported for human SAA as well. Optimization of the plate blocking protocol and washing and dilution buffers helped to reduce the non‐specific binding of SAA to the plate surface. Various mAb combinations were tested with serum samples from healthy animals and animals with acute inflammation using an optimized sandwich ELISA protocol. One mAb combination was selected which recognized SAA from canine, feline and equine serum with high sensitivity. In conclusion, monoclonal antibodies described in the present study can be used for the development of an immunoassay applicable for the SAA measurements in blood of three major companion animal species. P01 EFFECTS OF ASPIRIN AND CYCLOSPORINE ON CANINE T‐CELL CYTOKINES AND REGULATORY T‐CELLS Todd Archer1, Claire Fellman 1, John Stokes1, Evangel Kummari1, John Thomason1, Jenica Haraschak2, Lesya Pinchuk1, Robert Wills1, Andrew Mackin1 1Mississippi State University, Mississippi State, MS, USA, 2North Carolina State University, Raleigh, NC, USA Cyclosporine and aspirin are routinely used to treat immune‐mediated hemolytic anemia (IMHA) in dogs. Cyclosporine is a potent immunosuppressive agent that exerts its effects by decreasing cytokine production by lymphocytes, especially T‐cell production of IL‐2 and IFN‐γ. Recent research in dogs has shown that high dose cyclosporine reliably reduces expression of IL‐2 and IFN‐γ, while lower doses of cyclosporine affect cytokine expression more variably. Low‐dose aspirin is commonly used concurrently with immunosuppressive agents such as cyclosporine in dogs with IMHA to inhibit platelet function, since such patients are hypercoagulable and prone to life‐threatening thromboembolic disease. Regulatory T‐cells (Tregs) are necessary both for maintenance of a healthy immune system and for prevention of autoimmune pathology. Abnormalities in Treg number or function have been implicated in multiple autoimmune, infectious and allergic diseases. In rodents and in humans, aspirin and cyclosporine have both been shown to affect Treg number and function and T‐cell cytokine production. In dogs, however, it has not yet been determined if either drug influences Treg numbers, nor has it been determined if concurrent aspirin alters the immunosuppressive effects of cyclosporine. Our study was designed to determine the effects of cyclosporine, low‐dose aspirin, high‐dose aspirin and concurrent cyclosporine and low‐dose aspirin on canine Treg numbers and T‐cell cytokine expression. In a cross‐over study design, seven healthy young adult dogs were given either oral cyclosporine at an immunosuppressive dose (10 mg/kg Q12 hours), oral low‐dose aspirin (1 mg/kg Q24 hours), oral high‐dose aspirin (10 mg/kg Q12 hours), or low‐dose aspirin combined with cyclosporine, each for 8 days, with a washout of at least 4 weeks after each cross‐over. Blood was drawn prior to and on the last day of drug treatment for flow cytometric analysis. Heparinized whole blood was activated with PMA and ionomycin for evaluation of CD3+ T‐cell expression of IL‐2 and IFN‐γ. Percent CD4+/CD25+/FoxP3+ Tregs was evaluated in freshly drawn EDTA‐anticoagulated whole blood. The difference between pre‐ and post‐treatment values for each group, as well as the difference between treatment groups, was evaluated. Paired t‐tests comparing baseline and post‐treatment values within a treatment group for each outcome were performed. Differences between groups were assessed using mixed effects models. P ≤ 0.05 was considered significant. Cyclosporine significantly decreased IL‐2 and IFN‐γ expression when used alone or in combination with low‐dose aspirin. High‐dose aspirin, but not low‐dose aspirin, significantly decreased IL‐2 expression, although the decrease was not as marked as that seen with cyclosporine alone or combined with aspirin. Neither low‐dose nor high‐dose aspirin significantly affected IFN‐γ expression. Cyclosporine alone, unlike low‐dose or high‐dose aspirin alone, significantly decreased Treg numbers. Cyclosporine combined with low‐dose aspirin was also associated with a decrease in Treg numbers, although the magnitude of the reduction was not enough to achieve statistical significance (P = 0.06). Low‐dose aspirin given with cyclosporine creates the same degree of T‐cell cytokine suppression as does cyclosporine alone, suggesting that the two drugs can be used concurrently without altering cyclosporine's immunosuppressive effects. High dose aspirin also suppresses T‐cell IL‐2 expression, although suppression is not as marked as that seen with cyclosporine. Cyclosporine also significantly decreases Treg numbers in dogs. Since low Treg numbers have been associated with immune‐mediated diseases, our findings suggest that the effects of cyclosporine on the canine immune system may be more complex than immunosuppression alone. P02 MEASUREMENT OF PLASMA AMLODIPINE CONCENTRATION USING LC/MS/MS AT THE POINT OF BLOOD PRESSURE CONTROL IN HYPERTENSIVE CATS Esther Bijsmans 1, Rosanne Jepson1, Henry Pratt2, Hattie Syme1, Jonathan Elliott1, Ludovic Pelligand1 1Royal Veterinary College, London, UK, 2ABS Laboratories Ltd., Welwyn Garden City, UK Amlodipine is the first‐line antihypertensive agent used in cats, and a dose of 0.625–1.25 mg/cat/day is recommended. In human medicine the recommendation exists to initiate a more aggressive antihypertensive plan in patients with severe hypertension. In cats, monotherapy with amlodipine is standard practice with dose adjustment used to achieve control. The aim of this study was to investigate whether the dose of amlodipine required to control blood pressure adequately is dependent on individual variation in drug pharmacokinetics or pharmacodynamics. Plasma samples from hypertensive cats treated with amlodipine at two first opinion practices were identified retrospectively. Hypertension was diagnosed based on a systolic blood pressure (SBP) ≥170 mmHg on 2 consecutive visits or SBP ≥160 mmHg with concurrent evidence of hypertensive retinopathy. All cats were started on 0.625 mg amlodipine q24h and invited back for a blood pressure re‐check 1–2 weeks later. Adequate control was defined as SBP <160 mmHg and if SBP was ≥160 mmHg the dose was increased to 1.25 mg/cat/day, with another re‐check 1–2 weeks later. Blood samples were collected at the first visit SBP control was adequate. Plasma amlodipine concentration was determined using high‐performance liquid chromatography with tandem mass spectrometric (LC/MS/MS) detection. A linear mixed effects model with Bonferroni post hoc comparison was used to compare SBP at hypertensive visit and SBP at first controlled visit between cats on 0.625 mg (A) and 1.25 mg (B). Plasma amlodipine concentration was compared between A and B using a Mann‐Whitney U‐test. Results are presented as mean ± sd or median [IQR]. Univariable linear regression models were performed with absolute change in SBP with amlodipine treatment as the dependent variable and plasma amlodipine concentration, dose in mg/kg, time since last dosing, treatment time in days, and plasma potassium and creatinine concentration as explanatory variables. Variables were log transformed if needed to meet normality criteria and variables significant at the P < 0.1 level were included in the multivariable model. Plasma amlodipine concentration was determined for 97 cats with adequate SBP control (A: 57 cats; B: 40 cats). SBP at the hypertensive visit differed significantly between groups (A: 185.9 ± 14.6 mmHg; B: 208.5 ± 24.4 mmHg, P < 0.001), but comparable post‐treatment SBP was achieved (A: 142.4 ± 12.1 mmHg; B: 144.3 ± 10.3 mmHg, n.s.). Plasma amlodipine concentration was twice as high in cats on 1.25 mg compared to those receiving 0.625 mg (70.5 [48.9, 98.7] ng/mL vs. 33.3 [25.5, 54.9] ng/mL), as was the dose in mg/kg that the cats were receiving (A: 0.17 ± 0.04 mg/kg; B: 0.33 ± 0.09 mg/kg). Dose in mg/kg, plasma amlodipine concentration, and potassium were associated with the absolute change in SBP in the univariable analysis (P < 0.1). The plasma amlodipine concentration and the dose in mg/kg remained positively and significantly associated with decrease in SBP in the multivariable model (P < 0.05). This study showed that the decrease in SBP associated with amlodipine treatment was independently and positively associated with the plasma amlodipine concentration. Cats that required a higher dose to control their SBP below 160 mmHg had a significantly higher SBP at diagnosis of hypertension and therefore required a greater decrease in SBP in order to achieve normotensive control. Administration of the higher dose needed to achieve the target SBP was associated with proportionately higher plasma amlodipine concentrations, suggesting that the difference in dose of amlodipine required is unlikely to be due to a difference in individual cat pharmacokinetics or owner compliance. Based on these results it could be suggested that cats with a SBP≥200 mmHg should be started on 1.25 mg amlodipine daily. Further studies are warranted to determine whether differences exist in the pathophysiological mechanisms underlying hypertension in the more severely hypertensive group. P03 THE EFFECT OF CONCURRENT AND DELAYED SUCRALFATE ADMINISTRATION ON THE RELATIVE BIOAVAILABILITY OF FLUOROQUINOLONES IN GREYHOUNDS Sarah Guess, Kate Kukanich, Ellen Heinrich, Butch Kukanich Kansas State University, Manhattan, KS, USA Sucralfate is a gastroprotectant that has been reported to impair absorption of certain antimicrobials. Our objective was to determine if co‐administration of sucralfate and a fluoroquinolone impairs fluoroquinolone bioavailability in dogs and to determine the effect of a 2‐hour delay of sucralfate on fluoroquinolone pharmacokinetics. Five Greyhounds were incorporated into a randomized crossover design and administered either ciprofloxacin (25 mg/kg PO) or enrofloxacin (5 mg/kg PO) alone, concurrent sucralfate (1 g oral suspension PO) with ciprofloxacin or enrofloxacin, or sucralfate administered 2 hours after either fluoroquinolone. Fluoroquinolone concentrations were evaluated with liquid chromatography/mass spectrometry. Area under the curve (AUC), time to maximum plasma concentration (TMAX), and maximum plasma concentration (CMAX) were compared between groups. Dogs had variable absorption of ciprofloxacin (AUC range: 5.52–22.47 hour μg/mL) compared with enrofloxacin (3.86–7.50 hour μg/mL). Relative bioavailability (%F) of ciprofloxacin was 48% when administered concurrently with sucralfate (with one dog at 8%), and improved to 87% when sucralfate was delayed by 2 hours. In contrast, no significant difference in absorption was demonstrated based on AUC or CMAX when enrofloxacin was administered concurrently with sucralfate as compared with enrofloxacin alone, and %F for concurrent administration was 104%. This study confirmed that oral bioavailability of ciprofloxacin is variable, which could lead to treatment failures or drug toxicity in some treated dogs. When sucralfate and ciprofloxacin are both indicated, a 2 hour delay should be considered to allow improved bioavailability. On the contrary, no drug interaction was documented for concurrent sucralfate and enrofloxacin, and concurrent administration may improve compliance. P04 PHARMACODYNAMICS OF TRANSDERMAL MIRTAZAPINE IN HEALTHY CLIENT‐OWNED CATS Kellyi Benson, Lara Barron, Paul Lunghofer, Ryan Hansen, Luke Wittenburg, Dan Gustafson, Jessica Quimby Colorado State University, Fort Collins, CO, USA Transdermal mirtazapine (TM) would be beneficial for ill cats intolerant of pill administration. The purpose of this study was to evaluate the pharmacodynamics of transdermal mirtazapine (TM) in client‐owned cats. Ten client‐owned cats (1–7 years) with unremarkable CBC, chemistry and urinalyses were enrolled in a randomized double‐blind placebo‐controlled three‐way crossover study. Treatments consisted of 6 days of aural application of 7.5 mg/dose TM or placebo gel by owner, and 1.87 mg mirtazapine once orally in clinic (washout period: 7 days). Owners documented appetite, rate of food ingestion, begging, activity and vocalization daily in the home environment. On day 6 cats were observed for 8 hours in hospital; food consumed, activity and vocalization was documented hourly and trough and peak serum levels were obtained. Serum alanine aminotransferase (ALT) levels were measured after TM administration. Serum mirtazapine and compounded gel concentrations were measured using liquid chromatography/tandem mass spectrometry. Cats receiving TM for 6 days had a significant increase in appetite (P = 0.0078), rate of food ingestion (P = 0.0078), activity level (P = 0.0078), begging (P = 0.0078) and vocalization (P = 0.0039) when compared to placebo. On day 6, cats receiving TM had a significant increase in food consumption in hospital (P = 0.0435) when compared to placebo but vocalization and activity were not statistically different between treatments. Gel concentrations were 105 ± 11% of target dose. Average ± SD trough and peak serum mirtazapine concentrations after 6 days of TM administration were 21 ± 14.9 ng/mL and 35 ± 18.2 ng/mL respectively. Serum ALT remained normal after 6 days of TM administration. Average ± SD peak serum mirtazapine concentrations after one oral dose of 1.87 mg mirtazapine were 65.1 ± 33.5 ng/mL. Daily 7.5 mg TM administration in healthy cats resulted in a significant increase in appetite in comparison to placebo with serum concentration within the same range of single oral 1.87 mg mirtazapine administration. P05 COMPARISON OF TOLFENAMIC ACID AND ROBENACOXIB FOR POSTOPERATIVE ANALGESIA IN CATS AFTER OVARIOHYSTERECTOMY Naris Thengchaisri, Prangtip Phuwapattanachart Kasetsart University, Bangkok, Thailand Evidence suggested that nonsteroidal anti‐inflammatory drugs (NSAIDs) are helpful for post‐operative pain control in animals. The anti‐inflammatory properties of NSAIDs involve an inhibition of cyclooxygenase (COX) isoforms including COX‐1 and COX‐2. Nontheless, the efficacy of non‐selective COX inhibitor (tolfenamic acid) and selective COX‐2 inhibitor (robenacoxib) for post‐operative pain control in cats has not been compared. The purpose of a present study was to study the postoperative analgesic effects of tolfenamic acid and robenacoxib in cats undergoing ovariohysterectomy. Twenty four cats were divided into three groups including control (no medicine), tolfenamic (4 mg/kg/day) and robenacoxib (1 mg/kg/day) groups. NSAIDs were oral administered at 1 hour before induction with propofol 4–6 mg/kg IV. Anesthesia was then maintained with isoflurane inhalation. Buccal mucosal bleeding time (BMBT) was measured before surgery in each cat. Colorado pain scores (0–4) were recorded in blinded fashion before induction (base line) and 2, 5, 8, 24, 30, and 48 hours following the recovery. The opiate analgesia (morphine 0.2 mg/kg IM) was provided in control group based on clinical signs. The experimental protocol was approved by the Kasetsart University Animal Care and Use Committee. The average BMBT (mean ± SE) from control, tolfenamic acid and robenacoxib were not different (59 ± 15, 67 ± 20, 53 ± 14 second, respectively; P‐value > 0.05). The average baseline of Colorado pain scores from control, tolfenamic acid and robenacoxib groups were not different (0.4 ± 0.2, 0.3 ± 0.2, 0.3 ± 0.2, respectively; P‐value > 0.05). The average Colorado pain scores recordings to 48 hours postoperation from tolfenamic acid (0.3 ± 0.2) and robenacoxib (0.3 ± 0.1) groups were significantly lower than that of control group (0.6 ± 0.2; P‐value < 0.01), nonetheless, there was no significant difference comparing between tolfenamic acid and robenacoxib groups (P‐value = 0.86). In conclusion, preoperative administration of either tolfenamic acid or robenacoxib significantly reduces post‐operative pain in cats after ovariohysterectomy without affecting the bleeding time. A significant difference could not be detected for post‐operative pain control in cats comparing between non‐selective COX inhibitor (tolfenamic acid) and selective COX‐2 inhibitor (robenacoxib). P06 SAFETY AND EFFICACY OF INTRAVENOUS TRANEXAMIC ACID AS AN EMETIC IN CANINE PATIENTS Hitoshi Kakiuchi 1, Atsushi Nakamura2, Asako Kawarai‐Shimamura3, Atsushi Ogawa3, Kensuke Orito1 1School of Veterinary Medicine, Azabu University, Kanagawa, Japan, 2TRVA Animal Medical Center, Tokyo Jonan Regional Veterinary Medicine Promotional Association, Tokyo, Japan, 3Anicom Holdings, Inc., Tokyo, Japan In Japan, intravenous tranexamic acid (TA) has been used empirically to induce vomiting after accidental ingestion in dogs. We found that neurokinin‐1 receptors may play a pivotal role in TA‐induced emesis and revealed that TA induces emesis in a dose‐dependent manner in dogs. Understanding the efficacy and safety of TA‐induced emesis is critical for careful and precise management of accidental ingestion. The purpose of this study was to evaluate the emetic action and side effects of TA in a clinical setting. The study population comprised 137 dogs. A single intravenous TA dose was administered. If needed, an additional dose was administered. Data were collected on a case report form, which included age, breed, body weight, gender, anamnesis, clinical signs, time to emesis after intravenous TA, duration of emesis, number of incidences of emesis, and side effects. TA induced vomiting in 116 of 137 dogs (84.7%). The median time to onset of vomiting was 116.5 seconds (range 26–370 seconds), the median duration of emesis was 151.5 seconds (range 30–780 seconds), and the median number of incidences of emesis was 2 (range 1–8). In the second and third administrations, TA induced vomiting in 34 of 43 dogs (79.1%) and 3 of 4 dogs (75%), respectively. The side effects included one tonic‐clonic convulsion and hemostatic disorder in different dogs, both of which received medical care later and recovered. TA induced emesis with a high probability following both single‐dose and dose‐escalating administration. The side effects were considered mild and self‐limiting. P07 PHARMACOKINETICS AND PHARMACODYNAMICS OF TRANSDERMAL MIRTAZAPINE IN PURPOSE‐BRED CATS Kellyi Benson, Lara Barron, Paul Lunghofer, Ryan Hansen, Luke Wittenburg, Dan Gustafson, Jessica Quimby Colorado State University, Fort Collins, CO, USA Transdermal application of the appetite stimulant mirtazapine is an attractive option for cats resistant to pilling. The purpose of this study was to evaluate the pharmacodynamic and pharmacokinetics (PK) of transdermal mirtazapine (TM). Three purpose‐bred cats with unremarkable CBC, serum biochemistry and urinalyses were utilized. The following procedures were performed (minimum 7 day wash out): A) 3.75 mg and 7.5 mg TM once aurally with 48 hour PK sampling (Serum samples were obtained via jugular catheter at 0, 0.5, 1, 2, 5, 9, 12, 24, 36 and 48 hours), B) 7.5 mg TM and placebo daily aurally for six consecutive days then 48 hour PK C) 1.87 mg mirtazapine orally once with limited serum sampling at 1, 4 and 8 hours. Kibble weight was recorded every 2 hours during B and C. Serum mirtazapine and compounded gel concentrations were measured using liquid chromatography/tandem mass spectrometry. Non‐compartmental pharmacokinetic modeling was performed. Oral dose AUC values were estimated using a limited sampling model developed from previous study data. Daily 7.5 mg TM resulted in significantly more food consumed than placebo at 2, 4, 6 and 10 hours (P < 0.0001). Gel concentrations varied by 35–97% of target dose. 3.5 mg TM once 7.5 mg TM once 1.87 mg oral once 7.5 mg TM daily Placebo daily AUC0–48 hour ng/mL median (range) 108.8 (99.9–117.6) 231.3 (192–269.8) 104.0 (101.5–189.6) 538.9 (365.8–542.3) NP Average % food eaten ± SD NP NP 73 ± 12.2 93 ± 4.5 54 ± 18.6 TM achieves adequate serum concentrations in cats to stimulate appetite despite inconsistencies in compounding. P08 SAFETY AND TOXICOKINETIC PROFILES IN CATS ADMINISTERED GRAPIPRANT, A SELECTIVE PROSTAGLANDIN‐RECEPTOR ANTAGONIST Lesley Rausch‐Derra, Linda Rhodes Aratana Therapeutics, Inc., Kansas City, KS, USA Grapiprant is a new analgesic that selectively blocks the EP4 receptor of prostaglandin E2 and which may be useful in the treatment of osteoarthritis in dogs and cats. The objective of this study was to evaluate the feline safety and toxicokinetic profiles of grapiprant. Twenty four domestic short‐hair cats were randomly assigned to receive either placebo or 3, 9, or 15 mg/kg grapiprant in a capsule formulation (three males and three females per group) once daily for 28 days; all cats received their assigned medications as per protocol. Food consumption and behavior were observed daily, body weight measured weekly, and clinical pathology analyses on blood run on Days ‐7, 14, and 25, and urinalysis run on Days ‐7, and 25. Serial blood samples for toxicokinetic analyses were obtained following the Day 0 and Day 27 doses. Full necropsies and histopathological evaluations were performed following humane euthanasia on Day 28. Grapiprant was well tolerated, and no adverse effects were noted at daily doses up to 15 mg/kg for 28 days. All animals appeared normal throughout the study. Grapiprant caused no deaths or drug‐related effects on body weight, food consumption, clinical pathology, organ weight, gross pathology, or histopathology. Grapiprant rapidly reached peak serum concentrations and maintained substantial levels throughout the study. Grapiprant exposure varied with dose, but the 9 mg/kg and the 15 mg/kg groups showed similar mean AUClast values. By Day 27, maximum grapiprant serum concentrations ranged from 683 ng/mL to 4950 ng/mL, and these were reached by 1–4 hours after administration. Mean half‐lives on Day 27 ranged from approximately 3–14 hours, with a median value of approximately 5–6 hours. Grapiprant in a capsule formulation was well tolerated when given for 28 days at serum exposures up to 4950 ng/mL. P09 PHARMACOKINETIC COMPARISON OF BODY SURFACE AREA AND BODY WEIGHT BASED PREDNISOLONE DOSE ON SMALL AND LARGE BREED DOGS Hye Bin Wang, Sung Min Kim, Tae Heon Kim, Tae Sung Koo, Min Kyu Kim, Kyoung Won Seo, Kun Ho Song Chungnam National University, Daejeon, Korea Prednisolone (PDS) is commonly used oral glucocorticoids on dogs. However, total PDS dose could be high on large breed dogs. The aim of this study is comparing the pharmacokinetics of PDS dependent on body surface area (BSA) and body weight (BW) on small and large breed dogs. In group A, five beagle dogs were given 2 mg/kg of oral PDS twice a day for a week. In group B, five large breed dogs were given the same dose of oral PDS twice a day for a week. In group C, five large breed dogs were given the dose of 40 mg/m² of oral PDS twice a day for a week. On the 7th day, blood samples were collected 12 hour after oral administration. Pharmacokinetic curve was analyzed by HPLC/MS/MS determination. In all groups, the time to reach maximum concentration (Tmax) was similar by being 0.8–1.1 hour. In group A, the peak plasma concentration (Cmax) and area under curve (AUC) were 777.20 ng/mL and 2074.59 ng h/mL. In group B, Cmax and AUC were 1918.00 ng/mL and 5597.79 ng h/mL, which were extremely higher than group A and C. In group C, Cmax and AUC were 427.80 ng/mL and 1428.79 ng h/mL, which were notably lower than group B and slightly lower than group A. Pharmacokinetics of same BW‐based PDS dose were found different in small and large breed dogs. Therefore clinicians are suggested to adjust the dose by considering PK difference of PDS based on BSA or BW. R01 MAROPITANT (CERENIA®) AMELIORATES CLINICAL SIGNS BUT NOT INFLAMMATION ASSOCIATED WITH CANINE CHRONIC BRONCHITIS Megan Grobman, Carol Reinero Comparative Internal Medicine Laboratory and Department of Veterinary Medicine and Surgery College of Veterinary Medicine, University of Missouri, Columbia, MO, USA Chronic bronchitis (CB) is a common disease of dogs resulting in a chronic cough (≥ 2 months) and inflammation of the lower airways. Adverse effects and risk of secondary infection associated with lifelong steroid administration prompt investigation into alternative therapies. Neurogenic pathways mediated by tachykinins binding neurokinin (NK) 1 receptors may contribute to airway inflammation. Maropitant (Cerenia®), a NK‐1 receptor antagonist, has been used for CB based on anecdotal improvement without supporting evidence. We hypothesized that maropitant would blunt clinical signs and airway inflammation associated with CB. Client‐owned dogs (n = 7) with cough >2 months, thoracic radiographic evidence of airway disease and sterile airway inflammation (>7% non‐degenerate neutrophils/eosinophils) on bronchoalveolar lavage (BAL) were enrolled. Dogs received maropitant (2 mg/kg) every 48 hours for 14 days. Study endpoints included client perception of clinical signs (surveys at baseline and 14 days and a visual analogue scale (VAS) at baseline, 7 and 14 days), and objective assessment of BALF % neutrophils and eosinophils (baseline and 14 days). A One‐Way Repeated Measures ANOVA (VAS) and Mann Whitney Rank Sum tests (BAL cytology) were used. P < 0.05 was considered significant. Pilot data showed maropitant reduced clinical signs based on client surveys (mild improvement: 3/7, marked improvement: 4/7) and significantly decreased VAS scores (P = 0.002). No difference in BALF % neutrophils or % eosinophils were noted with treatment (P = 0.165, P = 0.710). Subjective clinical improvement could reflect antitussive properties of maropitant or a placebo effect. Preliminary results suggest a lack of reduction of airway inflammation, making maropitant unsuitable for treatment of canine CB. R02 COMPARISON OF FATIGUE CHARACTERISTICS OF TWO COMMERCIALLY AVAILABLE VETERINARY TRACHEAL STENTS Michal Crha 1, Alois Necas2, Milan Dvorak2 1Ceitec Vfu University of Veterinary and Pharmaceutical Sciences Brno, Brno, Czech Republic, 2Small Animal Clinic University of Veterinary and Pharmaceutical Sciences Brno, Brno, Czech Republic Stent placement has been shown to decrease clinical signs and improve quality of life in dogs with tracheal collapse when medical management fails. Stent fractures occasionally occur which complicates post‐operative care, raises cost, and decreases operator and owner satisfaction. Coughing likely plays a causative role but variations in clinical signs, disease severity, and operator preferences makes comparing stents based on clinical outcomes problematic. The aim of this study was to evaluate the fatigue properties of two types of tracheal stents in a controlled environment that simulated chronic coughing. Twelve tracheal stents of similar size from two manufacturers were tested. A fatigue‐testing fixture was used to simulate the force of a cough using parallel compression plates. Stents were pre‐compressed by 20% of their nominal diameter to simulate in vivo conditions and then stressed at a frequency of 50 Hz for the equivalent of 60 coughs per hour for 1 year. Stents completing the test intact were further cycled to fracture. Stents were inspected every 50,000 cycles. Data was compared using Fisher's exact and Wilcoxon rank sum tests. A P value < 0.05 was considered significant. Group A stents (Infiniti Medical) were significantly more durable, with 6/6 completing the trial intact compared to 0/6 in group B (Dextronix) (P = 0.002). The mean number of cycles to fracture in group A was 741 1667 compared to 83 333 in group B (P = 0.0034). Although in vivo stent performance is multi factorial, stent A appears better able to tolerate the stresses of chronic cough. R03 META‐ANALYSIS OF THE ASSOCIATION BETWEEN MYCOPLASMA SPECIES AND FELINE RESPIRATORY TRACT DISEASES Kevin Le Boedec University of Illinois College of Veterinary Medicine, IL, USA Mycoplasma is considered a primary pathogen in the lower respiratory tract (LRT) of cats, but its pathogenic role in the upper respiratory tract (URT) is debated, as it could be part of the normal flora. Seven databases were searched to identify studies comparing the rate of isolation of Mycoplasma species from the respiratory tract of symptomatic and asymptomatic cats. The quality of the evidence was assessed using a modified Newcastle‐Ottawa scale. The meta‐analysis was stratified by URT and LRT diseases. Of the 2002 reports retrieved initially, twelve met inclusion criteria. The quality of the evidence was moderate (4.1/8). Mycoplasma was not significantly more prevalent in the LRT of symptomatic cats compared to controls’ (odds ratio (OR): 1.56 [0.51, 4.76]). However, samples from the URT of symptomatic cats were twice as likely to be positive for Mycoplasma as healthy cats’ (OR: 2.02 [1.29, 3.16]). Not all the studies on URT showed this association (heterogeneity: 51%, P = 0.03). Heterogeneity could come from differences in populations (shelter cats, OR: 1.79 [0.90, 3.57], heterogeneity: 69%, P = 0.006; versus non‐shelter cats, OR: 2.48 [1.54, 3.99], heterogeneity: 0%, P = 0.79) and choice of the controls (asymptomatic in the past versus asymptomatic only in the present). The results of this meta‐analysis might suggest a pathogenic role of Mycoplasma in feline URT, with possible asymptomatic carriage after initial infection. Case and control populations should be carefully selected in future studies. More evidence is required to determine whether Mycoplasma is truly pathogenic in feline LRT. R04 SEVERE PRE‐CAPILLARY PULMONARY HYPERTENSION IN 11 SHIH TZU DOGS: (2007–2014) Amy Brida, Elizabeth Rozanski, Kristen Antoon, John Rush Tufts University, North Grafton, MA, USA Pulmonary hypertension (PHTN), defined as an increased in pulmonary arterial pressure >25 mmHg at rest may develop due to a variety of causes in dogs and is often classified into one of five mechanisms using the WHO classification scheme. When PHTN is due to left heart disease it is termed post‐capillary and is WHO classification II. Pre‐capillary causes of PHTN include WHO classification Types I, III, IV and V and may result from many causes including heartworm disease, COPD, and pulmonary vascular diseases. Pulmonary hypertension of all causes has been associated with a guarded prognosis in dogs. In our practice, pre‐capillary PHTN, typically without heartworm disease and without thrombotic disease, is a common cause of respiratory signs in Shih Tzu dogs. The purpose of this study is to describe the clinical attributes of a series of Shih Tzu dogs with severe pre‐capillary PHTN. The cardiology database was searched for Shih Tzu dogs evaluated by echocardiography and identified with PHTN. Pulmonary artery pressure was estimated by use of modified Bernoulli equation measuring the velocity of the regurgitant jet across the tricuspid value. PHTN was defined as severe if the estimated pulmonary artery pressure was >75 mm Hg. Dogs with evidence of left atrial enlargement, consistent with Type II pulmonary hypertension were excluded. The population characteristics and outcome were described. PHTN was documented in 70 Shih Tzu dogs. Of these, 50 were identified with pre‐capillary PHTN and 20 were classified with post‐capillary PHTN. Eleven of the 51 dogs with pre‐capillary PHTN had severe PHTN (median estimated PA pressure 97 mm Hg; range 80–140 mmHg). The median age of the affected dogs was 13 years (range 5–18), with six castrated males, 2 intact females, 2 spayed females and 1 intact male. The median survival time was 30 days (range 1–670 days). All dogs received supplemental oxygen during initial evaluation. Eight dogs were treated with sildenafil (median 1.5 mg/kg q 8–12 hours; range 1–3 mg/kg); four dogs also received pimobendan and clopidogrel. Four dogs survived more than 1 year. Three dogs died/were euthanized within 24 hours of diagnosis; none of these dogs were treated with sildenafil. Precapillary PHTN appears common in the Shih Tzu breed, is often severe, and is associated with a variable survival time. R05 THE EFFECT OF FUROSEMIDE ON TRANSVASCULAR FLUID FLUXES IN PULMONARY CIRCULATION IN DOGS UNDER GENERAL ANESTHESIA Olga Frlic1, Aleksandra Domanjko Petric1, Alenka Seliskar1, Rok Blagus2, George Heigenhauser3, Modest Vengust 1 1University in Ljubljana, Veterinary Faculty, Ljubljana, Slovenia, 2University of Ljubljana, Institute for Biostatistics and Medical Informatics, Ljubljana, Slovenia, 3McMaster University Medical Centre Hamilton, Hamilton, ON, Canada Pulmonary fluid dynamics and edema formation are governed by Starling's forces. Furosemide is often used for reduction of transmural pulmonary hydrostatic pressures to prevent or treat fluid accumulation in the pulmonary interstitium. Pulmonary transmural hydrostatic forces are reflected in pulmonary circulation transvascular fluid fluxes (J), which have never been defined in dogs. Hypotheses were tested that 1) the duration of anesthesia will influence J, and 2) that furosemide will reduce J. Twenty dogs (10 males, 10 females) that underwent castration/ovariectomy were randomly assigned to furosemide (FUR) (4 mg/kg) or placebo treated group (PLC). Dogs were premedicated with morphine (0.3 mg/kg sc), anesthesia was induced with midazolam (0.1 mg/kg iv) and propofol (3–4 mg/kg iv) and maintained with sevoflurane in oxygen. Thereafter dogs were instrumented with pulmonary artery and dorsal pedal artery catheters. Dogs were treated with Lactated Ringers solution during anesthesia (5 mL/kg/h iv). Baseline (BL) arterial and mixed venous blood were sampled just before treatment with FUR or PLC and then at 15, 30 and 45 minute post‐treatment. Cardiac output (Q), pulmonary artery (PAP) and mean arterial blood pressure (MAP) were measured. Transvascular fluid fluxes in pulmonary circulation were calculated from changes in plasma protein, hemoglobin, hematocrit and Q values. Variables were analyzed using random intercept mixed model (P < 0.05) Data are expressed as means with 95%CI. The duration of anesthesia had a significant effect on MAP (PLC: 70.7 mmHg (95%CI: 64.1–77.3) at BL to 84.6 mmHg (95%CI: 73.4–95.8) at 45 minute, P ≤ 0.0001), whereas Q and PAP were not affected by the duration of anesthesia. Treatment with furosemide had no effect on Q (P = 0.5), MAP (P = 0.2) and PAP (P = 0.9). At BL in PLC there was an efflux of fluid from the pulmonary circulation (3.5 mL/min, 95%CI: −0.4 to 10.9), which did not change at 15 minute (1.3 mL/min, 95%CI: −6.4 to 9.1), 30 minute (0.5 mL/min, 95%CI: −7.2 to 9.1) or 45 minute (2.9 mL/min, 95%CI: −3.9 to 8.7) of anesthesia (P > 0.05). At BL in FUR there was an influx of fluid into the pulmonary circulation (−6.9 mL/min, 95%CI: −13.9 to 0.2), which changed to an efflux at 15 minute (1.0 mL/min, 95%CI: −5.8 to 7.8), 30 minute (8.3 mL/min, 95%CI: 0.6–16.0) and 45 minute (1.0 mL/min, 95%CI: −2.1 to 2.3) of anesthesia. The duration of anesthesia did not affect J (P = 0.3). Treatment with furosemide also did not affect J (P = 0.3). Neither of our hypotheses were verified. Although there was a significant increase in MAP with time in PLC and FUR, the duration of anesthesia did not affect J. Similarly, FUR did not change transmural hydrostatic forces to an extent, which would change J in healthy dogs during general anesthesia. R06 THE PHARMACOKINETICS OF Β2‐ADRENOCEPTOR AGONIST IN TULOBUTEROL ON BEAGLE DOGS AFTER TRANSDERMAL AND INTRAVENOUS ADMINISTRATION Hyuck Joo Yang 1, Jae Hoon Kim1, Tae Heon Kim1, Hyung Jin Park2, Tae Sung Koo1, Kyoung Won Seo1, Kun Ho Song1 1Chungnam National University, Deajeon, Korea, 2Daegu Animal Medical Center, Deagu, Korea Tulobuterol, which contains a β2‐adrenergic agonist, is the first bronchodilator available as a transdermal patch. Pharmacokinetics of tulobuterol in veterinary medicine has not been well studied while studies on it with human are well investigated. In this study, the pharmacokinetics of tulobuterol on beagle dogs after transdermal and intravenous administration was compared to evaluate the utility of tulobuterol patch on dogs. Fifteen beagle dogs were used. For intravenous bolus injection, tulobuterol solution was prepared at the concentration of 1 mg/mL as intravenous dose in saline. Dose‐volume was 0.2 mL/kg for all dogs (Group A). For transdermal administration, 0.5 and 2 mg patches of tulobuterol were used. Doses were 0.2 mg/kg and 0.4 mg/kg (Group B and C). There was no systemic adverse effect on the dogs either during the treatment or during the following 7 days. Tulobuterol was rapidly absorbed after intravenous administration with a tmax of 0.017 hour. Tulobuterol was well absorbed after transdermal administration with an absorption lag‐time of about 4–8 hour. The Cmax of tulobuterol in dogs reached 2.09 ng/mL at 16.0 hours and reached 4.85 ng/mL at 13.6 hours after the transdermal administration of 0.2 and 0.4 mg/kg. Existing literature report that the effective blood level of tulobuterol in human medicine is 1–3 ng/mL. As evident from this study, the ideal blood drug concentration was reached by applying 0.2 mg/kg of the patch. It is believed that tulobuterol patch can be used in treating chronic respiratory diseases of dogs by sustained releasing. E01 ACCURACY OF NON‐INVASIVE OSCILLOMETRIC BLOOD PRESSURE IN NORMO‐HYPER AND HYPOTENSIVE ADULT STANDING HORSES Natalie Heliczer, Olivia Lorello, Daniela Casoni, Cristobal Navas de Solis University of Bern, Bern, Bern, Switzerland The study investigated the accuracy of an oscillometric blood pressure monitor (Cardell 9402), against invasive blood pressure (IBP), in standing horses before and during pharmacologically induced hyper‐ and hypotension. IBP was measured in eight mares using the facial or transverse facial artery. Non‐invasive blood pressure (NIBP), corrected to heart level, was measured simultaneously using the coccygeal artery. The heart rate and rhythm were recorded continuously using digital telemetry. Hyper and hypotension were induced by administration of phenylephrine (3 μg/kg/minIV for 15 minute) and acepromazine (0.05 mg/kg/IV) respectively. There was a significant correlation between NIBP and IBP (r = 0.84, 0.84, 0.88, P < 0.001 for systolic, diastolic and mean pressures respectively). The mean bias, [upper‐ lower limits of agreement] were 30.0[78.7–(−18.8)] mmHg for systolic, 8.9[43.3–(−25.5)] mmHg for diastolic and 16.4[48.9–(−16.1)] mmHg for mean blood pressure. Average NIBP was higher after phenylephrine [mean NIBP (mean±SD) 155.7 ± 30.0] and lower after acepromazine (78.6 ± 9.7) when compared to baseline (116.2 ± 19.4) (P < 0.05). When normo‐ hyper‐ and hypotensive phases were studied separately correlations were only detected for the normotensive phase (r = 0.85, 0.72 and 0.78, P = 0.01, 0.05, 0.02 for systolic, diastolic and mean pressure respectively). All horses showed frequent second degree atrioventricular and/or sinus block/arrest after phenylephrine administration. NIBP was strongly correlated with IBP and differentiated changes in pressure, but consistently underestimated blood pressure and had low accuracy. The device did not meet validation criteria proposed by ACVIM for small animals. Inaccuracies during hypertension and hypotension need further evaluation and phenylephrine induced arrhythmias could have affected NIBP. E02 ATTENUATION OF THE BLOOD PRESSURE RESPONSE TO EXOGENOUS ANGIOTENSIN I AFTER ORAL ADMINISTRATION OF BENAZEPRIL TO HEALTHY ADULT HORSES Tiago Afonso, Steeve Giguère, Gregg Rapoport, Michelle H. Barton, Scott A. Brown, Amanda E. Coleman University of Georgia, Athens, GA, USA Angiotensin‐converting enzyme inhibitors are used empirically in horses with cardiovascular disorders despite the paucity of data available. The objective of this study was to determine the lowest dose of benazepril that attenuates the rise in blood pressure after administration of exogenous angiotensin I (ANG‐I). Five healthy horses were instrumented for the direct measurement of arterial blood pressure. Each received 4 increasing oral doses of benazepril (0.5, 1.0, 2.0 and 4.0 mg/kg, intragastrically) with a washout period of 7 days between doses. Prior to and 2, 12 and 24 hour after benazepril administration, each horse received an IV bolus of ANG‐I at 20 ng/kg. Attenuation of the systolic arterial pressure (SAP) response (percentage of inhibition) was quantified. Data were analyzed using ANOVA for repeated measures. A mean (± SD) increase in SAP of 38.8 ± 3.7 mm Hg was obtained after administration of ANG‐I at baseline. There was a significant effect of benazepril dose (P = 0.004) and time (P = 0.004) on the percentage of inhibition of the systolic pressor response to ANG‐I. Regardless of dose, the percentage of inhibition was significantly greater 2 hour after administration of benazepril compared to 12 hour and 24 hour. The percentage of inhibition after administration of benazepril at 1.0 mg/kg (46.6 ± 18.9%) was significantly greater than that achieved after 0.5 mg/kg but not significantly different from that achieved after higher dosages. Benazepril at a dose of 1.0 mg/kg orally provides adequate attenuation of the rise in blood pressure after administration of ANG‐I. E03 EFFECT OF DIETARY INDUCED WEIGHT GAIN ON ORAL GLUCOSE TOLERANCE AND INSULIN RESPONSIVENESS IN SHETLAND PONIES Ellen de Graaf‐Roelfsema, Martha de Ruijter‐Villani, Simone de Vries, Tom Stout Department of Equine Sciences, Faculty of Veterinary Medicine, Utrecht University, Utrecht, The Netherlands Reasons for performing the study: obesity and hyperinsulinemia are both risk factors for the development of laminitis. It is not clear whether obesity induces hyperinsulinemia or hyperinsulinemia induces obesity. To examine the effects of inducing obesity by feeding a high energy, fat rich diet on insulin sensitivity and oral glucose tolerance in Shetland ponies. Seven Shetland mares (test group) were fed 200% of maintenance digestible energy (DE) requirements for 24 weeks to induce weight gain; six Shetland mares (control group) were fed 100% digestible energy requirements. Each pony underwent oral glucose tolerance testing at five time points (baseline = before starting dietary regime; period 1 = 5–8 weeks; period 2 = 10–13 weeks; period 3 = 17–18 weeks; period 4 = 22–23 weeks post onset of feeding regime). Ponies were weighed weekly. Statistical analyses were performed using a general linear mixed model with Bonferrroni as post hoc test. Differences were considered significant when P < 0.05. The test group showed a 26% increase in body weight during the course of the study. Area under the curve (AUC) for plasma glucose concentration was significantly reduced in the test group during periods 1, 2, 3 and 4 compared to the baseline, and was significantly lower than in the control group during periods 1, 2 and 4. Test group mares showed a greater (P < 0.05) OGTT insulin response during periods 3 and 4 than at baseline, and exhibited a significantly greater insulin response during periods 2, 3 and 4 than the control group. Inducing obesity by feeding a fat rich diet (200% of maintenance DE requirements) induced early stage hyperinsulinemia after oral glucose loading and resulted in an improvement in glucose metabolism, as evidenced by a significant decrease in the plasma AUC for glucose. E04 HORMONAL CHANGES: PREDICTIVE FACTORS OF CHRONIC FATIGUE IN SHOW HORSES Concetta Amato 1, Chantal Thorin2, Patrick Nguyen1, Brigitte Siliart1 1Nutrition and Endocrinology Unit, Oniris‐Nantes Atlantic College of Veterinary Medicine, Food Science and Engineering, NANTES, France, 2Animal Physiopathology and Functional Pharmacology Unit, Oniris‐Nantes Atlantic College of Veterinary Medicine, Food Science and Engineering, NANTES, France In horses the chronic fatigue is characterized by a reduction in the performance. The possibility that plasma hormone variations could be objective biomarkers of fatigue to exercise was tested by monitoring Iberian show horses (22 geldings and 15 stallions, 11 ± 4 years; bodyweight = 481 ± 34 kg, BCS = 2.8 ± 0.2) during a work season. Plasma adrenocorticotropic hormone (ACTH), cortisol, free thyroxine (FT4), testosterone, prolactin, insulin‐like factor‐I (IGF‐I), insulin and leptin concentrations were assessed monthly. The workload intensity score was determined according to exercise data and grouped into four categories: W1 to W4. Repeated‐measures ANOVA with post hoc Bonferroni's test were used to evaluate the effects of workload on plasma hormone concentrations and body condition parameters. Workload intensity significantly increased ACTH, testosterone, insulin and energy ration but significantly decreased cortisol, FT4, leptin and prolactin (P < 0.05). IGF‐1 increased significantly between W1 and W2, but decreased between W2 and W4 (P < 0.05). Testosterone/cortisol ratio and BCS increased significantly between W1 and W3, but decreased between W3 and W4 (P < 0.05). During heavy exercise, 49% of the horses showed a significant decrease in bodyweight despite a significant increase in energy intake (P < 0.05). ACTH, insulin, energy ration and BCS were positively correlated with workload. Cortisol, leptin, FT4, IGF‐1 and prolactin were negatively correlated with workload. Different hormonal responses are significantly linked to changes in physical performance. It might be possible to use these as markers to monitor exercise stress and indicate possible overtraining in the case of a negative energy balance in horses presenting signs of fatigue. E05 MULTIPLE ADRENOCORTICAL STEROID AND STEROID PRECURSORS RESPONSE TO ACTH STIMULATION TEST IN HOSPITALIZED FOALS Katarzyna Dembek 1, Kathryn Timko1, Lindsey Johnson1, Nathan Slovis2, Stephen Reed3, Kelsey Hart4, Ramiro Toribio1 1Department of Veterinary Clinical Sciences, The Ohio State University, Columbus, OH, USA, 2Hagyard Equine Medical Institute, Lexington, KT, USA, 3Rood and Riddle Equine Hospital, Lexington, KT, USA, 4College of Veterinary Medicine, University of Georgia, Athens, GA, USA Sepsis remains the main cause of death in neonatal foals. The hypothalamic‐pituitary‐adrenal axis regulates the response to sepsis‐associated stress. We and others have shown that relative adrenal insufficiency (RAI), characterized by an impaired cortisol response to ACTH, is common and associated with mortality in foals. Most studies evaluating adrenocortical function in foals have been focused on cortisol, while other adrenocortical steroids, steroid precursors, and neurosteroids have not been extensively investigated. We hypothesized that RAI in critically ill foals will involve multiple adrenocortical layers, resulting in decreased glucocorticoid and mineralocorticoid, and increased steroid precursor and neurosteroid secretion, which will be associated with severity of disease and mortality. We also proposed that the response of these steroids to exogenous ACTH will be altered. Foals <1 day of age were categorized into 3 groups: 11 septic, 9 sick non‐septic and 11 healthy. After baseline blood sample collection on admission (Time 0), each foal received 10 μg of ACTH (cosyntropin). Blood samples were also collected at 30 and 90 minute post ACTH (Time 30 and 90 minute). Steroid and steroid precursor concentrations were determined by immunoassays. The delta steroid was defined as the percent change in steroid concentration between Time 0 and 30 minute (Delta0–30) and Time 30 minute and 90 minute (Delta30–90). Septic foals had higher cortisol, pregnenolone, progesterone, 17α‐OH‐progesterone and androstenedione at all three time points and higher aldosterone at Time 0 and 30 minute compared to healthy foals (P < 0.05). Non‐surviving septic foals had lower cortisol concentrations at Time 0 and 30 minute compared to surviving septic foals (P < 0.05). The Delta0–30 was lower in septic compared to healthy foals for cortisol, progesterone, aldosterone, and 17α‐OH‐progesterone (P < 0.05), suggesting RAI. Progesterone and 17α‐OH‐progesterone decreased, however, DHEA increased in response to ACTH at Time 30 minute in septic foals. ACTH stimulation did not influence pregnenolone and androstenedione concentrations at Time 30 minute. The Delta30–90 was not different between groups of foals. In our study, RAI was indicated by altered steroidogenesis in response to ACTH in critically ill foals. Hypocortisolemia was associated with poor prognosis for survival in septic foals. E06 THE EFFECT OF PH ON LARGE INTESTINAL ION TRANSPORT AND TRANSEPITHELIAL RESISTANCE IN SUSPECTED INSULIN RESISTANCE HORSES Nicole Arana‐Valencia 1, Frank Andrews2, Marion Thibault3, Catherine Koch2, Rita Perry2, Adeline Thischmacher3, Nicholas Frank4 1School of Animal Sciences, Louisiana Agricultural Experiment Station, Louisiana State University, Baton Rouge, LA, USA, 2Equine Health Studies Program, Department of Veterinary Clinical Sciences, Louisiana State University School of Veterinary Medicine, Baton Rouge, LA, USA, 3Ecole Nationale Veterinaire De Toulouse, Toulouse, France, 4Department of Veterinary Clinical Sciences, Tufts University School of Veterinary Medicine, North Grafton, MA, USA Endocrinopathies, such as equine metabolic syndrome (EMS) and pituitary pars intermedia dysfunction (PPID) can lead to abnormal distribution of adipose tissue, hyperinsulinemia, glucose intolerance, and insulin resistance. Recent studies have shown that intravenous administered insulin and glucose cause laminitis. In addition, during insulin infusion, some of these horses develop mild colic signs and diarrhea. It might be that hyperinsulinemia causes alterations in large intestinal ion transport and transepithelial resistance. These factors might be critical when colon pH is decreased because of intake of feeds with high soluble carbohydrate content. The purpose of this study was to determine the effect of the pH on colonic mucosa in insulin resistant horses using an in‐vitro Ussing chamber system. Tissue samples of the right dorsal colon were collected in suspected insulin resistant (n = 4) and normal horses (n = 5) immediately after euthanasia. The colonic mucosa was dissected from the muscular layers and mounted in Ussing chambers. Ussing chambers are used to measure ion transport and transepithelial resistance in gut mucosa, as an indicator of viability and permeability. Colonic tissue was exposed to pH 4, 6, or 7 (control) on the mucosal side and pH 7.0 on the submucosal side. Potential difference (PD) and short circuit current (Isc) were measured every 15 minutes. Tissue transepithelial resistance (TER) was calculated from the open‐circuit PD and Isc, using Ohm's Law (R = PD/Isc). The mean Isc and TER did not change in the colonic tissues from control horses when exposed to different pH. However, in insulin resistant horses, mean Isc decreased rapidly in right dorsal colon tissues exposed to pH 4.0, but this was not significant. No change was seen at pH 6 or 7. Also, mean TER decreased significantly (P < 0.015) in colon tissues exposed to pH 4.0 from insulin resistant horses, when compared to tissues exposed to pH 6 or 7. At a low pH, colonic mucosal ion transport and transepithelial resistance were decreased in the insulin resistant horses, compared to control horses. Insulin resistant horses might be predisposed to alterations in mucosal ion flux and transepithelial resistance, especially when there is a drop in pH after feeding a diet high in soluble carbohydrates. Decreased ion fluxes and TER might result in the colic and diarrhea seen in horses afflicted with EMS and PPID. These data are preliminary and further research is indicated to confirm these findings. E07 EPIDEMIOLOGICAL CHARACTERISTICS OF HORSES WITH PPID AT INITIAL DIAGNOSIS Steven Grubbs 1, Dwana Neal1, Thomas Keefe2 1Boehringer ingelheim Vetmedica, Inc, St. Joseph, MO, USA, 2Colorado State University, Fort Collins, CO, USA Pituitary pars intermedia dysfunction (PPID) has been described as the most common endocrinologic disorder of aged horses. Few studies exist that describe the epidemiological characteristics of horses with PPID. Additional epidemiologic studies are needed to determine the prevalence of PPID in a larger population, not just aged horses. The purpose of this study was to obtain epidemiological information that included age, breed, sex, clinical signs, and insulin/glucose status at initial PPID diagnosis (new cases) from a large population of horses. Horses of any age, breed, and sex from the continental US were eligible for study enrollment as long as they were documented to be exhibiting one or more of the following clinical signs: generalized or regional hypertrichosis, muscle wasting, abnormal fat distribution, lethargy, laminitis (unknown etiology), polyuria, polydipsia, susceptibility to infections, abnormal sweating, and/or inappropriate lactation. Normal horses were excluded from the study. At initial visit, demographic data, signalment, a physical examination was conducted. Clinical signs were documented and blood was drawn for basal ACTH, fasting insulin, and glucose. Blood samples were processed and shipped overnight to the Animal Health Diagnostic Center, Cornell University, Ithaca, NY for analysis. The association between PPID status, based on ACTH and insulin results, and each of the demographic variables (age, sex, breed, and geographic region), the two test result variables insulin and glucose were statistically evaluated individually using the Pearson chi‐square test. Odds ratios for significant predictors of PPID status were computed using corresponding 95% confidence intervals when applying multiple logistic regression analysis. Five‐hundred fifteen of nine‐hundred eighty two horses with complete epidemiological information were included in the final data analysis. Of the 515 horses, 222 (43%) were PPID+, 203 (39.4%) were IR+. Further, horses were stratified into 1 of 4 groups based on ACTH and insulin laboratory results (PPID+/IR‐), (PPID+/IR+), (PPID‐/IR+), and (PPID‐/IR‐). Of the 515 horses, 115 (22.3%) were PPID+/IR‐, 97 (18.8%) were PPID+/IR+, 106 (20.6%) were PPID‐/IR+, and 197 (38.3%) were (PPID‐/IR‐). Of the 515 horses evaluated, 18.5% of horses <15 years of age were PPID+, 33% of horses 15 to 19.9 years of age were PPID+, 56.3% of horses 20 to 24.9 years of age were PPID+, and 74% of horses >25 years of age were PPID+. In particular, the odds of PPID for horses between 15 and 19.9 years of age was approximately twice (2.2) that for horse <15 years old. Compared to horses in the youngest age group, the odds ratio (OR) for PPID increased to 4.6 for horses between 20 and 24.9 years of age and to 14.0 for horses ≥25 years of age. Even after adjusting for the age of the horses, insulin‐resistant horses were seen to have approximately twice the odds (1.9) of PPID compared to horses with normal insulin levels. Of the 515 enrolled horses, 61.7% were diagnosed with an endocrine disorder (PPID and/or hyperinsulinemia). As expected, PPID prevalence increased significantly (P < 0.001) by age and was significantly higher in horses with above‐normal levels of both insulin and glucose (P = 0.014 and P = 0.016, respectively). Further, forty‐three percent of PPID horses were also hyperinsulinemic. Therefore, when evaluating horses with suspected endocrine disease, at a minimum, ACTH, insulin and glucose should be evaluated. Long term studies need to be conducted in large populations of horses to further evaluate endocrinopathies in horses. E08 EFFECTS OF FASTING AND THE ORAL SUGAR TEST ON THYROTROPIN‐RELEASING HORMONE STIMULATION TEST RESULTS IN HORSES Melissa Restifo, Pilar Hermida, Alfredo Sanchez, Nicholas Frank Cummings School of Veterinary Medicine at Tufts University, North Grafton, MA, USA It is often necessary to test for pituitary pars intermedia dysfunction (PPID) and insulin dysregulation in the same horse. We hypothesized that results of the thyrotropin‐releasing hormone (TRH) stimulation test would not be affected by feeding conditions or the oral sugar test (OST). Thirty adult horses (aged 6–21 years of age; 8 mares and 22 geldings) from the same facility were tested on 4 occasions across a 6‐week period. All horses underwent a TRH stimulation test under fed (day 1) and then fasted (day 14) conditions, an OST alone (day 28), and then a combined OST/TRH stimulation test (day 42). Test results were compared using the Wilcoxon rank‐sum test or paired student's t‐test, and agreement was assessed by Bland‐Altman analysis. Fasting or feeding did not affect plasma adrenocorticotropin hormone (ACTH) concentrations at 0 (P = 0.968) or 10 minute (P = 0.598) in the TRH stimulation test. Baseline (time 0) ACTH concentrations did not differ (P = 0.167) when TRH stimulation tests were performed alone (fasted) or in combination with the OST. Plasma ACTH concentrations increased after TRH administration in all tests, but were lower at 10 minute (P = 0.004) in the combined test, compared to the TRH stimulation test (fasted); with a mean bias of 26 pg/mL (95% confidence interval; −60 to 111 pg/mL) detected. Feeding conditions did not impact TRH stimulation test results, but ACTH concentrations increased by a lower magnitude when testing was combined with the OST. Combination testing is not recommended at this time. E09 EVALUATION OF AGREEMENT OF FOUR COMMON DIAGNOSTIC TESTS FOR INSULIN RESISTANCE IN ADULT LIGHT‐BREED HORSES Laura Dunbar, Kaitlin Mielnicki, Katarzyna Dembek, Ramiro Toribio, Teresa Burns The Ohio State University, Columbus, OH, USA Several tests have been evaluated in horses and humans for quantifying systemic insulin resistance, but many of these tests have not been evaluated for agreement. This study evaluated the agreement between four methods of insulin sensitivity testing in horses, including basal serum insulin and glucose concentration, oral sugar test (OST), combined glucose‐insulin test (CGIT), and the insulin‐modified frequently sampled intravenous glucose tolerance test (FSIGTT). Each of the above tests was performed once in random order on 12 healthy light‐breed horses. Minimal Model analysis of the FSIGTT classified seven horses as insulin resistant (IR) and 5 as insulin sensitive (IS) (IR = SI < 1.0 × 10−4 L/mU/min). In contrast, basal insulin and OST classified all horses as IS (Insulin60 min <60 μIU/mL). Calculated Kappa statistics for the CGIT and FSIGTT were 0.274 (IR = PP‐Dglu >45 minutes) and 0.25 (IR = Insulin45 min > 100 μIU/mL). Sensitivity of the CGITPP‐Dglu>45 min was 85.7% and specificity was 40%. Sensitivity of the CGITIns>100μIU/mL–45 min was 28.5% and specificity was 100%. Area under the glucose curve (AUCg0–120) of the OST, CGIT, and FSIGTT showed significant linear relationships, but Bland‐Altman comparison showed lack of agreement between tests. Current criteria for diagnosis of IR using basal insulin and the OST are highly specific but lack sensitivity. While these tests may be useful in diagnosing severely IR individuals, they may not be adequate in detecting mildly affected horses. The CGIT displayed better sensitivity and specificity; however, more liberal criteria may be necessary to improve agreement with Minimal Model analysis. E10 VITAMIN D METABOLITES AND THEIR ASSOCIATION WITH SEVERITY OF ILLNESS AND MORTALITY IN HOSPITALIZED FOALS Ahmed Kamr 1, Katarzyna Dembek1, Bonnie Barr2, Michele Frazer3, Ahmed Zaghawa4, Thomas Rosol1, Ramiro Toribio1 1College of Veterinary Medicine, The Ohio State University, Columbus, OH, USA, 2Rood and Riddle Equine Hospital, Lexington, KY, USA, 3Hagyard Equine Medical Institute, Lexington, KY, USA, 4College of Veterinary Medicine, Sadat City University, Sadat City, Egypt Hypocalcemia is a frequent finding in septic foals that is associated with mortality. Hypovitaminosis D in critically ill people has been linked to hypocalcemia and mortality; however, information on vitamin D metabolites and their association with clinical and laboratory information, as well as mortality in critically ill foals is lacking. We hypothesized that hypovitaminosis D, hypocalcemia, hyperphosphatemia, and increased parathyroid hormone (PTH) concentrations will be frequent in critically ill hospitalized foals. We also proposed that vitamins D metabolites will be associated with severity of disease and likelihood of mortality in septic foals. A total of 100 neonatal foals ≤72 hours old divided into hospitalized (n = 83; 59 septic, 24 sick non‐septic [SNS]) and healthy (n = 17) groups were included in this study. Blood samples were collected on admission. Serum 25‐hydroxyvitamin D3 [25(OH)D3], 1,25‐dihydroxyvitamin D3 [1,25(OH)2D3], and PTH concentrations were measured by immunoassays. Calcium and phosphorus concentrations were determined by colorimetric methods. Data were analyzed by nonparametric methods and univariate logistic regression. The prevalence of hypovitaminosis D [25(OH)D3 < 9.51 ng/mL] was 63% (53/83) for hospitalized, 64% (38/59) for septic and 63% (15/24) for SNS foals. Serum 25(OH)D3 and 1,25(OH)2D3 concentrations were significantly lower in septic and SNS compared to healthy foals (P = 0.001; P = 0.037). Septic foals had significantly lower calcium and higher phosphorus and PTH concentrations than healthy and SNS foals (P < 0.05). Low 1,25(OH)2D3 concentrations <7.09 pmol/L were associated with increased PTH concentrations in hospitalized foals. Septic foals with 25(OH)D3 <9.51 ng/mL and 1,25(OH)2D3 <7.09 pmol/L were more likely to die [OR = 3.62; OR = 5.41] compared to septic foals with normal 25(OH)D3 and 1,25(OH)2D3 concentrations [9.51–16.89 ng/mL; 7.09–16.48 pmol/L], respectively. However, vitamin D metabolites were not associated with mortality in SNS foals (P = 0.34). Hypovitaminosis D, hypocalcemia, hyperphosphatemia and elevated PTH concentrations were frequent in critically ill foals. The association between vitamin D metabolites with severity of disease and mortality in septic foals supports a role for vitamin D in the pathogenesis of equine neonatal disorders. E11 USE OF METABOLOMIC APPROACHES FOR INSIGHT INTO METABOLIC PERTURBATIONS IN WELSH PONIES WITH INSULIN DYSREGULATION Sarah Jacob 1, Raymond Geor1, Jill McCutcheon1, Nichol Schultz2, Molly McCue2 1Michigan State University College of Veterinary Medicine, East Lansing, MI, USA, 2University of Minnesota College of Veterinary Medicine, St. Paul, MN, USA In human medicine, the rapidly developing area of metabolomics has emerged as a promising approach to identify biomarkers of diabetes and insulin resistance before the development of overt pathology. To date, however, there has been minimal application of metabolomics in equine medicine. The objective of the study reported here was to examine the serum metabolome of Welsh ponies with and without evidence of insulin dysregulation before and during an oral sugar test (OST). Twenty Welsh ponies (12.6 ± 8 years; mean ± SD) were selected from a cohort of 300 ponies in a study examining genetic and metabolic risk factors for metabolic syndrome and laminitis. Ponies were classified as healthy [CON] (n = 10, insulin <30 mU/L) or having insulin dysregulation [ID] (n = 10, insulin >60 mU/L) at 75 minute post administration of Karo syrup (0.15 mL/kg per os). Metabolomic analysis, completed by a commercial laboratory (Metabolon®) using a combination of gas chromatography mass spectrometry (GC‐MS) and liquid chromatography mass spectrometry (LC‐MS), was performed on serum samples obtained before (baseline) and at 75 minute during the OST. Plasma glucose (glucose oxidase) and serum insulin (radioimmunoassay) concentrations also were measured at baseline and 75 minute. Data were analyzed by two‐way repeated measures ANOVA following log transformation, with significance set at P ≤ 0.05 to identify differences between groups and time points. Mean insulin concentrations during the OST were significantly higher in ID when compared to CON ponies at baseline (mean ± SD; ID 15.4 ± 5.6 mU/L, CON 4.0 ± 2.6 mU/L) and at 75 minute (ID 112.7 ± 29.1 mU/L, CON 16.3 ± 8.3 mU/L) whereas mean glucose concentrations did not differ between groups at baseline (ID 73.0 ± 8.3 mg/dL, CON 73.4 ± 6.7 mg/dL) or at 75 minute (ID 108.3 ± 12.4 mg/dL, CON 106.4 ± 23.7 mg/dL). The serum metabolomic profile comprised 646 biochemical compounds, of which 506 were of known identity. Statistically significant differences between ID and CON were identified for >130 biochemicals, particularly metabolites involved in the tricarboxylic acid (TCA) cycle, fatty acid metabolism, and branched‐chain amino acid (BCAA) oxidation. Mannose, produced by enzymatic isomerization of glucose, was higher (P < 0.05) in ID than in CON at baseline. In both groups fructose, pyruvate and mannose were significantly higher at 75 minute when compared to baseline. Citrate, fumarate and malate, metabolites associated with the mitochondrial use of pyruvate, were lower in ID than in CON at 75 minute, suggesting lower TCA cycle activity in ID ponies. Several long‐chain fatty acids (e.g. palmitoleate, eicosapentanoate) and acylcarnitines (e.g. palmitolycarnitine, laurylcarnitine) were 2‐to‐3‐fold higher in ID vs. CON at baseline and at 75 minute. In addition, intermediates in BCAA metabolism including 4‐methyl‐2‐oxopentanoate, 3‐methyl‐2‐oxoalerate and isovalerylglycine were significantly higher in ID when compared to CON ponies, suggesting reduced BCAA oxidation in ID ponies. In this cohort of Welsh ponies, examination of the serum metabolome revealed differences in the TCA cycle, fatty acid metabolism and BCAA oxidation when comparing animals with and without evidence of insulin dysregulation. Collectively, these differences may represent a state of mitochondrial dysfunction in the ID when compared to CON ponies. E12 CAN MOTION CAPTURE DETECT ATAXIA? Emil Olsen 1, Nathalie Fouche2, Hannah Jordan3, Richard J. Piercy4, Thilo Pfau3 1Department of Clinical Sciences, Cornell University Hospital for Animals, Cornell University, Ithaca, NY, USA, 2Institut suisse de médecine équine ISME, Vetsuisse Fakultät, Bern University, Bern, Switzerland, 3Structure & Motion Laboratory, The Royal Veterinary College, North Mymms, UK, 4Comparative Neuromuscular Diseases Laboratory, Department of Clinical Sciences and Services, The Royal Veterinary College, London, UK A thorough neurological examination of questionably ataxic horses is crucial to differentiate normal from abnormal and to help with neuroanatomical localization. Unfortunately agreement amongst experienced clinicians can be poor when assessing presence and severity of ataxia, especially when signs are mild; consequently, objective gait characterization might be beneficial. We hypothesize that assessment of variability of spatial characteristics and swing duration derived from motion capture (kinematics) have good sensitivity and specificity and can detect ataxia in horses. Twenty‐one horses underwent a gait laboratory measurement and live neurological assessment by 4 to 6 raters. In the gait laboratory the horses were walked across a runway surrounded by a 12‐camera motion capture systemi with a sample rate of 240 Hz. They were walked normally and with a blindfold in at least three trials each, with 3–4 strides subsequently processed using semi‐automated custom‐written Matlabii scripts. Displacement of reflective markers on head, fetlock, hoof, L4, tuber coxae and sacrum were processed. The horses were considered ataxic if they had a median rater ataxia grade ≥ 2/4. The statistical software R was used to generate receiver operator characteristics to determine the diagnostic sensitivity (Se), specificity (Sp) and area under the curve (AUC). Objective measurement of pelvic limb coefficient of variation for the time point where the fetlock joint reached maximum vertical displacement had greater diagnostic value than that in thoracic limbs. Blindfolding exacerbated objective signs of ataxia when compared to normal horses (AUC of 0.76, Se of 82% and Sp of 80%). Coefficient of variation for duration of swing during normal walk had an AUC of 0.70 and a Se of 64% with a Sp of 90%. We conclude that motion capture can be useful when assessing and quantifying ataxia objectively. Furthermore there is greater variation of the pelvic limbs gait parameters for ataxic horses when walking blindfolded compared to non‐ataxic horses when walking blindfolded. Objective assessment during the neurological examination might in future improve diagnostic accuracy. i:Qualisys AB, Gothenburg, Sweden, ii:The Mathworks inc., Natick, MA, USA. E13 INTRAVENOUS PLASMA ADMINISTRATION EVOKES A BALANCED SYSTEMIC INFLAMMATORY RESPONSE IN HEALTHY NEWBORN FOALS Eleonora Po, Brian M. Aldridge, Pamela A. Wilkins, Elena Bichi Ruspoli Forteguerri, Breanna Sheahan, Kara M. Lascola, Scott M. Austin, Kevin H. Kline, Dennis D. French University of Illinois, Urbana, IL, USA Fresh frozen plasma (FFP) transfusions are common in critically ill and healthy neonatal foals. While anaphylactic reactions are reported, the extent and nature of FFP transfusion‐associated immune activation in foals is unknown. We hypothesized that FFP transfusion evokes a detectable systemic inflammatory response in healthy foals. Thirty‐six healthy, Standardbred foals (24–36 hours old) were allocated to either group 1 (n = 18) receiving a 1L commercial FFP transfusion, or group 2 (n = 18) receiving 1L of isotonic crystalloids, by intravenous infusion. Jugular blood was collected before transfusion and at 2, 6 and 24 hours post‐administration, and evaluated for serum amyloid A ([SAA], turbidometric assay), fibrinogen ([FIB], Modified Clauss method), and TNF‐α, IL‐1β, IL‐10 (commercial ELISA kit). Statistical comparisons between groups and across time, was performed using Chi squared and repeated measures ANOVA. [FIB] was significantly increased (P < 0.05) at 24 hours compared to baseline in foals receiving FFP. [SAA] was increased at 24 hours in foals receiving FFP compared to crystalloids (P < 0.05). Serum IL‐1β was undetectable at all times. Compared to baseline, many foals showed increased TNF‐α (6 fold, P < 0.001) and IL‐10 (1.3‐fold, P < 0.001) concentrations at 2, 6, and 24 hours. At 24 hours, the number of foals with these increases was significantly greater in group 1 (15/18 TNF‐α; 10/18 IL‐10) than in group 2 (6/18 TNF‐α; 1/18 IL‐10) (P < 0.001). In combination, these findings indicate that FFP transfusion evokes a balanced (pro/anti‐inflammatory), systemic immune response in healthy foals. The presence of inflammation in some control foals implies a dualistic immunogenic mechanism. E14 A COMPREHENSIVE EVALUATION OF TRADITIONAL AND NOVEL IGG CUT‐OFF VALUES AND THEIR ASSOCIATION WITH MORTALITY IN NEONATAL FOALS Rachel Liepman, Katarzyna Dembek, Nathan Slovis, Stephen Reed, Ramiro Toribio 1The Ohio State University, Columbus, OH, USA 2Hagyard Equine Medicine Institute, Lexington, KY, USA 3Rood and Riddle Equine Hospital, Lexington, KY, USA Serum immunoglobulin G (IgG) assessment in neonatal foals is considered standard of care in equine hospitals to determine immunity and overall health. However, cut‐off values of IgG to predict complete or partial failure of transfer of passive immunity (FTPI) were developed 30 years ago, are largely empiric, with little prospective or statistical data to support their use or association with outcome. Our objectives were to comprehensively evaluate traditional cut‐off values of IgG in the assessment of FTPI (IgG <800 mg/dL), to generate novel IgG cut‐off values, and determine their association with likelihood of non‐survival. This was a multicenter, prospective, cross‐sectional study. We evaluated clinicopathologic variables in 597 foals ≤7 days old including serum IgG, hematology and serum biochemistry. Foals were divided into three groups by diagnosis: healthy (n = 45), sick non‐septic (n = 341) and septic (n = 211). Five IgG cut‐off values were generated. Univariate and multivariate logistic regression were used to compute odds ratios for non‐survival in these foals. Our findings support the use of the traditional cut‐off value of >800 mg/dL as adequate transfer of passive immunity (ATPI). Stepwise decreases in IgG values <800 mg/dL were proportionally associated with non‐survival. Hospitalized foals with any degree of FTPI were 3.5–10 times more likely to die. There were no statistical differences in non‐survival if foals had IgG concentrations between 801–1200 mg/dL compared to those with >1200 mg/dL. Foals with hyperproteinemia also had higher mortality rates. We recommend immediate assessment of IgG concentrations in hospitalized foals and those with FTPI should receive prompt immunotherapy. E15 COMPARISON OF IMMUNOLOGIC RESPONSES FOLLOWING INTRANASAL AND ORAL VACCINATION WITH A USDA APPROVED, LIVE‐ATTENUATED STREPTOCOCCUS EQUI VACCINE Katherine Delph 1, Elizabeth Davis1, Nora Bello1, Kevin Hankins2, Melinda Wilkerson1, Catherine Ewen1 1Kansas State University, Manhattan, KS, USA, 2Zoetis Inc, Florham Park NJ, USA The purpose of this pilot study was to compare both systemic and local immune responses following intranasal or oral vaccination for Streptococcus equi var equi. Eight healthy horses with Streptococcus equi M protein titers <1:1600 were vaccinated with a live‐attenuated vaccine intranasally per label instructions or transmucosally (orally) in a two‐vaccine series. Both serum and nasal secretions were collected for baseline and 2‐ and 4‐week post booster vaccinations for detection of IgG and IgA. Antibody assessment was performed using a commercially available ELISA for M protein titers and using a novel microbead assay analyzed by the Luminex™ Magpix instrument which utilizes flow cytometric principles. Horses vaccinated by the intranasal route demonstrated a substantial increase in both serum IgG and serum IgA M protein antibody levels post vaccination (P = 0.0006 and P = 0.007, respectively). Horses that were vaccinated by the oral route also demonstrated an increase in the serum IgG M protein antibody levels post vaccination (P = 0.0150), though the increase was of approximately one third the magnitude of that observed in intranasal vaccinates. Orally‐vaccinated horses did not show any evidence of response in serum IgA post vaccination (P = 0.15). Because of the small scale of the study, study design limitations, and the unknown protective level of IgG against S. equi, the findings of this study should be interpreted carefully. Results from this study indicate that horses that received the live‐attenuated Pinnacle® Streptococcus equi var equi vaccine either through intranasal or oral routes showed an increase in serum IgG M protein titer, though the magnitude of the response differed between administration routes. E16 MAJOR HISTOCOMPATIBILITY COMPLEX EXPRESSION IN MUSCLE OF HORSES WITH IMMUNE MEDIATED MYOSITIS Sian Durward‐Akhurst 1, Carrie Finno2, Jan Shivers1, Stephanie Valberg1 1University of Minnesota, St Paul, MN, USA, 2University of California‐Davis, Davis, CA, USA Major histocompatibility complex (MHC) I and II expression is not detected in normal muscle and appears to be a prerequisite for the autoimmune cytotoxicity that occurs in muscle from humans and dogs with immune‐mediated myositis (IMM). The purpose of this study was to characterize lymphocyte subtypes in muscle biopsies of suspected IMM horses and determine if MHC I and II expression occurs in myocytes. Immunohistochemical staining for MHC class I, II, CD3+, CD4+, CD8+, and CD20+ lymphocytes was performed on muscle samples from 18 Quarter Horses with a history of recent muscle atrophy. The degree of lymphocyte infiltration in muscle or around vessels was scored for each lymphocyte subtype. Sarcolemmal MHC expression was detected in all muscle biopsies with a majority (n = 14) of horses having more MHC class I than II expression. A strong positive correlation occurred between MHC class I and II expression (r 2 = 0.62). Scores for all lymphocyte subtypes infiltrating myocytes significantly correlated with MHC I and II expression (range r 2 0.24–0.51: P < 0.05). In 11/18 horses, CD4+ rather than CD8+ lymphocytes predominated. In conclusion, an immune‐mediated basis for equine IMM is supported by MHC I and II expression on myocytes. Myocytes expressing MHC class I and II will present processed self‐ and non‐self antigenic peptides to CD8+ and CD4+ T‐lymphocytes, respectively, activating an immune response within skeletal muscle. E17 CHARACTERIZATION OF THE RESPIRATORY TRACT MICROBIOTA OF HEALTHY HORSES Gaelle Hirsch 1, Henry R. Staempfli1, Marcio C. Costa2, Luis G. Arroyo1, J. Scott Weese1 1University of Guelph, Guelph, ON, Canada, 2Universidade Estadual de Londrina, Londrina, PR, Brazil This study describes and compares for the first time the bacterial compositions of different locations in the equine respiratory tract using high throughput sequencing. Samples were collected from 8 healthy adult horses immediately after euthanasia. The upper respiratory tract was sampled by nasal swab, nasopharyngeal swab and guttural pouch lavage. Bronchoalveolar lavage (BAL) and bronchial protected brush sampling was then performed through surgically prepared tracheostomy. High throughput sequencing of 16S rRNA gene PCR products (V4 region) was performed. Sequences were classified into operational taxonomic units (OTUs). A total of 3,530,730 sequences passed all quality control tests. Firmicutes and Proteobacteria were the main phyla found at all sites (ranges 46.8–60.2% and 33.6–46.6%, respectively). Relative abundances between sites were overall similar. Members of the Gammaproteobacteria (42.5–50.1%) and Bacilli (32.4–48.2%) classes and members of the Moraxellaceae (26.2–38.1%) and Bacillaceae (26.5–37.5%) families were most commonly found at all sites. Acinetobacter was the main genus present at all sites (range 24.9–38.0%). Staphylococcus (P = 0.026) and Anoxybacillus (P = 0.037) were over represented in the nasal swabs compared to bronchial brushes. As expected there were significantly greater observed and estimated (chao1) OTUs in the nasal cavities compared to other sites (all P < 0.05). After sub‐sampling of 5440 reads per site, 169, 99, 66, 76 and 70 OTUs were detected from nasal swabs, nasopharyngeal swabs, guttural pouch lavages, bronchial brushes and BAL, respectively. There were no differences in evenness (Shannon evenness index, P = 0.509) and diversity (inverse Simpson's index, P = 0.812). There was a significant difference in community membership (Jaccard index) between nasal swabs compared to other sampling sites (Parsimony test P < 0.001), but no difference in community structure (Yue & Clayton index of dissimilarity, P = 0.555). When samples were categorized as upper or lower respiratory tract, similar results were obtained, with a difference in membership (P = 0.030) but not structure (P = 0.910). The difference in membership but not in structure, along with the greater richness in the nasal cavities compared to other sites may be the result of addition of a large number of members that are of low abundance, something that could reflect transient colonization or contamination from the environment. Compared to other organ systems, the respiratory tract harbors a more homogenous and less rich microbiota that decreases in richness but varies little in community structure from upper to lower airways. This study establishes baseline data for the healthy equine airway microbiota. E18 THE RELATIONSHIP BETWEEN SERUM 25‐HYDROXY VITAMIN D LEVELS AND INFLAMMATORY AIRWAY DISEASE (IAD) IN HORSES Daniela Bedenice, Ashley Houtsma, Elizabeth Rozanski, Melissa Mazan Cummings School of Veterinary Medicine, Grafton, MA, USA Human studies have previously documented a strong association between vitamin D deficiency and bronchial asthma, a condition similar to equine inflammatory airway disease (IAD). The goal of this study was, therefore, to evaluate the relationship between serum 25‐Hydroxyvitamin‐D levels and IAD in horses. Twenty‐three horses fulfilling the diagnostic criteria of IAD, as previously defined (2007 ACVIM consensus), were recruited for this study. Additionally, 13 asymptomatic control animals were sampled from comparable stabling environments. Serum 25‐Hydroxyvitamin‐D analysis (nmol/L), broncho‐alveolar‐lavage fluid (BALF) cytology and non‐invasive pulmonary function testing (forced oscillatory mechanics – FOM, or respiratory inductive plethysmography – RIP) for the assessment of airway hyperreactivity (AHR) were performed in all horses. BALF slides were stained with modified Wright stain and Toluidine Blue. All data were presented descriptively as mean (± standard deviation) based on normality of data distribution. Categorical data were compared between groups using Chi‐square analysis, and continuous variables using independent samples t‐test, with P < 0.05 considered significant. A diagnosis of IAD was either associated with AHR (18/23; 78%) or abnormal BALF analysis (21/23, 91%). Sixteen IAD horses showed airway neutrophillia (mean BALF neutrophils: 17.6 ± 16.6%), 2 demonstrated airway eosinophilia (mean BALF eosinophils: 0.2 ± 0.5%), and 15 horses had increased BALF mast cells (mean: 3.5 ± 3.4%). Serum 25‐Hydroxyvitamin‐D levels were significantly lower in IAD (12.1 ± 4.3 nmol/L) versus control horses (17.1 ± 5.9 nmol/L, P = 0.016), and were statistically associated with airway neutrophillia (neutrophils ≥5%; P = 0.035). In conclusion, serum 25‐Hydroxyvitamin‐D levels were inversely associated with a diagnosis of IAD in horses, and were related to BALF neutrophillia. E19 SEPTIC PLEUROPNEUMONIA IN 97 HORSES Maria Arroyo1, Nathan Slovis2, George Moore1, Sandra Taylor 1 1Purdue University, West Lafayette, IN, USA, 2Hagyard Equine Medical Institute, Lexington, KY, USA To investigate signalment, history, physical examination findings, clinicopathological and bacteriological data, treatments and outcome in horses diagnosed with septic pleuropneumonia. A retrospective study was performed. Horses were included in the study if they were diagnosed with sepsis (based on criteria for SIRS and confirmed infection) at admission, and demonstrated pleural effusion via transthoracic ultrasonography. The study period was from 2001 to 2014. Ninety‐seven horses met the inclusion criteria. The median age was 2.5 years, and travel within 6 weeks of admission was the most common risk factor. Common historical complaints included fever (84% of cases), lethargy (78%) and inappetance (77%), while common physical examination findings at admission included tachycardia (75%), tachypnea (60%) and fever (43%). Hyperfibrinogenemia and hypoalbuminemia were present in 79% and 69% of cases, respectively. The most common bacteria isolated from tracheal aspirates and pleural fluid was Streptococcus equi subsp. zooepidemicus. Within individual animals, only 20% of tracheal aspirates and pleural fluid samples yielded the same results, with tracheal aspirate culture more sensitive than pleural fluid culture. Laminitis was diagnosed in 8% of horses. The survival rate to discharge from the hospital was 67%. A thoracotomy was performed in 11 horses (11%), and was positively associated with survival (P = 0.030). Azotemia was negatively associated with survival (P < 0.001) and return to work (P = 0.034). Horses with azotemia were 11 times more likely to die than horses without azotemia, with every 1 unit increase in creatinine increasing the risk of death 5‐fold. Racehorses with a recent history of traveling are at risk of developing septic pleuropneumonia. Upper respiratory tract flora continues to represent the most common bacteria isolated from the lower airway of horses with pleuropneumonia. Tracheal aspirate culture is superior to pleural fluid culture, and laminitis is uncommon in horses with pleuropneumonia. Azotemia is a negative prognostic indicator and likely reflects severity of illness upon admission. Finally, thoracotomy should be considered in horses with pleuropneumonia. E20 FACTORS AFFECTING COMPLICATION RATES WITH SUBPALPEBRAL LAVAGE CATHETER USE IN HORSES Sophie Cornelissen1, Elizabeth Finding2, Mark Bowen1, Gayle Hallowell 1 1School of Veterinary Medicine and Science, University of Nottingham, Sutton Bonington, Leicestershire, UK, 2Royal Veterinary College, North Mymms, UK Subpalpebral lavage (SPL) systems are commonly used to facilitate treatment of a wide variety of ophthalmic conditions in the horse. The objectives of this retrospective study were to evaluate whether there were different complication rates between placement in the upper versus the lower eyelid and what other factors influenced the development of complications leading to catheter replacement. Data was extracted from clinical records from the Royal Veterinary College between 2000–2013. These cases were identified from invoices where SPL catheters were billed. During this time period only one brand of SPL catheter was used (Mila, Kentucky, USA). Data collated included age, breed, gender, clinician, stage and type of training, ophthalmic condition, pharmacological agents administered, catheter position, duration of placement, whether the catheter was managed in the hospital or at home and any complications reported and action(s) taken. This data was then entered and coded into a spreadsheet. Data was initially interrogated using Chi‐squared tests. Two binary logistic regression models were built in a step‐wise fashion to evaluate the factors associated with position of catheter insertion and complication rates. Data was extracted from 135 horses (138 SPL catheters) aged 6 days to 30 years with gender reflective of hospital admission (54 females and 81 males). A range of ophthalmic conditions were treated with SPL's, but the most common were corneal ulceration (n = 51;37%), squamous cell carcinoma (n = 25;18%) and uveitis (n = 22;16%). Duration of SPL was 3–60 days (median 8; IQR 7–15). Complication rates were lowest for medicine residents (10%), when compared with interns (42%), medicine clinicians (27%) and surgeons (16%). There was no difference in complication rate between SPL management at home (15.4%) compared to the hospital (16.1%; P = 0.94). Position of catheter placement were influenced by stage of training (P = 0.04; boarded clinicians were more likely to place catheters in the lower eyelid) and associated with duration of treatment (P = 0.03; longer treatment times associated with SPL's in the upper eyelid). Complications were associated with catheter position (P = 0.03; upper eyelid 12.8% and lower eyelid 22.2%), antimicrobial administration (P = 0.008; 13.2% complication rate when used and 25.0% when not) and plasma administration (P = 0.004; 30.4% complication rate when used and 13.0% when not). In conclusion, SPL placement in the upper eyelid was associated with fewer complications, use of antimicrobials appeared protective against complications and plasma should only be administered when indicated as it increased the likelihood if complications. E21 ESTRADIOL 17β AND ESTRONE SULFATE IN MARES WITH EXPERIMENTALLY INDUCED ASCENDING PLACENTITIS Igor Canisso 1, Barry Ball2, Alejandro Esteller‐Vico2, Edward Squires2, Mats Troedsson2 1University of Illinois Urbana Champaign College of Veterinary Medicine, Urbana, IL, USA, 2Maxwell Gluck Equine Research Center, University of Kentucky, Lexington, KT, USA Placental steroids have been suggested as useful diagnostic markers for pregnancy loss in mares; however, there are few controlled studies that have examined changes in steroids in mares with placentitis. The objectives were to compare plasma concentrations of estradiol 17β and estrone sulfate in mares with placentitis with gestationally age‐matched control mares. Pregnant mares (260–280d of gestation) were allocated to a control group (n = 17) and to a group with experimentally induced placentitis (n = 17). Induction of placentitis was achieved via intracervical inoculation of Streptococcus equi ssp zooepidemicus. Plasma was collected at inoculation for mares in the placentitis group or commence for control mares (day = 0), and then daily until abortion or for 12 days post inoculation (DPI). Estrone sulfate concentration was determined by ELISA, whereas estradiol 17β was determined by a chemiluminescence immunoassay. Data were analyzed by mixed models, for DPI and for days from abortion (DFA) for the different steroids. Mares with experimentally induced placentitis aborted within 5–25 DPI. There were significant effects of time (DPI, P < 0.0001; DFA P = 0.0008) and time by group interactions (DPI or DFA P < 0.001) for the concentrations of estradiol 17‐β. Concentrations of estradiol 17‐β were significantly reduced at 4 DPI and at – 2 DFA (P < 0.05). Analyses of estrone sulfate revealed effects of time (DPI P < 0.003; DFA P = 0.01) and time by group interactions (DPI P = 0.002; DFA P < 0.0001). Estrone sulfate concentrations were significantly reduced 6 and 7 DPI and on the day of abortion (0 DFA) for mares with experimentally induced placentitis. Concentrations of estrogens changed significantly in mares with experimentally induced ascending placentitis. E22 IGM AND IGG RESPONSES IN HORSES AND PONY FOALS AFTER VACCINATION FOR WEST NILE VIRUS AND EASTERN EQUINE ENCEPHALITIS Frank Andrews 1, Dylan Shannon1, Sara Lyle1, Michael Keowen1, Robert Keene2, Steve Grubbs2 1Louisiana State University, Baton Rouge, LA, USA, 2Boehringer Ingelheim Vetmedica, St. Joseph, MO, USA West Nile (WN) and Eastern Equine Encephalitis (EEE) Viruses are endemic in the US and cause neurologic disease in horses. Diagnosis of WNV and EEE encephalitis is made based on clinical signs and a serum capture ELISA IgM titer of >1:400. Vaccination of horses and foals with WNV and EEE vaccines might illicit an IgM antibody response that could interfere with the capture ELISA diagnostic test, especially if horses and foals are tested after vaccination. The hypothesis of this study is that horses and foals vaccinated with a multivalent vaccine containing WN and EEE antigens will not elicit an IgM antibody response of >1:400. Six adult horses (8–25 years) and six 4 month old pony foals and their dams were used in the study. Both horses and foals were vaccinated with a commercially available combination vaccine (Vetera® Gold, Boehringer Ingelheim Vetmedica, St. Joseph, MO, USA) containing killed WN and EEE antigens. Adult horses and dams of the ponies had a previous history of yearly vaccination, whereas the pony foals had not been previously vaccinated. Blood samples were collected from the jugular vein of the horses, dams and foals on Days 0 (prior to vaccination), 7, 14, and 21 days after the 1st vaccination. On day 21, foals were given a booster vaccination and additional blood samples were obtained days 7, 14 and 21 post booster vaccination. In addition, blood samples were obtained from the dams of the foals at the same times and were used as sentinels for virus exposure, as the foals and dams were housed together in a pasture at the Louisiana State University Veterinary Sciences Farm. Blood samples were centrifuged and serum collected and frozen at −80°C until shipment. Serologic assays for WN and EEE, including IgM antibody capture ELISA (MAC‐ELISA) and Plaque Reduction Neutralization Assay (PRNT), were performed at the National Veterinary Services Laboratory (NVSL) in Ames, IA, USA. All MAC‐ELISA titers for WN and EEE were negative at 1:400 for the horses, foals and dams of the foals during the study period. In the adult horses, PRNT titers ranged from 1:10 to >1:100 for EEE and >1:100 for WN at all times. In the pony foals, PRNT titers to EEE and WN ranged from negative to 1:10 prior to vaccine administration to >1:100 after the second vaccination. One foal had negative PRNT titers to EEE throughout the experiment. Vaccination with a multivalent vaccine containing WN and EEE elicited an antibody response in horses and pony foals. However, vaccination for WN and EEE did not increase IgM titers >1:400, 7 days after vaccination in adult horses and ponies and after initial and 2nd vaccination in naïve pony foals. Therefore, vaccination does not interfere with serologic testing for naturally occurring WN and EEE infections. E23 FACTORS ASSOCIATED WITH EQUINE SHEDDING OF MULTI‐DRUG RESISTANT SALMONELLA ENTERICA AND ITS IMPACT ON HEALTH OUTCOMES Brandy A. Burgess 1, Kristin Bauknecht2, Nathan M. Slovis2, Paul S. Morley3 1Virginia Tech, Blacksburg, VA, USA, 2Hagyard Equine Medical Institute, Lexington, KY, USA, 3Colorado State University, Fort Collins, CO, USA Salmonella enterica is an important factor in healthcare‐associated epidemics and zoonotic disease in veterinary hospitals – with outbreaks of multi‐drug resistant (MDR) Salmonella among equine patients resulting in high case fatality rates and substantial financial cost. The objectives of this study were, 1) to determine factors associated with fecal shedding of MDR‐Salmonella; and 2) to determine what effect Salmonella shedding may have on health outcomes of previously hospitalized horses and their stablemates. Patients eligible for this case‐control study included those having fecal cultures for S. enterica as part of a surveillance program from January 2011 through December 2012. Data regarding exposures of interest were collected retrospectively from medical records. Information on long‐term outcomes was obtained by an owner phone survey. Multivariable regression techniques were used to determine factors associated with shedding MDR‐Salmonella and subsequent health outcomes. Equine patients enrolled in this study included 94 culture‐positive and 279 culture‐negative (on at least 3 fecal samples) horses from 199 farms. Horses experiencing diarrhea during hospitalization were more likely to shed Salmonella compared to horses without diarrhea; and horses having decreased feed intake during hospitalization were more likely to shed MDR‐strains compared to horses with normal feed intake. In general, shedding did not increase long‐term risk for non‐survival, colic or abnormal feces after discharge of hospitalized horses nor did it increase risk for hospitalization or abnormal feces in stablemates. Despite these findings, it is still recommended to manage horses shedding Salmonella separately from other resident horses and to employ rigorous personal and environmental hygiene. E24 ESTIMATING THE SENSITIVITY AND SPECIFICITY OF REAL‐TIME QUANTITATIVE PCR OF FECAL SAMPLES FOR DIAGNOSIS OF RHODOCOCCUS EQUI PNEUMONIA IN FOALS Sarah Shaw 1, M. Keith Chaffin1, Glenn Blodgett2, Melissa Syndergaard2, Danny Hurych1, Noah D. Cohen1 1Equine Infectious Disease Laboratory, Department of Large Animal Clinical Sciences, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University, College Station, TX, USA, 26666 Ranch, Guthrie, TX, USA Many field veterinarians eschew diagnostic testing for Rhodococcus equi pneumonia because of the time, labor, and risk to foals associated with obtaining a tracheobronchial aspirate. A real‐time, quantitative PCR (qPCR) method for detection of R. equi in feces has been developed as a non‐invasive, rapid diagnostic test for R. equi pneumonia, but has not been evaluated in a large population of foals. The objective of this study was to evaluate the clinical utility of fecal PCR as a diagnostic test for detecting foals with R. equi pneumonia using receiver operating characteristic (ROC) methods. The study included 186 foals born in 2011 on an R. equi‐endemic ranch in Texas. Fecal samples were collected every 2 weeks for all foals and at the time of onset of clinical signs in affected foals (n = 31). Foals with pneumonia were matched by age and date‐of‐birth to healthy (n = 31) and subclinical (n = 124) control foals. The concentration of R. equi in feces, determined by qPCR, differed significantly (P < 0.05) among groups. The area under the ROC curve for using fecal qPCR for diagnosis of R. equi pneumonia was 89% (95% CI, 83% to 99%) with a sensitivity of 94% and a specificity of 72%. These data indicate that real‐time, quantitative PCR on feces may be useful for the diagnosis of R. equi in foals with clinical signs of pneumonia. E25 BAYESIAN GEOSTATISTICAL AND ENVIRONMENTAL RISK FACTORS OF PIGEON FEVER Courtney Boysen, Elizabeth Davis, Ram Raghavan, Laurie Beard, Brian Lubbers, Monica Farfan Kansas State University, Manhattan, KS, USA During the summer and fall months of 2012 Kansas and its neighboring states in the Central Midwestern United States witnessed an unprecedented number of pigeon fever cases, which is typically an arid region disease in horses. This study presents a retrospective Bayesian geostatistical analysis and environmental risk factors associated with the pigeon fever cases received at the Kansas State University Veterinary Health Center and the Kansas State Veterinary Diagnostic Lab. Disease caused by Corynebacterium pseudotuberculosis in horses were determined by positive test results for characteristic abscess formation, positive bacterial culture on purulent material obtained from a lanced abscess (n = 82), or positive serologic evidence of exposure to organism (>1:512) (n = 11). Horses negative for these tests (n = 172) were considered absence of disease. Horse demographics, stabled location as well as stabling history 3 months prior to development of clinical signs were obtained through review of medical records or contact with client via telephone. Environmental covariate information was collected from publicly available sources using Geographic Information Systems (GIS) and remote‐sensing methods; soil attributes from USDA State Soil Geographic Survey (SSURGO), land use/land cover attributes from the USGS National Land Cover Dataset (NLCD), and climate data from NASA Moderate Resolution Imaging Spectroradiometer (MODIS) and Prediction of Worldwide Renewable Resources (POWER). The candidate covariates were screened using a linear univariate regression model with individual testing of each covariate and those with significance (P < 0.2) were kept. A Bayesian geostatistical model was performed both with and without covariates. The DIC was lower when covariates were added, giving them some spatial autocorrelation in pigeon fever status among horses. The covariates that remained statically significant after the addition of the covariates into the Bayesian model included: soil moisture, total edge contrast index, and land surface temperature (>33°C). Increased soil moisture with an OR of 0.53 appeared protective while both total edge contrast (OR 1.27) and land surface temperature >33°C (OR 1.31) were risk factors for development of disease. This data may lead to future investigations regarding prevention or better management of disease outbreaks. E26 RISK FACTORS FOR EQUINE INFECTIOUS ANEMIA IN CANADA Sara Higgins 1, Katharina Lohmann1, Krista Howden2, Carolyn James3, Tasha Epp1 1Western College of Veterinary Medicine, University of Saskatchewan, Saskatoon, SK, Canada, 2Canadian Food Inspection Agency, Edmonton, AB, Canada, 3Canadian Food Inspection Agency, Ottawa, ON, Canada This study represents a small component of a larger study looking to identify potential risk factors for Equine Infectious Anemia (EIA) in Canada. This smaller study used information that accompanied serum sample submissions to the Canadian Food Inspection Agency (CFIA) National Reference Laboratory between 2009 and 2012 for the purposes of disease investigation, export testing to countries other than the US and Mexico (excluding testing of feedlot horses), and confirmatory testing of laboratory referrals. All disease investigation and export samples were taken by CFIA veterinarians or veterinary inspectors and all laboratory referral samples were taken by CFIA‐accredited private veterinarians. An animal was considered positive for EIA when it tested positive by agar gel immunodiffusion (AGID), and only one positive test per animal was retained for the analysis. The association between potential risk factors (equid species, breed, sex, age, coat color, month and year of submission, province and region (Eastern vs. Western Canada)) and a positive test result was examined using binary logistic regression with binomial distribution to calculate odds ratios (OR) with 95% confidence interval (CI). A final multivariable model based on non‐modifiable horse characteristics was created from the dataset and P < 0.05 was considered significant. Samples with missing data were excluded from the analysis. Information was available for 5981 submissions, which included 420 animals that tested positive. Based on univariate analysis, breed, province, region, year and month of submission were significant risk factors for a positive EIA result. Age and breed were retained in the final multivariable model and showed significant interaction. Conclusion: Breed and province/region may be a risk factor for EIA in Canada and may indicate effects of use‐specific management factors. E27 DEPLETION OF PHENYLBUTAZONE RESIDUES FROM EQUINE URINE AND TISSUES Patricia Dowling 1, Ron Johnson2, Joe Boison3 1Western College of Veterinary Medicine, Saskatoon, SK, Canada, 2Ontario Veterinary College, Guelph, ON, Canada, 3Canadian Food Inspection Agency, Saskatoon, SK, Canada Phenylbutazone (PBZ) is a nonsteroidal anti‐inflammatory drug commonly used to treat musculoskeletal pain and inflammation in horses. It is approved for use in horses in Canada and the United States, but product labelling carries the warning “not for use in horses intended for human food”. This is not a legal prohibition against the use of PBZ in horses intended for slaughter in Canada, but rather a disclaimer that there is no legal Maximum Residue Limit (MRL) established for PBZ in horse tissues, so the detection of any amount of PBZ constitutes a residue violation. Globally, the use of PBZ in horses that may eventually be used for food has been very controversial and the detection of undeclared horse meat in food products in the European Union (EU) in 2013 caused much concern that people were exposed to unsafe residues of PBZ. Because it is very effective, inexpensive and generally safe in horses, PBZ continues to be extensively used in many countries. Because of the human safety concerns and lack of residue depletion data and MRLs for PBZ, the European Union adopted the “precautionary principle” and banned PBZ for use in all food animals. Because of the valid and wide spread use of PBZ in North American horses, a lifetime ban is impractical and detrimental to equine welfare. This project was undertaken to provide information on the depletion of PBZ from urine and tissues of horses and to address concerns regarding the risk of PBZ residues in horse meat. Validated liquid chromatography‐mass spectrophotometry (LC‐MS/MS) assays were developed for the detection of PBZ and its metabolite oxyphenbutazone (OXPBZ) in equine serum and urine (Maxxam Analytics) and tissues (Canadian Food Inspection Agency). The limits of quantification (LOQ) for the assays were 1 ng/mL for PBZ and 2 ng/mL for OXPBZ residues in serum and urine and 0.5 ng/g for PBZ and OXPBZ in all equine tissues. These LOQs are well below the EU's Minimum Required Performance Limit (MRPL) for analytical methods for phenylbutazone of 5 ng/g. Utilizing horses destined for slaughter and a high dose PBZ treatment regimen (8.8 mg/kg/day for 4 days), we conducted a pilot study to determine the sampling time points for a full drug depletion study. As we predicted, pilot study results indicated that blood concentrations of PBZ and OXPBZ depleted rapidly and did not reflect tissue concentrations. Both PBZ and OXPBZ depleted rapidly from equine muscle, being readily detectable only at the first slaughter time point of 7 days post‐PBZ administration. In the full depletion study, using 20 horses slaughtered in groups of 5, we confirmed that even with a very high dose treatment regimen, PBZ depleted very rapidly from equine muscle and that liver is the tissue from which PBZ residues deplete at the slowest rate. While OXPBZ concentrations exceeded PBZ concentrations in urine and were still detected 42 days post‐PBZ administration, no residues of OXPBZ were detectable in muscle, kidney, or liver from 21 days post‐PBZ administration. Our results indicate PBZ should be the regulatory marker compound and liver should be the target tissue for regulatory enforcement. In conclusion, even after a high dose PBZ treatment regimen, PBZ and OXPBZ deplete rapidly from equine muscle and a lifetime ban on its use in horses intended for human food is not justified. If the medication history of a horse is questionable, pre‐slaughter testing of urine or post‐slaughter testing of liver will prevent PBZ contaminated horse meat from entering the human food chain. E28 EFFECT OF RIFAMPIN ON ERYTHROMYCIN‐INDUCED ANHIDROSIS IN FOALS Amy Stieler, L. Chris Sanchez, Martha Mallicote, Alexis Muniz, Jim Burrow, Robert J. MacKay University of Florida, College of Veterinary Medicine, Gainesville, FL, USA Previous work by the authors showed that erythromycin‐associated hyperthermia is caused by impaired sweating. The effect of rifampin on erythromycin‐induced anhidrosis is unknown. This study compared terbutaline‐induced sweating in erythromycin‐treated foals with sweating responses in foals given rifampin alone or in combination with erythromycin. In separate experimental sessions, 12 pony‐cross foals (nine female, three male), 1 month old at enrollment, were each treated for 5 days with erythromycin base (25 mg/kg orally, three times daily), rifampin (10 mg/kg orally, twice daily), or a combination of the 2 medications according to a duplicated 6 × 3 counterbalanced measures design. Quantitative intradermal terbutaline sweat tests were performed on each of three successive days before treatment (baseline) and on days 1, 2, 5, 9, 24, and 39 after treatment began. There was significant (P < 0.0005) suppression of sweating in foals given erythromycin with or without rifampin but no effect (P = 0.800) of rifampin alone. Compared with terbutaline‐induced sweating at baseline, values were significantly reduced (P < 0.0005) in foals given erythromycin with or without rifampin at all evaluation days except d 39. The addition of rifampin treatment to the standard erythromycin regimen did not change overall sweating responses to intradermal terbutaline (P = 0.071). Results show that, despite evidence that rifampin reduces bioavailability of a related macrolide (clarithromycin) by more than 90%, the potent anti‐sudorific effect of erythromycin is undiminished by concurrent rifampin administration. E29 EX VIVO EQUINE SUBCONJUCNTIVAL INJECTION CHARACTERISTICS AND IN VITRO RELEASE PROFILES OF VORICONAZOLE FROM A PLGA‐PEG‐PLGA THERMOGEL Rosemary Cuming, Sue Duran, Allison Stewart, Anne Wooldridge, William Ravis, Yuh‐Jing Lin, Eva Abarca Auburn University, Auburn, AL, USA Development of a voriconazole‐containing thermogel for subconjunctival space (SCS) injection could allow for sustained drug delivery to target tissues and improved success in treating keratomycosis. The purpose of this study was to determine if sustained‐release of voriconazole occurs when combined with a thermosensitive Poly(lactic‐co‐glycolic acid)‐b‐Poly(ethylene glycol)‐b‐Poly(lactic‐co‐glycolic acid) (PLGA‐PEG‐PLGA) hydrogel in vitro and to characterize SCS thermogel injection in ex vivo equine eyes. Six groups were defined: 1.) thermogel, 2.) 1 mg voriconazole + thermogel, 3.) 1 mg voriconazole in ethanol + thermogel, 4.) 5 mg voriconazole + thermogel, 5.) 5 mg voriconazole in ethanol + thermogel and 6.) 5 mg voriconazole. Thermogel samples were maintained at 34.5°C in phosphate buffered saline (PBS). PBS was collected daily for 28 days and voriconazole concentrations determined via high‐performance liquid chromatography. Ten normothermic ex vivo horse eyes were injected with 300 μL thermogel and evaluated via high‐resolution (50 MHz) ultrasound. Voriconazole was successfully combined with the thermogel in crystalline form and ethanol solution. Voriconazole release from the thermogel followed first order kinetics with t1/2 ranging from 1.82 (Group 3) to 4.22 days (Group 2). Voriconazole concentrations released exceeded the target MIC (0.5 μg/mL) for 28 days in Groups 2 and 4 and 16 days in Groups 3 and 5. Addition of ethanol inconsistently altered voriconazole release. A well‐demarcated, hypoechoic structure formed in the SCS following thermogel injection. Voriconazole is released in a sustained manner from PLGA‐PEG‐PLGA thermogel in vitro. SCS injections of liquid thermogel are feasible in equine eyes and result in formation of a well‐defined gel deposit at body temperature. E30 POST‐TRANSFUSION SURVIVAL TIME OF CROSS‐MATCH INCOMPATIBLE RED BLOOD CELLS IN ADULT HORSES Joy Tomlinson 1, Ester Taberner Brugue2, Raymond Boston2, Sean Owens3, Rose Nolen‐Walston2 1Cornell University, Ithaca, NY, USA, 2University of Pennsylvania, Kennett Square, PA, USA, 3University of California Davis, Davis, CA, USA In species such as humans and cats, there is a markedly reduced half‐life of transfused red blood cells (RBC) when the donor and recipient blood cross‐reacts. Only 10–20% of horses have naturally occurring alloantibodies and cross‐match is not always performed prior to whole blood transfusion. We hypothesized that cross‐match incompatibility predicts shortened RBC survival time after autologous transfusion in horses. Twenty healthy adult horses were used. Blood type, anti‐RBC antibody screen and major and minor cross‐match determined 10 pairs. Two pairs were cross‐match compatible, and the remainder were incompatible. Donor blood (4L) was collected into citrate phosphate dextrose adenine‐1, labelled with NHS‐biotin, and transfused into recipients. Post‐transfusion samples were collected at 1 hour and 1, 2, 3, 5, 7, 14, 21, 28, and 35 days. Proportions of biotinylated RBC were detected by flow cytometry using streptavidin‐phycoerythrin. Cross‐match incompatibility was significantly associated with decreased survival time (P < 0.001). The half life of transfused incompatible (cross‐match >1+) allogeneic equine RBC was 4.7 days versus 33.5 days for compatible RBC. Cross‐match >1+ was significantly associated with febrile transfusion reaction (P = 0.0083). Cross‐match testing prior to transfusion is recommended whenever feasible despite the low incidence of naturally occurring alloantibodies in horses. E31 TEMPORAL EFFECTS OF INTRAVENOUS ANTIBIOTICS ON THE EQUINE FECAL MICROBIOME Rachel Liepman, Jacob Swink, Greg Habing, Ramiro Toribio The Ohio State University, Columbus, OH, USA Acute colitis is the most common and devastating complication of antibiotic therapy in horses. Fecal culture often fails to identify etiologic agents and has low sensitivity in detecting organisms that are difficult to cultivate. Metagenomic studies utilize the 16S ribosomal RNA (rRNA) sequence for bacterial identification and classification and can produce profiles of genetic diversity from microbial communities. There are limited data to explain how antibiotics affect the fecal microbiome of horses over time. Our objectives were to provide a comprehensive description of bacterial phylum structures within feces of normal horses, to determine how commonly used antibiotics affect these communities over time and if 5 days of treatment is long enough to detect changes. We also investigated if these changes persisted after a 30‐day period. Sixteen healthy horses with no recent history of underlying disease were used for this study. They were randomly assigned to one of four treatment groups (n = 4 horses per group). Each group was treated intravenously with antibiotics (ceftiofur, enrofloxacin, oxytetracycline) or saline (control) at standard doses for five consecutive days. Fecal samples were collected once daily during treatment. Bacterial DNA was extracted, PCR amplified and analyzed by 454 pyrosequencing of the 16S rRNA gene from baseline (before treatment) three, five and 30 days post‐treatment. The composition and diversity of the microbiota were determined and compared using MOTHUR. Linear mixed models were used to analyze differences between treatment groups and time points using subject as a random effect. Overall, 18 phyla were identified in all horses and an average of 19.9% 16S sequences remained unclassified. Diversity within subjects (alpha diversity) was not significantly altered by treatment or over time. However, community structure between samples changed with most differences observed in Proteobacteria, Firmicutes, Bacteroidetes and Verrucomicrobia (P < 0.05). The treatment duration (5 days) was sufficient to identify changes in fecal microbiota, and changes in community structure were frequently specific to the antibiotic. Ceftiofur and enrofloxacin treatment resulted in a >90% reduction in Proteobacteria compared to other treatments. Ceftiofur and oxytetracycline treatment led to the most significant changes in major phyla over time. Within treatment groups, several changes in phylum structure persisted at 30 days, however some returned to near baseline. E32 PROSPECTIVE EVALUATION OF THE ACCURACY OF THE MFLASH ULTRASOUND PROTOCOL AND TRANSRECTAL ABDOMINAL PALPATION FOR PREDICTING SPECIFIC SURGICAL DIAGNOSES IN CASES OF EQUINE ACUTE COLIC Emilie Setlakwe, Barbara Dallap Schaer, Helen Aceto, JoAnn Slack University of Pennsylvania‐New Bolton Center, Kennett Square, PA, USA There is scarce literature regarding the accuracy of either transabdominal ultrasound or transrectal abdominal palpation compared to surgical diagnosis in cases of equine acute abdominal pain. The aims of this study were to evaluate the level of agreement between 1) a modified FLASH protocol (mFLASH, thoracic and left middle third windows excluded) and the surgical diagnosis and 2) transrectal abdominal palpation and the surgical diagnosis. Specific diagnoses of interest included nephrosplenic entrapment, right dorsal displacement, large colon volvulus, and surgical small intestinal lesion. Sequential cases of client‐owned horses presenting for evaluation of acute abdominal pain were enrolled in the study. Only horses that were ≥1 year of age and large enough for transrectal abdominal palpation were included. Transrectal abdominal palpation was performed first by the attending clinician and the descriptive findings and tentative diagnosis were recorded in a standardized recording sheet. An mFLASH was then completed and descriptive findings and ultrasound diagnosis recorded. Horses that underwent exploratory celiotomy within 6 hours of evaluation were included in the statistical analysis. Kappa two‐way agreement comparisons were performed to assess agreement. Sixty‐one cases fit the inclusion criteria. Results showed that mFLASH was in moderate agreement with the surgical diagnosis (Kappa = 0.42, P < 0.001), while transrectal abdominal palpation was in slight agreement with the surgical diagnosis (Kappa = 0.18, P = 0.005). In conclusion, mFLASH is a more accurate tool than transrectal abdominal palpation in predicting specific surgical diagnoses in cases of equine acute abdomen. E33 PREVALENCE OF AND RISK FACTORS FOR GASTRIC ULCERATION IN SHOWJUMPING WARMBLOODS Sarah Pedersen 1, Claire Windeyer1, Emma Read1, Dan French2, Alastair Cribb1, Heath MacLeod1, Heidi Banse1 1University of Calgary, Calgary, AB, Canada, 2TD Equine Veterinary Group, Calgary, AB, Canada Gastric ulcers are a common finding in performance horses. Recognition that ulcers may impair performance has led to investigations of prevalence in multiple performance disciplines. Previous research has focused primarily on risk factors for squamous ulcers, while the role of specific risk factors in the development of glandular ulcers remains unclear. The purpose of this study was to determine prevalence of and identify risk factors for the development of squamous and/or glandular gastric ulcers in a population of actively competing Warmblood showjumping horses. Eighty‐nine Warmbloods from 14 barns were included in the study. Gastroscopies were performed between April and July 2014 after a 12–16 hour fast. Any glandular or squamous ulceration was recorded and scored on a scale of 0–4 (EGUS Council). Trainers completed management surveys that included questions regarding feeding, housing, competition level, training practices, and previous medical history, including medications administered. Continuous variables were examined using Student's t‐test and categorical data examined using Chi squared test or Fisher's Exact test (SPSS). The prevalence of glandular ulceration was 72%, with 49% of horses having grade 2 or greater glandular ulceration. The prevalence of squamous ulceration was 40%, with 28% of horses having grade 2 or greater ulceration. Concurrent glandular and squamous ulceration was identified in 34% of horses. Preliminary risk factors identified for grade 2 or greater ulceration of any kind were exercising ≥6 days per week (P < 0.05) and feeding hay once daily (P = 0.04). For glandular ulceration of grade 2 or greater, horses competing at national‐level events showed an increased risk compared to those competing at FEI level (P = 0.03). This study suggests that once daily hay feeding and increased exercise frequency are associated with an increased risk of gastric ulceration. The decreased risk of glandular ulceration observed in FEI‐level horses is unexpected, but may suggest that glandular ulcers are associated with reduced performance. E34 THE EFFECT OF TRANSPORT, FASTING AND ANESTHESIA ON THE COMPOSITION OF THE EQUINE MICROBIOTA Angelika Schoster 1, Martina Mosing1, Henry Staempfli2, Mohammad Jalali3, Scott Weese3 1Vetsuisse Faculty, Equine Departement, University of Zurich, Zurich, Switzerland, 2Departement of Clinical Studies, Ontario Veterinary College, University of Guelph, Guelph, ON, Canada, 3Ontario Veterinary College, Departement of Pathobiology, University of Guelph, Guelph, ON, Canada The intestinal microbiota is important for health and disease. Factors that disturb the equine intestinal microbiota are incompletely studied. The objective of this study was to determine effects of transportation, fasting and anesthesia on the fecal microbiota of healthy adult horses using next generation sequencing Fecal samples from eight horses at baseline, after transportation, after 12 hour of fasting and 24, 48 and 72 after 6 hours of anesthesia were studied. Next generation sequencing of the V4 region of the 16S rRNA gene followed and was used to assess the microbial composition. Alpha diversity, phylogenetic structures and beta diversity were assessed using MOTHUR. There were few changes in the relative abundance of different taxonomic groups at higher taxonomic levels. At the phylum level horses had significantly less Cyanobacteria after fasting (P = 0.012) and anesthesia (P = 0.012) than baseline. There were more changes at lower taxonomic levels. Most notably the order Clostridiales was decreased after anesthesia (P = 0.003). Alpha diversity was not significantly different between time points (all P > 0.21). When parsimony analysis was applied, anesthesia had a significant effect on both community membership (Jacquard index P = 0.02) and community structure (Yue & Clayton index P = 0.02). There was an effect of fasting and anesthesia on the structure of the microbiota of healthy horses. A reduction of Clostridiales has also been seen in horses affected with colitis. This indicates that these are stress factors for the equine intestinal microbiota and could be related to development of disease in the post‐operative period. E35 WHITE COAT HYPERTENSION, INTERDAY VARIATION OF BLOOD PRESSURE AND INTERDAY AND INTEROBSERVER VARIATION OF SPLENIC VOLUME Olivia Lorello, Natalie Heliczer, Daniela Casoni, Cristobal Navas de Solis University of Bern, Bern, Bern, Switzerland This study investigated changes in non‐invasive oscillometric blood pressure (NIBP) between a clinic setting and home environment (white coat hypertension), normal daily variation of NIBP and splenic volume and interobserver variation of splenic volume. NIBP, measured in 38 horses in clinical and home environment on consecutive days, was compared. NIBP and spleen volume were measured three consecutive days and splenic volume repeated by two ultrasonographers each time in 8 mares. NIBP, corrected to heart level, was measured using the coccygeal artery. Systolic and mean NIBP were significantly (P = 0.01 and P < 0.001 respectively) lower in home (mean ± sd 129 ± 14 and 103 ± 13 systolic and mean NIBP respectively), than hospital environment (138 ± 10 and 109 ± 11 systolic and mean NIBP respectively). Repeated measures ANOVA (P > 0.05) showed no difference in splenic volume and NIBP between days. There was no correlation in the changes of splenic volume and NIBP in different days. There was no difference in splenic volume between observers. The bias [upper‐ lower limits of agreement] for splenic volume between observers was 1.8 [5.9‐(−2.3) liter]. The power of the ANOVA calculations were lower than 0.8. Horses can have white coat hypertension similarly to other species. A marker of sympathetic tone (change in splenic volume) was not correlated with the physiologic day to day changes in blood pressure. The lack of interday and interobserver differences should be interpreted with caution due to the risk of type II error. E36 CORRELATIONS OF BLOOD PRESSURE, SPLENIC VOLUME AND HEMATOCRIT AFTER PHENYLEPHRINE AND ACEPROMAZINE ADMINISTRATION Daniela Casoni, Olivia Lorello, Natalie Heliczer, Cristobal Navas de Solis University of Bern, Bern, Bern, Switzerland The objective was to determine correlations of invasive blood pressure (IBP), splenic volume and hematocrit in standing horses after administration of drugs that cause hypertension, splenic contraction and increase in hematocrit (phenylephrine) and hypotension, splenic relaxation and decrease in hematocrit (acepromazine). IBP, ultrasonographic splenic volume and hematocrit were measured in 8 mares before and after phenylephrine (3 μg/kg/min IV for 15 minute) and acepromazine (0.05 mg/kg IV). IBP was measured using the facial or transverse facial artery. Differences between time points were assessed using paired t tests and correlations using Pearson's product correlation. The values before intervention were (mean±SD) 16640 ± 4994 mL for splenic volume, 171 ± 26 mmHg for systolic, 111 ± 17 mmHg for diastolic, 133 ± 20 mmHg for mean IBP and 37.9 ± 3.4% for PCV. Phenylephrine caused a significant decrease in splenic volume (2797 ± 618 mL) and significant increase in systolic (237 ± 18 mmHg), diastolic (148 ± 16 mmHg) and mean (181 ± 12 mmHg) IBP, and hematocrit (50.9 ± 2.1%) (P < 0.01). Acepromazine induced a significant decrease in systolic (118 ± 9 mmHg), diastolic (73 ± 7 mmHg) and mean (88 ± 7 mmHg) IBP (P < 0.01) but non‐significant increase in splenic volume (18113 ± 2812 mL) and decrease in hematocrit (34.6 ± 4.4%) (P > 0.1). Baseline splenic volume was correlated with mean IBP after phenylephrine administration (P = 0.03, R = 0.8), in addition to a trend for the correlation of baseline splenic volume with the hematocrit after phenylephrine (P = 0.06, R = 0.73). IBP and hematocrit after phenylephrine (P = 0.05, r = 0.75) were correlated. Changes in blood pressure and hematocrit after phenylephrine administration can be partially explained by changes in splenic volume as previously suggested. Smaller changes and non‐significant correlations were found for the changes after acepromazine administration. E37 COMPARISON OF MORPHOMETRIC DATA AND PROXY MEASUREMENTS OF INSULIN RESISTANCE TO MINIMAL MODEL ANALYSIS IN ADULT LIGHT‐BREED HORSES Laura Dunbar, Kaitlin Mielnicki, Katarzyna Dembek, Ramiro Toribio, Teresa Burns The Ohio State University, Columbus, OH, USA Proxy measurements calculated from baseline insulin and glucose concentrations have been correlated to Minimal Model analysis of the insulin‐modified frequently sampled intravenous glucose tolerance test (FSIGTT) in horses and humans, but correlation of these values with several morphometric parameters measured in the same cohort of equids has not been reported. This study compared morphometric data and proxy measurements from insulin resistant (IR) and insulin sensitive (IS) light‐breed horses with results of the FSIGTT. Several morphometric measurements were obtained from 12 adult light‐breed horses in which Minimal Model analysis of the FSIGTT was performed. Proxy measurements of insulin resistance (reciprocal of the square root of insulin [RISQI], modified insulin to glucose ratio [MIRG], quantitative insulin sensitivity check index [QUICKI], insulin to glucose ratio [I:G], and homeostasis model assessment [HOMA]) were calculated. Body condition score (BCS), cresty neck score (CNS), mean neck circumference (MNC), and tailhead fat depth (TFD) were significantly correlated with proxy measurements. However, no significant correlation was observed between morphometric data and insulin sensitivity (SI), and no significant difference was observed between IR and IS horses with respect to morphometric parameters. Significant correlations were observed between SI and QUICKI, RISQI, I:G, HOMA, and basal insulin concentration, and acute insulin response to glucose (AIRg) was significantly correlated to MIRG. While morphometric data was not different between IR and IS horses, many morphometric measurements correlated with proxy measurements of insulin sensitivity. CNS displayed the strongest correlation, consistent with its reported utility in identification of horses at risk for Equine Metabolic Syndrome. E38 HAIR CORTISOL IN HORSES WITH PPID Heidi Banse 1, Michel Levy1, Fitsum Getachew1, Alycia Chrenek2, Candice Crosby3, Casey Gruber2, Suzon Schaal3, Ashley Whitehead1, Judit Smits1 1University of Calgary, Calgary, AB, Canada, 2Moore Equine, Rocky View County, AB, Canada, 3TD Equine Veterinary Group, Calgary, AB, Canada Pituitary pars intermedia dysfunction (PPID) is the most common endocrine disorder of aged horses, affecting 15–30% of horses >15–20 years of age. A diagnostic test with good sensitivity and specificity in early disease remains to be identified. Cortisol accumulates in hair and reflects chronic cortisol levels (i.e., over the entire period of hair growth). Therefore, hair may be more useful than other matrices in evaluating chronic hypercortisolemia. Hair cortisol concentrations have been shown to be higher in both people and dogs with hyperadrenocorticism. We hypothesized that horses with PPID have higher hair cortisol concentrations compared to aged control horses. For this study, six horses with PPID and six control horses were included. Horses were considered to have PPID if resting ACTH concentration (in fall) was >11 pmol/L (range, 24–354) and exhibited hypertrichosis or delayed shedding. Horses were considered to be healthy aged horses with ACTH concentrations <11 pmol/L (range, 3.7–10.8) with no evidence of hypertrichosis or delayed shedding. Hair was collected from the jugular furrow using a #40 clipper blade between late August and late September. Hair cortisol was measured using an enzyme immunoassay. Hair cortisol concentrations were not normally distributed. A Mann‐Whitney U test was used to compare hair cortisol concentrations between aged control horses and PPID‐affected horses. No difference in hair cortisol was identified between horses with PPID and aged control horses (P = 0.21). However, when hair from the PPID horses was divided into axial and proximal segments, the proximal segments of hair from horses with PPID (median, 11.2; IQR 6.7–16 pg/mL) was higher than the entire fragment of hair from the aged control horses (median, 5.6; IQR, 5.3–7.5; P = 0.04). The findings of increased cortisol in the proximal fragment of horses with PPID may be a reflection of enhanced seasonal accumulation of hair cortisol. E39 DYNAMICS OF ADRENAL STEROIDS, STEROID PRECURSORS AND NEUROSTEROIDS IN NEONATAL FOALS DURING HOSPITALIZATION Katarzyna Dembek 1, Lindsey Johnson1, Kathryn Timko1, Kathy Macgillivray2, Bonnie Barr3, Kelsey Hart4, Ramiro Toribio1 1Department of Veterinary Clinical Sciences, The Ohio State University, Columbus, OH, USA, 2Hagyard Equine Medical Institute, Lexington, KT, USA, 3Rood and Riddle Equine Hospital, Lexington, KT, USA, 4College of Veterinary Medicine, University of Georgia, Athens, GA, USA Sepsis is a major cause of morbidity and mortality in neonatal foals. Relative adrenal insufficiency (RAI), defined as an inadequately low cortisol concentration for the severity of disease, has been associated with sepsis in newborn foals. In addition to cortisol, adrenal gland synthesizes a number of biologically important steroids, steroid precursors and neurosteroids. However, the dynamics of multiple adrenocortical hormones during hospitalization and their association with severity of disease and mortality in critically ill foals have not been evaluated. We hypothesized that in addition to cortisol, foals with RAI will have altered concentrations of adrenal steroids and their precursors. We also proposed that septic foals will have higher concentrations of adrenal steroids than healthy foals and steroid concentrations will be persistently increased during hospitalization in non‐surviving septic foals. Foals <3 days of age were categorized into three groups: 29 septic, 22 sick non‐septic (SNS) and 10 healthy. Blood samples were collected on admission (Time 0), 24 hour (Time 24 hour) and 72 hour post‐admission (Time 72 hour). Concentrations of cortisol, progesterone, pregnenolone, aldosterone, dehydroepiandrosterone (DHEA), 17 α‐OH‐progesterone and androstenedione were measured by immunoassays. Cortisol, progesterone, pregnenolone, aldosterone, DHEA, and 17 α‐OH‐progesterone were higher in septic compared to healthy foals, at Time 0 (P < 0.01). In septic foals at Time 0, DHEA and cortisol were also significantly higher than in SNS foals (P < 0.05). Cortisol and DHEA were elevated and 17 α‐OH‐progesterone lower in septic compared to healthy foals at 72 hour (P < 0.05). At Times 24 and 72 hour, cortisol concentrations increased by 44% and 51% in non‐surviving foals but decreased by 63% (−63%) and 204% (−204%) in surviving foals, respectively. Critically ill neonatal foals had an appropriate response to sepsis associated stress characterized by increased concentrations of steroids, steroid precursors and neurosteroids on admission. Decreased steroid concentrations in all groups can be attributed to adrenal dysfunction (septic foals) as well as physiological post‐natal steroidogenesis (healthy foals). Persistently increased cortisol concentration was associated with non‐survival in septic foals. E40 REPEATABILITY OF THE ORAL SUGAR TEST AND GLUCAGON‐LIKE PEPTIDE‐1 AND HIGH MOLECULAR WEIGHT ADIPONECTIN CONCENTRATIONS IN HORSES WITH EQUINE METABOLIC SYNDROME Nicholas Frank 1,2, Donald Walsh1,2 1Cummings School of Veterinary Medicine at Tufts University, North Grafton, MA, USA, 2Homestead Veterinary Hospital, Pacific, MO, USA Insulin status is assessed in horses by performing an oral sugar test (OST) and incretin hormones and adipokines can also be measured to investigate mechanisms of hyperinsulinemia. However, the repeatability of these measures has not been determined. We therefore hypothesized that glucose, insulin, active (aGLP‐1) and total glucagon‐like peptide 1 (tGLP‐1), and high molecular weight (HMW) adiponectin results from oral sugar teats would be repeatable, as defined by a within‐horse coefficient of variation (CV) <20%. Twenty‐three horses with suspected EMS from a Tennessee research facility were evaluated, along with 15 horses with suspected EMS and 15 controls from a private practice in Missouri. Two oral sugar tests were performed 7–14 days apart under fasting conditions and plasma glucose and insulin concentrations were measured at 0, 60, and 75 minute. Incretin hormones and HMW adiponectin were measured. Within‐horse CV values and measures of agreement were calculated. Median (range) within‐horse CV values were 6.9% (0.0–79.2%) and 15.3% (0.3–78.5%) for glucose and insulin concentrations, respectively. Active GLP‐1 (r s = 0.78; P < 0.001) and tGLP‐1 (r s = 0.81; P < 0.001) concentrations at 75 minute and resting HMW adiponectin concentrations (r = 0.99; P < 0.001) were positively correlated between days. There was 91% and 83% agreement in test interpretation for insulin results in Tennessee and Missouri groups, respectively. Our hypothesis was supported and we conclude that OST results have acceptable repeatability. Incretin hormone concentrations were not as strongly correlated between test days as resting HMW adiponectin concentrations. E41 CORTISOL BINDING GLOBULIN CONCENTRATIONS IN SEPTIC FOALS, HEALTHY FOALS, AND HEALTHY HORSES Melanie Fratto, Kelsey Hart, Natalie Norton, Madison Berger University of Georgia, Athens, GA, USA Cortisol binding globulin (CBG) binds circulating cortisol in plasma, and regulates the amount of free cortisol available in circulation by influencing metabolic clearance rate. Free cortisol is the only form that is able to enter cells to bind cytosolic cortisol receptors, and thus has long been presumed to be the biologically active form of cortisol. However, recent research in rodents and humans suggests CBG‐bound cortisol may interact with cell‐surface receptors on neutrophils to mediate some of cortisol's anti‐inflammatory effects. Previous research has shown that neonatal foals have dramatically increased serum free cortisol levels compared to adult horses, but many septic neonatal foals still exhibit exaggerated pro‐inflammatory responses consistent with critical‐illness related cortisol insufficiency (CIRCI). This disparity between increased circulating free cortisol but clinical evidence of cortisol insufficiency in some septic neonatal foals could be explained by limited CBG availability in foals, if CBG‐bound cortisol is important for regulating some of cortisol's anti‐inflammatory effects in foals as in other species. However, our current understanding of cortisol binding dynamics in neonatal foals is limited, and measurement of CBG concentrations in septic foals has not been reported to date. The objective of this study was to compare CBG concentrations among healthy foals of various ages, septic foals, and healthy adult horses. We hypothesized that sick foals have significantly lower CBG concentrations than healthy foals or adult horses, and that healthy foals have significantly lower CBG concentrations than adult horses, with the lowest concentrations being at birth. CBG concentrations in serum samples were measured using a commercial equine CBG ELISA validated for this study (% recovery = 92.86%, intra‐assay coefficient of variation = 8.80%, inter‐assay coefficient of variation = 7.32%). Healthy foals (n = 7) were sampled at birth, 36–48 hour, 1 week, 2 weeks, and 8 weeks, and healthy horses (n = 7) were sampled once. For the septic foal group (n = 29), archived serum samples collected from sick foals ≤7 days of age with a diagnosis of sepsis based on a positive blood culture and/or sepsis score ≥ 11 at hospital admission were used. Data were compared with ANOVAs, with statistical significance set at P < 0.05. There was no significant effect of age on CBG concentration in healthy foals (P = 0.390). Healthy foals at all ages had significantly lower serum CBG concentrations than adult horses (P < 0.001). Septic foals had significantly lower serum CBG concentrations than adult horses or healthy age‐matched foals (P < 0.001). These results support our hypothesis that septic foals exhibit CBG deficiency compared to healthy foals and adult horses. CBG‐deficiency results in decreased circulating CBG‐bound cortisol in foals, and if CBG‐bound cortisol is critical for mediating some of cortisol's anti‐inflammatory effects in foals as in other species, this CBG‐deficiency could contribute to the exaggerated and potentially detrimental pro‐inflammatory responses seen in some septic foals with CIRCI. Further prospective clinical and experimental studies are needed to confirm these findings in larger numbers of septic foals and to investigate the effects of CBG‐bound cortisol on equine leukocyte function. E42 CLINICAL SIGNS ASSOCIATED WITH PPID STATUS IN A LARGE POPULATION OF HORSES Steven Grubbs 1, Dwana Neal1, Thomas Keefe2 1Boehringer Ingelheim Vetmedica, Inc, St. Joseph, MO, USA, 2Colorado State University, Fort Collins, CO, USA Pituitary pars intermedia dysfunction (PPID) has been considered the most common endocrinologic disorder of aged horses. Few studies exist that describe the epidemiological characteristics of horses with PPID. Additional epidemiologic studies are needed to determine the prevalence of PPID in a larger population, not just aged horses. The purpose of this study was to obtain epidemiological information at initial PPID diagnosis (new cases) from a large population of horses that included age, breed, sex, clinical signs, and insulin/glucose status. Horses of any age, breed, and sex from the continental US were eligible for study enrollment as long as they were documented to be exhibiting one or more of the following clinical signs: generalized or regional hypertrichosis, muscle wasting, abnormal fat distribution, lethargy, laminitis (unknown etiology), polyuria, polydipsia, susceptibility to infections, abnormal sweating, and/or inappropriate lactation. Normal horses were excluded from the study. At initial visit, demographic data, signalment, a physical examination was conducted, clinical signs documented, and blood was drawn for basal ACTH, fasting insulin, and glucose. Blood samples were processed and shipped overnight to the Animal Health Diagnostic Center, Cornell University, Ithaca, NY for analysis. Non‐PPID horses (PPID‐) were characterized with ACTH levels <35 pg/mL (November‐July) and ACTH <100 pg/mL (August‐October). PPID+ horses were characterized with ACTH levels >35 pg/mL (November through July) and ACTH >100 pg/mL (August‐October). Hyperinsulinemia was characterized as fasting insulin >20 ulU/mL. Glucose was considered to be elevated if >113 mg/mL. In order to evaluate the combined effects of these variables on PPID status, multiple logistic regression (MLR) analysis was applied to PPID status using backward elimination of variables at the 5% level of significance. Odds ratios for the significant predictors of PPID status in the MLR analyses were computed along with their corresponding 95% confidence intervals. The prevalence of PPID was significantly (P < 0.015) greater among horses found to have 5 of the 11 clinical signs present: abnormal sweating, decreased athletic behavior, delayed shedding, loss of muscle mass, and weight loss. Although not statistically significant (P > 0.25), PPID prevalence was greater when three of the remaining clinical signs was present (excessive thirst, excessive urination, and laminitis) and was actually less when four of the remaining signs was absent (cresty neck, fat pads, pot belly, weight gain, and recurrent infections). Based on only the data on clinical signs, both delayed shedding and weight loss were found to be significantly associated with PPID status (P ≤ 0.002). Specifically, the odds of PPID among horses showing delayed shedding was four times that for horses not showing delayed shedding, and the odds of PPID among horses showing weight loss was approximately twice (1.9) among horses not showing weight loss. Based on the combined data on demographic variables, insulin levels, and all eleven interpretable clinical signs, age and insulin status were still seen to be significantly associated with PPID status, but only one clinical sign, delayed shedding, was found to be a significant predictor of PPID after accounting for age and insulin status. The odds ratio of PPID for horses showing delayed shedding (3.4) decreased only slightly after accounting for age and insulin status of the horses. Based on the data from this population of horses, delayed hair coat shedding and weight loss were significant predictors of PPID status. The prevalence of PPID was also significantly greater when certain clusters of clinical signs were present. Long term studies need to be conducted to further evaluate the occurrence and progression of clinical signs in horses with PPID and other endocrine diseases. E43 EVALUATION OF WELLBEING PARAMETERS IN NEW YORK CITY CARRIAGE HORSES Sarah Mercer‐Bowyer, David Kersey, Joseph Bertone College of Veterinary Medicine, Western University of Health Sciences, Pomona, CA, USA The purpose of this study was to provide an objective evaluation of the wellbeing of a population of working horses. This study investigated wellbeing parameters in a population of working New York City carriage horses. Samples were collected over the course of a 3 day period (August 3–5, 2014). The parameters measured included fecal cortisol, salivary cortisol, and infrared thermography (IRT) of the medial canthus. The parameters used in this study have all been utilized in previous investigations to quantify stress levels in horses in a variety of settings. Collection Day 1, 2 and 3 involved 11, 7 and 8 horses, respectively. Collection days included four collection time points. At time point 1 (TP1), subjects were at rest in their stalls prior to work or the arrival of employees (06:00–08:00am EST). TP2 occurred as harnessed subjects were being hitched to carriages and prepared for work. TP3 occurred immediately after being placed in their stall at conclusion of work day. TP4 occurred 1 hour after horses were returned to their stall. Fecal cortisol samples were collected at TP1 and are used to evaluate chronic glucocorticoid levels. Salivary cortisol and IRT were collected at all four time points and reflect a more acute state of wellbeing. Preliminary evaluation of the salivary cortisol show differences among the time points (χ2 3 = 8.7; P = 0.03) with TP3 (0.96 ± 0.06 ng/mL) being greater (t 25 = 2.5; P = 0.02) than TP4 (0.77 ± 0.07 ng/mL); however, all other time point comparisons were not different (P > 0.05). For the IRT, student t‐test indicated that measures between the left and right eye did not differ within time assessments (T = 141.5–144.512,12; U = 63.5–66.5; P = 0.644–0.773); therefore measures between left and right eye were averaged for comparisons of eye changes among the sampling time points. A One Way Repeated Measures Analysis of Variance revealed no difference (F = 0.114; P 0.952) among time points (TP1‐4) eye measures. A similar trend was not evident between the cortisol and IRT data sets. The salivary cortisol variation among TP3 and 4 is a reflection of increased metabolic activity driven by exercise rather than stress. Stress as the driving factor behind salivary cortisol increase would have resulted in a similar increase in IRT which was not observed. Evidence collected thus far indicate no significant perturbation in the state of wellbeing of this population of working horses. E44 PROSPECTIVE EVALUATION OF A COLIC PREDICTION MODEL Frank Andrews 1, Jonuel Cruz‐Sanabria1, Mustajab Mirza1, Kimberly Pescosolido1, Adele Groue2, Michael Kearney1 1Louisiana State University, Baton Rouge, LA, USA, 2Ecole Nationale Veterinaire de Toulouse, Toulouse, France Colic is common in horses. Accurate and timely interpretation of clinical and laboratory data at admission are important to determine the need for surgery and the probability of survival. Many studies have evaluated clinical and laboratory findings as indicators for surgery and survival, but few studies were as comprehensive as published by Reeves et al in 1989 and 1990.1,2 The model was based on a retrospective evaluation of medical records from horses presenting to referral institutions. The model was validated using a subpopulation of horses taken from the medical records, but was never validated in a prospective population of horses with acute colic. The purpose of this study was to prospectively validate surgical and prognosis colic prediction models.1,2 We hypothesized that the previously developed surgical and prognosis colic model will accurately predict the probability of surgery and survival in horses presented to Louisiana State University Veterinary Teaching Hospital (LSU‐VTH) with acute colic. Horses presenting to the LSU‐VTH with acute colic from 8‐20‐12 to 7‐13‐13 were studied. Clinical parameters including age, sex, breed, and rectal exam (normal or abnormal), frequency of abdominal pain (none, intermittent, or continuous), peripheral pulse character (normal or abnormal) and frequency of abdominal sounds (normal, decreased, absent, increased) were used to validate the surgical model; whereas peripheral pulse character, packed cell volume, heart rate, capillary refill time and medical vs. surgical treatment were used to validate the survival (prognosis) model. Horses that were euthanatized due to financial constraints were excluded from the survival model. Survival was defined as discharged from the VTH. Bayes theorem was utilized that calculated the post‐test probability of both surgery and survival. The hosmer‐Lemeshow goodness of fit chi square (GOFCS formula was used to assess if each model fit the data. A chi square value of >15.51 using 8 degrees of freedom was considered significant if it yielded a P < 0.05. Seventy‐one horses presenting with acute colic were included in the study. Thirty‐five of the 71 horses (49%) were admitted from June 1 through August 1, 2014. Of the 71 horses evaluated for acute colic, 25/71 ((35%) had surgery and 46/71 (65%) were treated medically. Also, 17/71 (24%) of horses were humanely euthanatized and 1/25 (4%) died after surgery. In addition, 6/71 horses were euthanatized at surgery due to financial reasons. Therefore, 53/65 horses (82%) survived to discharge. The parameters used in the model accurately estimated surgery (5.18; P < 0.05), however, the model did not accurately estimate of the probability of survival (15.88; P > 0.05), because only one horse admitted to the study died. Parameters, including an abnormal rectal examination, continuous or intermittent abdominal pain, decreased or absent abdominal sounds and weak peripheral pulse character, successful identified horses needing surgery in this population of horses. Whereas, the parameters used for survival failed to predict survival. Clinical parameters are important in predicting those horses that need surgical treatment for acute colic. Whereas, cardiovascular parameters may accurately predict survival in horses with acute colic but this could not be proven in the study presented here because only one horse died after treatment. 1Reeves MJ, Curtis CR, Salman MD, et al Prognosis in equine colic patients using multivariable analysis. Can Vet J Res 1989; 53:87–94., 2Reeves, M.J., Curtis C.R., Salman, M.D., et al. (1990) A Multivariable prognostic model for equine colic patients. Preventative Veterinary Medicine 9:241–257. E45 DIAGNOSTIC VALUE OF GASTRIC MUCOSAL BIOPSIES IN HORSES WITH GLANDULAR DISEASE Samantha Crumpton, Kerstin Baiker, Jocelyn Habershon‐Butcher, Gayle Hallowell, Mark Bowen School of Veterinary Medicine and Science, University of Nottingham, Nottingham, UK Equine Gastric Glandular Ulceration Syndrome (EGGUS) is a highly prevalent condition, for which the underlying pathophysiology is largely unresolved. Endoscopic mucosal biopsies have been proposed as a method for adapting therapy for individual horses. The aim of this study is to evaluate the diagnostic information obtained from endoscopic mucosal biopsies in horses. Twenty‐one clinically normal horses undergoing elective humane slaughter were subjected to gross examination of the glandular mucosa post‐mortem. Glandular pathology was graded using Equine Gastric Ulcer Council guidelines from images obtained using a digital camera. Mucosal biopsies were obtained using a ‘single‐bite’ technique using fenestrated endoscopic biopsy forceps that were 1.8 mm (Group A), 2.4 mm (group B) or a ‘double bite’ technique using the 2.4 mm instruments. Full thickness biopsies were also obtained from adjacent tissue to the biopsy site that included any lesions identified using the naked eye. All tissue was formal fixed, processed and stained using standard protocols. Histological appearance of the lesions was determined from each sample type and sample quality and artefacts were recorded. The appearance of inflammatory infiltrates was graded as mild, moderate or severe. Degree of inflammation was compared to ulcer grade. For a subset of animals that had shown moderate or severe gastritis from the full thickness sample (n = 5), inflammatory cell counts per high power field (HPF) were compared between full thickness biopsies and each of the samples obtained using endoscopic instruments. Agreement between cell counts from full thickness samples and mucosal biopsies were subsequently compared by calculation of intra‐class correlation coefficient (ICC). EGGUS was present in 12 horses (Table 1) Samples from three animals were excluded from analysis due to processing or sampling errors. Full thickness samples were free of artefact and allowed visualisation of mucosa, submucosa, glands, and villi. Mucosal biopsy samples contained mucosa in all samples, submucosa in 55% (group C), 61% (group A) and 66% (group B) of samples, glands in 50% (group B), 66% (group A) and 100% (group C). Group A samples were often too small for histological assessment (33%) and tissue damage was more commonly seen in group A and B (n = 8 and n = 10) compared to group C (n = 3). Horses with normal glandular appearance (EGGUS grade 0; n = 7) often demonstrated mild gastritis (n = 5) while only one ‘normal’ animal had no inflammation and one had moderate gastritis. Severe gastritis was identified in animals with mild EGGUS (grade 1 or 2), while mild and moderate gastritis was identified in all EGGUS grades. There was no histological evidence of ulceration or erosion. There was poor agreement between cell numbers between sampling techniques (ICC <0.29). EGGUS Grade Number of horses No gastritis Mild gastritis Moderate gastritis Severe gastritis Excluded from study 0 9 1 5 1 0 2 1 3 0 2 0 1 0 2 4 0 0 2 1 0 3 3 0 1 2 0 1 4 2 0 1 1 0 0 Table 1: Occurrence of EGGUS in 21 horses and presence of histological appearance of gastritis in these animals. Three horses were excluded from histological analysis due to processing or sampling errors. EGGUS was common in this group of horses, consistent with previous studies. These data shows lack of ulcerative pathology in horses where EGGUS is identified, which actually represents inflammation. Nomenclature should be reviewed to reflect this underlying pathology. Lesion appearance is a poor indicator of underlying severity, and lack of gross lesions should not be viewed as lacking any gastric pathology. Gastric mucosal biopsies offer limited value in predicting underlying disease in horses with EGGUS, however a double bite technique, with largest possible biopsy instruments should provide the highest quality samples for histological review. E46 WATER IN INGESTA INFLUENCES BLOOD GLUCOSE CONCENTRATION AFTER EATING IN THE HORSE Michael Katz, David Lewis University of Massachusetts, Amherst, MA, USA Osmoregulation by the intestine is one of the controlling mechanisms in gastric emptying. This study is an attempt to examine the effect of water content of ingested food on intestinal glucose absorption. Six miniature horses with an average body weight of 135 kilos were fed 3 pounds of timothy hay pellets after an 8‐hour fast. On a different day they were fed the same amount of hay pellets mixed with 4.5 pints of water. Rate of consumption of the pellets was the same whether the pellets were fed dry or moist. Frequent samples were taken and whole blood glucose was measured during the hour before feeding and then every fifteen minutes for 2 hours. (Pellet composition: moisture 7.2%, water soluble carbohydrate 12.5%, ethanol soluble carbohydrate 7.0%, starch 1.4%.) Data were analyzed using Student's t‐test. When fed dry or moist pellets there was an increase in blood glucose within fifteen minutes after the onset of feeding. There was a significantly larger increase in blood glucose when fed moistened pellets than when fed dry pellets: the glucose concentration‐time curve (AUC) was significantly (16% – 100%) larger when fed moistened pellets. Water in the food seems to encourage more rapid presentation of glucose to the intestine from the stomach, which may indicate an increased rate in gastric emptying. E47 PREVALENCE OF AND RISK FACTORS FOR GASTRIC ULCERATION IN POLO HORSES Heath MacLeod 1, Claire Windeyer1, Candice Crosby1, Heidi Banse2 1University of Calgary, Calgary, AB, Canada, 2TD Equine Veterinary Group, Calgary, AB, Canada Gastric ulcer prevalence has been evaluated in many equine performance disciplines. Polo horses are a unique equine population that are often housed and exercised in groups. Therefore, ulcer occurrence and risk factors may differ in this performance discipline compared to previously studied populations. The purpose of this study was to determine the prevalence of gastric ulceration in polo horses and identify potential risk factors for squamous and glandular ulceration. Sixty‐three polo horses from nine barns were included in this study. All gastroscopies were performed between June and August, 2014. Grooms filled out a survey on management practices, including feeding, housing, training, competition level, and medications administered. Horses underwent gastroscopy following a 16 hour fast. Sixty‐nine percent of horses had glandular ulcers, and 54% of horses had squamous lesions. Ulcers were dichotomized into grade ≥2 and grade <2 for statistical analysis. Thirty one percent of horses had grade ≥2 or glandular ulcers, while 37% of horses had grade ≥2 squamous ulceration. Preliminary analysis was performed using a Chi square test to identify possible risk factors for gastric ulcers. Ulcers were dichotomized into grade ≥2 and grade <2 for analysis. There was no association between the presence of grade ≥2 squamous and glandular ulceration (P = 0.07). Horses performing at a >6 goal level had a lower prevalence of glandular (P = 0.04) and squamous (P = 0.02) ulcers than those performing at < 6 goal. Paddock turnout was also associated with a decreased risk of squamous ulcers (P = 0.03). In this study, only housing factors and performance level were associated with the presence of gastric ulcers. The increased prevalence of squamous and glandular ulceration in low goal polo horses suggests that gastric ulceration may be associated with reduced performance. E48 A COMPARISON OF GASTRIC ULCER PREVALENCE IN FERAL AND DOMESTICATED HORSES IN THE UK: AN ABATTOIR STUDY Sophia Ward1, Benjamin Sykes 2, Hieke Brown1, Annie Bishop1, Lee‐Ann Penaluna1 1Oxford Brookes University, Oxfordshire, UK, 2BW Sykes Consultancy, Upper Orara, NSW, Australia Equine squamous gastric disease (ESGD) has been reported in 50–100% of performance horses in various studies while the prevalence of equine glandular gastric disease (EGGD) has been reported as being between 30–65% in a number of different horse types. To date little information is published on the prevalence of either in feral horses. The purpose of this study was to compare the prevalences of ESGD and EGGD between feral and domesticated horses presenting to an abattoir for slaughter. Horses included in this study were randomly selected during two separate visits to a certified abattoir in the United Kingdom (UK) between the months of June and August. Horses were classified as feral on the basis that they were free ranging with little or no human interaction, and classed as domesticated on the basis that they had been under the care of humans with a designated purpose as assessed by an abattoir employee. The feral population originated from the Dartmoor and Exmoor regions of the UK. History on the horses such as age, sex, breed or discipline for the horses was not available. Once classified as feral or domesticated horses were slaughtered in accordance with appropriate UK legislation. Following slaughter, the gastrointestinal tract was removed and the stomachs separated by an abattoir employee. Each stomach was then dissected along the dorsal plane from the lower oesophageal sphincter to the pyloric sphincter and rinsed with water. Photographs were captured for each stomach and numbered for future analysis. Grading of the gastric lesions was performed by a single investigator blinded to the group allocation of the horse. Lesions were graded using the four point EGUS council system with the squamous and glandular regions of the stomachs graded separately. Horses were dichotomously classified as being ESGD or EGGD positive or negative if the assigned grade for the squamous of glandular mucosa was ≥ II/IV or ≤ I/IV, respectively. A Chi‐Squared test was used to compare the prevalence of ESGD and EGGD between feral and domesticated horses. Significance was set at P ≤ 0.05. A total of 60 domesticated horses and 29 feral horses were randomly selected. Data from two feral and nine domesticated horses was excluded due to inconclusive photographic documentation. Fifty‐one domesticated and 27 feral horses were included in the final analysis. The lesion distribution for ESGD and EGGD are shown in the table below. When dichotomously classified, ESGD was present in 22.2% and 60.8% of feral and domesticated horses, respectively (P = 0.001). Similarly, EGGD was present in 29.6% and 70.6% of feral and domesticated horses, respectively (P ≤ 0.001). Horse Type Lesion Type Grade 0/IV Grade I/IV Grade II/IV Grade III/IV Grade IV/IV Domesticated ESGD 18 2 25 5 1 EGGD 11 4 25 10 1 Feral ESGD 20 1 5 1 0 EGGD 19 0 4 4 0 The results of this study suggest that the prevalences of both ESGD and EGGD are higher in domesticated horses than feral horses. While it is well recognized that the risk of ESGD increases with increased intensity of management this report is, to the authors' knowledge, the first report documenting that a similar effect may be present for EGGD. E49 THE USE OF AN INDWELLING PERCUTANEOUS GASTROTOMY TUBE FOR THE MEASUREMENT OF INTRA‐GASTRIC PH IN THE HORSE Benjamin Sykes 1, Claire Underwood1, Catherine McGowan2, Paul Mills1 1School of Veterinary Sciences, The University of Queensland, Gatton, Qld, Australia, 2Institute of ageing and Chronic Disease, University of Liverpool, Liverpool, Liverpool, UK The purpose of this report is to describe a method of intra‐gastric pH measurement using a pH probe connected to a continuous data logger that is fitted in a retrograde manner into an endoscopically placed, indwelling percutaneous gastrotomy tube (PEG) in the horse. Two healthy adult Thoroughbred horses were used. Horses were fasted for 20 hours and sedated with detomidine (10–20 μg/kg IV). Gastroscopy was performed as per standard protocol and the stomach fully insufflated with air. The stomach was identified by percutaneous ultrasound at the 11th inter‐costal space. A 14 G × 5.25” catheter (Angiocath, Becton Dickinson, Utah, USA) was inserted through the skin and into the stomach until it could be observed endoscopically. A 3 m length of high tensile fishing line was passed through the catheter from the outside and into the stomach. Biopsy forceps were used to grasp the fishing line and draw it into the endoscope which was then withdrawn, drawing the fishing line out the nostril. A 20 mm button was attached to the fishing line. Gentle traction was then placed on the fishing line to draw the button back through the nostril and down the esophagus and into the stomach. The endoscope was reinserted to confirm the positioning of the button against the stomach wall. A second button was placed approximately 3–5 cm away from the first button using the same technique. The ends of the two lines were then tied together to create a gastropexy, similar to the abomasal toggle technique described in cattle. A PEG tube was then placed endoscopically using a similar technique, as previously described. Horses received antimicrobials and anti‐inflammatories for 10 days peri‐operatively. Three weeks following the PEG tube placement their position was confirmed endoscopically. In both horses the tube was located within the ventral glandular fundus approximately 10 cm below the margo plicatus adjacent to the lesser curvature. At this point a pH probe (Comfortec PLUS, Sandhill Scientific, CO, USA) was fitted retrograde through the PEG tube with its location confirmed endoscopically. The probe's insertion distance was noted for future placement. The horses were then adapted to one of two diets; a hay only diet (HAY) or a high grain/low fiber (HG/LF) diet. The HG/LF diet consisted of 1% BW each of grain and hay per day, divided equally into two meals. Each horse was then monitored on their diet for a period of 5 consecutive days. To achieve this; the pH probe was attached to a continuous data logger (Zephyr pH, Sandhill Scientific, CO, USA). The probe was calibrated prior to placement each day and then inserted the previously determined distance and secured. Data recording commenced at 8 am and continued for 23 hours each day. The probes contained two measurement points, 5 cm apart. The location of probe 1 was approximately 1 cm from the glandular mucosa while the second probe sat 5 cm deeper, presumably within the ingesta. Data (median (IQR)) for each horse is shown below. Horse Diet Probe 1 (Mucosal) Probe 2 (Ingesta) Mean pH Median pH % time <4 Mean pH Median pH % time <4 1 HAY 1.8 (1.6:1.9) 1.6 (1.6:1.8) 100 (97:100) 1.9 (1.7:2.0) 1.9 (1.7:2.0) 100 (99:100) 2 HG/LF 1.9 (1.9:1.9) 1.6 (1.5:1.7) 92 (91:94) 1.8 (1.5:2.2) 1.6 (1.4:1.9) 94 (91:98) This report describes the successful instrumentation of horses with permanent PEG tubes and their use for the measurement of intra‐gastric pH using a technique that allows measurement at multiple points within the stomach. Further, the technique allows for the monitoring of intragastric pH without disturbance to the horse's normal eating behavior, a major limitation of many other models. The procedure can be performed in the standing, sedated horse with minimal specialized equipment. E50 SEVERE MYELOSUPPRESSION AND ALOPECIA IN HORSES ASSOCIATED WITH TRICHOTHECENE INGESTION Rachel Liepman 1, Nina Zitzer1, M. Judith Radin1, Michael Geiger2, Teresa Burns1 1The Ohio State University, Columbus, OH, USA, 2Horizon Equine Veterinary Service, Wooster, OH, USA This report describes severe myelosuppression in four American Quarter Horses associated with ingestion of mycotoxin‐contaminated hay. The index case (11 year old mare) presented with lethargy, acute perineal edema, hematochezia, fever, patchy alopecia, and petechiation. The horses had recently been fed a new batch of hay (which was later revealed to be improperly cured), which initially caused feed aversion. Bloodwork from the mare showed severe neutropenia and thrombocytopenia and prolonged prothrombin time. Profound myeloid and megakaryocytic hypoplasia was found on evaluation of bone marrow. The mare developed septic peritonitis and was treated aggressively for 8 days with anti‐inflammatories, antibiotics, and blood transfusions prior to acute decompensation necessitating euthanasia. Of the remaining horses, two developed similar hematologic findings and all three had alopecia. A second case (19 year old gelding) was referred for treatment with milder signs of petechiation and patchy alopecia and was noted to have similar bone marrow findings to the mare; this horse recovered after treatment with anti‐inflammatories and antibiotics. The two remaining horses (8‐ and 6‐year old mares) were managed conservatively and recovered. Toxicological testing on the hay revealed high levels of mycotoxins, including T‐2 toxin and vomitoxin (both known myelotoxicants; 1627 ppb and 1.1 ppm, respectively). The myelosuppression in these cases was consistent with trichothecene intoxication, which is poorly documented both in horses and from a forage source. The clinical course of these cases suggests that spontaneous recovery is possible following severe myelosuppresion due to tricothecene intoxication in horses. E51 PLASMA LACTATE AS A MARKER OF HEALTHY IN FOALS BORN FROM MARES WITH PLACENTITIS Lorena Feijó, Bruna Curcio, Fernanda Pazinato, Bruno Almeida, Carlos Eduardo Nogueira Universidade Federal De Pelotas, Pelotas/Rs, Brazil Placentitis is an important cause of pregnancy loss and neonatal death in mares. Fetal hypoxia and placental abnormality can be associated with hyperlactatemia in foals; however, data on field cases is scant. The objective of this study was to measure lactate in plasma of foals born from mares suffering with placentitis and from mares carrying and delivering normal pregnancies. A total of 58 Thoroughbred newborn foals were included in this study. Immediately after placental passage, fragments of chorioallantois and amnion were collected and preserved in formalin for histopathologic evaluation. Immediately post‐delivery and by 24 hour of birth, each foal underwent clinical examination and blood sampling Plasma lactate was determined by spectrophotometry using a commercial colorimetric kit. Based on placental histopathology and clinical evaluation foals were allocated into four groups as follows: group I (healthy control n = 22) no placental lesions and apparently health foals, group II (placentitis and apparently healthy n = 12), group III (placentitis and abnormal upon clinical examination); and group IV (immature foals and placenta with histopathological lesions compatible with placentiti n = 10). Normality of the data was assessed by Shapiro‐Wilk test. One‐way ANOVA was used to evaluate the lactate concentrations in different groups Significance was set at P < 0.05. Concentrations of lactate were significantly increased in groups III and IV at 24 hours after birth; however, foals in groups I and II concentrations presented a reduction in lactate from birth to 24 hours post‐delivery. Our findings demonstrate that lactate can be used to identify sick foals which screen apparently normal foals born from mares with placentitis. It is important to emphasize that groups I and II presented lactate concentrations within normal ranges, demonstrating full capacity of neonatal adaptation of group II, despite placental pathology. E52 HUMORAL RESPONSE OF BORRELIA BURGDORFERI OUTER SURFACE PROTEIN A (OSPA) VACCINE VIA THE SUBCUTANEOUS AND TRANSDERMAL ROUTE IN HORSES Kathryn Slaughter1, Spring Halland1, Danielle Demel1, Leslie Schur2, Bettina Wagner3, Joseph Bertone 1 1College of Veterinary Medicine, Western University of Health Sciences, Pomona, CA, USA, 2Desert Pines Equine Medical and Surgical Center, Las Vegas, NV, USA, 3Cornell University, Ithaca, NY, USA Borrelia burgdorferi causes Lyme disease via Ixodes scapularis and pacificus tick vectors. A safe and effective vaccine is essential in the prevention of equine Lyme borreliosis. A USDA canine approved outer surface protein A (OspA) vaccine may be promising in this regard and has been used by many equine practitioners. The purpose of this study was to identify the humoral immunogenicity of this vaccine in horsess Forty‐three healthy horses were identified for this study. These horses were selected for their improbable contact with Borrelia burgdorferi due to their life long desert habitat. Serological testing for OspA+C+F was completed prior to vaccination. Group‐TD (20 horses, mixed breed, five females, 15 geldings) was administered 1 mL (1/2 mL/site) vaccine transdermally over the pectoral region. Group–SQ (20 horses, mixed breed, four females and 16 geldings) were administered 2 mL at the right cervical region. Group–C (3 horses, mixed breed 2 females and 1 gelding) were untreated. Vaccine was administered on day 0 and 21, and 225. Antibodies to OspA + C + F were quantified on day −45, 22, 42, 93, 179, 244, 313, 361 using diagnostic laboratory multiplex assay. At day 41 all horses responded. Median Fluorescent Intensity (MFI) was 15829.3 ± 2031.7 (Mean±SE) in the Group‐TD with a 39 ± 9.0 fold increase over baseline. Group‐SQ responded with a 14580.2 ± 1671.1 MFI which translated to a 44.43 ± 7.6 fold increase over baseline. Group‐TD responded at Day 244 (post‐vaccine 3 at 225 days) with 15682.4 ± 2773.38 MFI, which translated to a 44.43 ± 7.6 fold increase over baseline. Group‐SQ responded on Day 244 with 16140.1 ± 2431.1 MFI which translated to a 39.99 ± 6.7 fold increase over baseline. There were no evident differences in OspA between age, sex, breed or route. The lack of difference between the routes would suggest the lesser dose via the transdermal route in the pectoral region elicits an equivalent OspA response as a 2 mL dose administered subcutaneously. Minor reactions (<4 mm swelling) occurred at the vaccination sites in Group‐ TD and Group‐SQ at 24.6% and 34.5% respectively. The reactions occurred most commonly in the same horses at 75% and 53% of the time in TD and SQ groups respectively. This data may aid in the development of an effective vaccine and strategic vaccination protocols for Lyme disease in horses. E53 DEVELOPMENT OF PULMONARY IMMUNOCOMPETENCY IN THE FOAL Brett Sponseller, Shannon Hostetter, Jessica Gilbertie, Douglas Jones, Sandra Clark, Sarah Wiechert, Mahesh Bhandari, Beatrice Sponseller, Cody Alcott, David Wong, Bryan Bellaire, Prashanth Chandramani Iowa State University, Ames, IA, USA Pneumonia is the most common cause of morbidity and mortality in foals <6 months of age. In contrast, pulmonary immunocompeteny in adult horses appears more robust given resistance to the same diseases that cause clinical manifestations in foals. We speculated that there would be quantitative and qualitative differences in the cell populations recovered in bronchoalveolar lavage fluid (BALF) that could provide insight into the progressive age‐related gain in immunocompetency. Serial BALF was collected from a cohort of healthy foals at 1 week; 1, 2, 6, 12 and 18 months of age. BALF from foals was then compared to BALF collected from healthy adult horses. Cellular morphologic evaluation and 300 cell differentials were performed on cytocentrifuged BALF samples. Foals had significantly higher percentages of macrophages than adult horses at 1 week, 1 month, 2 months, and 6 months of age; this difference was no longer noted at 12 months of age. Foals had significantly lower percentages of lymphocytes in BALF than adult horses (week 1, months 1, 2, and 6). Furthermore, foals had a significantly higher percentage of macrophages compared with adults, characterized by a progressive increase in percentage of macrophages concurrent with an increase in age. To better define potential qualitative differences in pulmonary‐alveolar macrophages (PAMs) between adult horses and foals, a longitudinal analysis of reactive intermediate activity between adult and foal PAMs was performed with samples from a subset of foals (n = 3) at age 1 week, 1, 6 and 12 months. In vitro measurements of superoxide and peroxynitrite were obtained using oxidation of dihydroethidium (DHE) and dihydrorhodamine (DHR) as indicators, respectively. PAMs were gated and sorted as live cells expressing CD172a, seeded in triplicate into 96‐well plates, and allowed to adhere for 2 hours. PAMs were then stimulated with LPS (2 μg/mL) + IFNγ (10 ng/mL); opsonized zymosan (0.5 mg/mL) or R. equi antigen (75 μg/mL) and incubated at 37 °C for 30 hours. DHE or DHR was added 30 minutes prior to fluorimetry (BMG Labtech FluoStar Omega). For all stimulants, we documented an age‐related progressive increase in production of superoxide and peroxynitrite with foals exhibiting lower levels of reactive intermediates at week 1 compared to month 12 (P < 0.05). Taken together, these results reveal a decreasing shift in percentage of PAMs in BALF as foals age along with a concurrent age‐related increase in reactive intermediate production by PAMs. These results support the notion that the pulmonary cell population changes as foals age, with a large percentage of macrophages having decreased capacity for reactive intermediate production early in life. Collectively, these findings are consistent with age‐related susceptibility of foals to pneumonia caused by the macrophage‐tropic bacterium, R. equi. E54 CORYNEBACTERIUM PSEUDOTUBERCULOSIS SEROPREVALENCE IN HEALTHY HORSES IN A NON‐ENDEMIC STATE Marta Barba, Allison Stewart, Thomas Passler, Anne Wooldridge Auburn University, Auburn, AL, USA Corynebacterium pseudotuberculosis biovar equi infection in horses, also called pigeon fever, has not been reported in Alabama. Our previous studies had surprisingly identified positive synergistic hemolysis inhibition (SHI) titers in a number of healthy horses. The purpose of this study was to investigate the seroprevalence of C. pseudotuberculosis in this non‐endemic population; and to determine the association of detectable titers with exposure to small ruminants with or without history of caseous lymphadenitis, or with previous travel to endemic states. Serum SHI C. pseudotuberculosis titers from 297 horses from 29 Alabama counties were analyzed. A questionnaire was completed by each owner at the time of blood collection. 137/297 horses (46.1%) had titers <1:8 (negative); however, positive titers were found in 54/297 (18.2%) at 1:8, 63/297 (21.2%) at 1:16, 28/297 (9.4%) at 1:32, 9/297 (3%) at 1:64, 4/297 (1.3%) at 1:128, 1/297 (0.5%) at 1:256 and 1/297 (0.5%) at 1:512. There does not appear to be a correlation between having exposure to small ruminants and having a high antibody titer. 4/5 horses with a titer ≥1:128 had a history of travel outside Alabama. The cause of the high proportion of low‐level positive SHI C. pseudotuberculosis titers in healthy naive horses was not apparent. The high prevalence of positive titers in a naive population questions the accuracy of the SHI antibody titer test. Possible false positives caused by cross‐reaction with antibodies against phospholipases from soil Corynebacterium spp. or C. pseudotuberculosis biovar ovis warrants further investigation. E55 GENOME WIDE ASSOCIATION STUDY OF EQUINE HERPESVIRUS TYPE 1‐INDUCED MYELOENCEPHALOPATHY Margaret Brosnahan 1, Mohammed Al Abri2, Douglas Antczak1, Nikolaus Osterrieder3, Samantha Brooks4 1Cornell University, Baker Institute for Animal Health, Ithaca, NY, USA, 2Cornell University, Department of Animal Science, Ithaca, NY, USA, 3Freie Universität, Institut für Virologie, Berlin, Germany, 4University of Florida, Department of Animal Sciences, Gainesville, FL, USA Equine herpes virus type 1‐induced myeloencephalopathy (EHM) is a neurologic disease of horses that occurs as a complication of infection. Neurologic signs result from endothelial cell damage causing vasculitis and ischemic insult to the central nervous system. This disease causes considerable animal suffering and economic loss for the horse industry. Genetic variation in the virus explains some, but not all, of the disease's occurrence. A total of 129 horses (61 affected, 68 unaffected) from experimental and natural infections were included in a genome wide association study (GWAS) to identify host genetic variation associated with EHM. Genotyping was performed using the Illumina SNP50 and SNP70 arrays and a custom Sequenom array. Mixed linear model (MLM) analysis using a dominant model identified one marker that surpassed the threshold for genome‐wide significance (P = 9.50 × 10−7) after Bonferroni correction. The marker is in an intron of TSPAN9, a gene expressed in endothelium and platelets. Four of the top 20 markers (P ≤ 2.58 × 10−4) were in an immune gene cluster including an interferon gamma receptor. A separate Bayesian analysis of the same marker set identified four windows that each accounted for >1% of the genetic variance and harbored significant quantitative trait loci. This GWAS identified several regions in the horse genome that are associated with EHM in the study population, and should be further explored. Understanding how host variation contributes to EHM will enhance our knowledge of disease pathophysiology, improving treatment of individual cases and management of outbreaks. E56 WNV IGG ANTIBODY RESPONSE IN FOALS (WITH AND WITHOUT MATERNAL ANTIBODY) FOLLOWING VACCINATION WITH A COMBINATION WNV VACCINE Britt Conklin1, Glenn Blodgett2, Amanda Adams3, Steven Grubbs 1, George Milliken4 1Boehringer Ingelheim Vetmedica, Inc., St. Joseph, MO, USA, 2Burnett Ranches, LLC (6666 Ranch), Guthrie, TX, USA, 3University of Kentucky, Lexington, KY, USA, 4Kansas State University, Manhatten, KS, USA Maternal antibodies have been reported to exert a significant inhibitory effect (maternal antibody interference) on the immune response of foals to various antigens contained in commercial vaccines. Based on previous information, it has been recommended that primary immunization of foals born to vaccinated mares should not begin prior to 6 months of age following the decline of maternally‐derived antibodies. The objective of this study was to evaluate the WNV antibody response in foals (with and without WNV maternal antibody) following vaccination with Vetera ® VEWT+WNV with proprietary adjuvant Carbimmune ® . Fifty‐one, sixteen‐week old Quarter Horse foals born to mares that had received a booster dose of a multicomponent vaccine during the last 2 months of gestation were randomized into five groups based on the pre‐vaccination WNV IgG level determined by a WNV‐IgG ELISA. The WNV assay has been described previously in detail for WNV‐specific IgM detection using the anti‐equine IgM mAb. Here, an IgG1 mAb was used instead of anti‐IgM to capture IgG1 from equine serum. ELISA results are expressed as P/N ratios. A non‐vaccinated foal CONTROL group was included to confirm field exposure to WNV did not occur. The four vaccinated groups were designated Group A with pre‐vaccination response <2, Group B with pre‐vaccination response between 2 and 4, Group C with pre vaccination response between 4 and 8, and Group D with pre‐vaccination response between 8 and 16. Foals were vaccinated on T17 and T21 and blood was drawn on week 16 (pre vaccination), then on weeks 19, 21, 22 and 24. A repeated measures analysis of variance was used to analyze the data where a covariance matrix with unequal variances and unequal correlations between the draw weeks. Statistical significance was determined at P ≤ 0.05. The CONTROL group WNV IgG antibody levels were 0 at each time point throughout the study. For all vaccinate foal groups (with and without maternal antibody), the means of WNV IgG at week 16 (pre‐vaccination) were all significantly (P < 0.05) different from each other. WNV IgG levels at weeks 21, 22, and 24 for all foal groups were not significantly different from each other. Foal groups A, B, and C (pre‐vaccination ELISA WNV IgG ≤8) all had significant (P < 0.05) increases in WNV IgG antibody following the first vaccination and these antibody levels continued to increase at each time point throughout the study. The WNV IgG antibody did not increase in foals in the maternal group D (pre‐vaccination ELISA WNV IgG >8) until after the second vaccination. These findings suggest that beginning the initial series of vaccinations in foals 4 months of age is a viable option although foals with the higher level of WNV maternal antibodies may require 2 doses before a response is observed. These findings also suggest that waiting to vaccinate until WNV maternal antibodies have disappeared may not be required. E57 THERAPEUTIC MONITORING OF PLASMA AMINOGLYCOSIDE CONCENTRATIONS AND THEIR RELATION TO MINIMUM INHIBITORY CONCENTRATIONS OF LOCAL GRAM NEGATIVE BACTERIA IN HORSES AND FOALS Angelika Schoster 1, Lanja Saleh2, Morena Amsler1, Colin Schwarzwald1 1Vetsuisse Faculty, Clinic for Equine Internal Medicine, University of Zurich, Zurich, Switzerland, 2Institute for Clinical Chemistry, University of Zurich, Zurich, Switzerland Currently, 6.6 mg/kg q24 hour gentamicin and 21 mg/kg q24 hour amikacin, respectively, are commonly used to treat infections in horses and foals. Doses of 7–9 mg/kg gentamicin and 25 mg/kg amikacin have been suggested to reach adequate peak concentrations of 8–10× minimum inhibitory concentration (MIC), although these doses have not been critically evaluated in clinical studies. The objective of this study was to measure peak and trough plasma concentrations in horses treated with 9 mg/kg gentamicin q24 hour IV and in foals treated with 25 mg/kg of amikacin q24 hour IV and to compare the peak concentrations to the MICs of local gram‐negative bacteria. 46 horses treated with gentamicin and eight foals treated with amikacin were included. Peak and trough plasma concentration were measured 60 minute and 20 hour after administration. The MICs of gram‐negative bacteria isolated in the hospital in 2014 were determined and compared to plasma concentrations. Backward stepwise logistic regression was used to identify factors associated with inadequate aminoglycoside peak concentrations. 43/46 (93%) horses reached a gentamicin peak of >16 μg/mL and 46/46 (100%) horses had a trough <2 μg/mL. 7/8 foals reached an amikacin peak of 32–40 μg/mL and 8/8 had a trough concentration of <2 μg/mL. Age (P = 0.006) and treatment failure (P = 0.03) were significantly associated with inadequate gentamicin peak concentrations, whereas presence of endotoxemia (P = 0.48) or prophylactic vs therapeutic use (P = 0.55) were not. 40/61 (65%) of gram negative bacteria were resistant to gentamicin (MIC ≥ 2 μg/mL). 47/61 (77%) of bacteria had a MIC of ≤4 μg/mL for amikacin. A dose of 9 mg/kg gentamicin q24h IV and 25 mg/kg amikacin q24h IV was not sufficient in all cases to reach 10× MIC for the majority of susceptible gram negative bacteria in our hospital. Using once daily dosing, the trough concentration fell below the renal protective threshold in all animals 4 hour before the next dose was administered. E58 EFFECT OF BLANKETING ON VITAMIN D STATUS IN HORSES AT PASTURE Sara Azarpeykan 1, Erica K. Gee1, Jonathan C. Marshall1, Peter Elder2, Keith G. Thompson1, Els Acke1, Keren E. Dittmer1 1Institute of Veterinary, Animal & Biomedical Sciences, Massey University, Palmerston North, New Zealand, 2Steroid & Immunobiochemistry Unit, Canterbury Health Laboratories, Christchurch, New Zealand The ability of skin to synthesise vitamin D3 can be influenced by factors including melanin pigmentation, season, latitude/altitude, type of clothing, and sunlight exposure. Equine vitamin D metabolism and factors influencing vitamin D synthesis remain poorly understood. The majority of horses in New Zealand spend a large proportion of time outside in paddocks often with blankets on, which may alter their ability to synthesise vitamin D3. This study aimed to determine if blanketed horses have lower serum vitamin D concentrations than un‐blanketed horses. Twenty one mature horses at pasture were included; five were covered with standard horse blankets including neck rugs. They were fed ad libitum grass pasture and supplementary hay when required. Blood and pasture samples along with climate data were collected monthly for 13 months and analysed for 25‐hydroxyvitamin D2, D3 (25OHD2, D3) and 1,25‐dihydroxyvitamin D (1,25(OH)2D). Statistical analysis consisted of generalized additive mixed models. The main form of 25OHD detected in serum was 25OHD2. The concentration of 25OHD3 was zero. There were no statistically significant differences in serum 25OHD2 and 1,25(OH)2D concentrations between the groups (P < 0.05). Vitamin D concentration in pasture was high (˃ 6.6 IU/BW), exceeding the minimum daily recommended intake requirements. Serum 25OHD2 and pasture vitamin D concentration were directly correlated with average monthly sunshine. These results suggest that blanketing does not affect the ability of mature horses at pasture to synthesise vitamin D3 in skin and diet is an important factor influencing serum vitamin D concentrations in horses. E59 EQUINE REFERRING VETERINARIANS' EXPECTATIONS OF EQUINE VETERINARY SPECIALISTS AND EQUINE REFERRAL CARE CENTERS Colleen Best, Jason Coe, Joanne Hewson, Michael Meehan, David Kelton Ontario Veterinary College, University of Guelph, Guelph, ON, Canada When an equine patient is referred, the veterinary healthcare team extends from the primary care veterinarian and their clinic to include the specialists and staff at the referral center. This increases the complexity of the communication among individuals and of integrating the work of both teams. The objective of this study was to explore referring veterinarians' expectations of equine veterinary specialists and equine referral centers. Six focus groups with equine referring veterinarians (n = 48) were conducted and the conversations were analyzed using thematic analysis, a recognized form of qualitative research. The overarching theme of participants' discussions was that the specialist and referral center should act as an extension of the care the referring veterinarian provides to their clients and patients. The relationship that participants had with their clients was a significant driver of their expectations of the specialist and referral care. Participants spoke about wanting to remain involved in the horse's care when the horse was under the specialist's care. Participants spoke about desiring mutual respect, collegiality, professionalism, and trust in the relationships they had with specialists. Communication played a central role in many of the positive experiences participants shared, and it was perceived to be an essential component of the referral process. Displeasure and frustration were voiced about referral centers that competed with the referring veterinarian's practices or solicited their clients. Strengthening the relationship of referring veterinarians and specialists has the potential to improve the referral process, as well as patient and client outcomes. E60 ENDOTHELIAL PROGENITOR CELL ISOLATION IN HORSES Anne Wooldridge 1, Ashley Sharpe1, Wen Jun Seeto2, Elizabeth Lipke2 1Auburn University College of Veterinary Medicine, Auburn, AL, USA, 2Samuel Ginn College of Engineering, Auburn, AL, USA Endothelial progenitor cells (EPCs) function in vascular formation and repair, and number and function of EPCs are emerging biomarkers of endothelial dysfunction and vascular health. Function of EPCs (such as vascular tubule formation) have been shown to be altered in people with metabolic syndrome, indicating that EPCs may be integral to endothelial dysfunction and microvascular disease in those patients. Our group has recently isolated and characterized EPCs using a whole blood adherence method from 5 mL jugular venous samples in horses, and 3/24 horses (12.5%) produced colonies. The purpose of this study was to optimize the isolation of EPCs from horses so that future diagnostic and therapeutic uses could be explored. Thirty mL heparinized blood samples were collected both the cephalic and jugular veins of 8 healthy adult horses. EPCs were isolated using both density gradient centrifugation (DGC) and a whole blood (WB) adherence method. Isolated colonies were characterized using vascular tubule formation, uptake of acetylated low density lipoprotein (LDL), and indirect immunofluorescence (IF). EPCs were isolated from 7/8 horses (88%). Collection from the cephalic vein yielded EPCs in 6/8 horses versus 2/8 horses from the jugular vein. There was no difference between DGC and WB methods. Isolated EPCs formed vascular tubules, showed uptake of LDL, and expressed endothelial markers through IF. The vessel diameter may influence the sample population of circulating cells. Collection of blood from both sites is non‐invasive and dramatically increases the yield of EPCs. Consistent yield of EPCs from peripheral blood will allow future studies evaluating diagnostic and therapeutic uses in horses. E61 A COMPARISON OF THE EFFECTS ON SWEATING OF THREE MACROLIDE ANTIBIOTICS USED IN FOALS Amy Stieler, L. Chris Sanchez, Martha Mallicote, Susan Westerterp, Jim Burrow, Robert J. MacKay University of Florida College of Veterinary Medicine, Gainesville, FL, USA Previous work by the authors showed that erythromycin‐associated hyperthermia is caused by impaired sweating. Whether or not the macrolides clarithromycin and azithromycin also induce anhidrosis is unknown. This study evaluated the sweat response in foals treated with azithromycin, clarithromycin and erythromycin. In serial experiments, 12 foals (four females, eight males), aged 1 month, were treated orally for 5 days with erythromycin (25 mg/kg, 3 times daily), azithromycin (10 mg/kg, once daily) and clarithromycin (7.5 mg/kg, twice daily) according to a masked, duplicated, 6 × 3 counterbalanced design. Quantitative intradermal terbutaline sweat tests were performed on 3 successive days before treatment (baseline) and on days 1, 2, 5, 9, 24, and 39 after treatment began. There was significant (P < 0. 05) suppression of sweating when foals were given any of the 3 macrolides. Compared with terbutaline‐induced sweating at baseline, values were significantly lower on all evaluation days when foals were given erythromycin, on days 2 and 5 after beginning treatment when given clarithromycin, and on days 1, 2, and 5 when given azithromycin. Overall suppression of sweating by erythromycin at the dose used in the study was significantly greater than that induced by either clarithromycin or azithromycin; however, there was no difference between the effects of clarithromycin and azithromycin. Results show that all of the macrolides commonly used for treating or preventing Rhodococcus equi pneumonia in foals suppressed sweating responses to varying degrees. Foals treated with any of these antibiotics should be considered at risk for hyperthermia. E62 ENDOBRONCHIAL BIOPSY SCORES RELIABLY ASSESS DISEASE SEVERITY IN HORSES WITH HEAVES Michela Bullone 1, Pierre Hélie2, Jean‐Pierre Lavoie1 1Department of Clinical Sciences, Faculty of Veterinary Medicine, Université de Montréal, St‐Hyacinthe, QC, Canada, 2Department of Pathology and Microbiology, Faculty of Veterinary Medicine, Université de Montréal, St‐Hyacinthe, QC, Canada Lower airway inflammatory diseases (Inflammatory Airway Disease and Heaves) are common conditions of variable severity in horses. While alterations in bronchoalveolar lavage fluid (BALF) cytology is a characteristic finding in affected horses, the degree of lower airway inflammation is not correlated with the severity of lung function impairment. Furthermore, currently, little is known of the inflammatory and remodeling processes affecting the central airways in these horses. We developed a semi‐quantitative scoring system for the assessment of endobronchial biopsy morphology and hypothesized that it would allow differentiating horses with heaves from controls, independently from their clinical status (exacerbation or remission). Six healthy horses and 12 horses with heaves (6 in remission, HR, and 6 in exacerbation, HE) were studied. Horses underwent respiratory mechanic measurements using an impulse oscillometry system (IOS). Six endobronchial biopsies were obtained from each horse and processed for histology. Selected morphological and inflammatory parameters were blindly assessed by three operators. The scores of HE were significantly higher than those of HR and controls (P < 0.001). Parameters differentiating HE from HR were primarily those related to inflammation and to the presence/extension of submucosal mucous glands. The scores of HE and HR were significantly correlated with expiratory reactance values at 0.5 Hz (P = 0.05 and P = 0.03, respectively). A semi‐quantitative scoring system which reliably assesses bronchial tissue remodeling and inflammation in horses with heaves was developed. Further optimization is required for differentiating HR, and possibly horses with Inflammatory Airway Disease, from controls. E63 EVALUATION OF ANGIOGENESIS IN THE AIRWAYS OF HORSES WITH HEAVES Nicolas Herteman, Michela Bullone, Jean‐Pierre Lavoie Department of Clinical Sciences, Faculty of Veterinary Medicine, Université de Montréal, Saint‐Hyacinthe, Canada Heaves is a severe and debilitating inflammatory airway disease of horses associated with permanent structural changes of the bronchial wall. Whether the chronic inflammation present in heaves promotes the formation of new vessels, a phenomenon known as angiogenesis, is unknown. Narrow band imaging (NBI) endoscopy is a non‐invasive technique that enhances the visualization of submucosal vessels and the detection of angiogenesis. It employs filters producing light of specific blue and green wavelengths corresponding to the peak light absorption of hemoglobin. Using NBI, we hypothesized that the central airways of horses with heaves sustain angiogenesis. Horses with heaves in exacerbation of the disease (n = 5) and control animals (n = 6) were studied. A library of NBI images was established from recorded endoscopies. Four sites were studied: larynx (40 images), trachea (117 images), carina (53 images), and bronchi (153 images). Using a dedicated stereology software (NewCAST, Visiopharm), the volume density of small and large vessels was calculated at each site for all horses. In the trachea, the volume density of small vessels was increased in horses with heaves compared to controls (P = 0.01). No difference was found for large vessels (P > 0.1). The volume density of both small and large vessels in the larynx, carina or bronchi was similar in the two groups (P > 0.1 for all). NBI imaging of the airways was easily performed in standing sedated horses. The significance of the increased tracheal vascularity in heaves remains to be ascertained. F01 LUNG CONSOLIDATION IN VEAL CALVES METAPHYLACTICALLY TREATED WITH TILDIPIROSIN: PREDICTIVE FACTORS AND IMPACT ON WEIGHT GAIN Julie Berman, David Francoz, Jocelyn Dubuc, Sébastien Buczinski Faculté de Médecine Vétérinaire, Université de Montréal, Saint‐Hyacinthe, QC, Canada Bovine respiratory disease (BRD) complex is a major problem in veal calves rearing units. Presence of lung lesions and inflammation was shown to negatively affect subsequent growth. Ultrasonography is an ante‐mortem diagnostic tool that can be used to assess the extent of lung lesions. The first objective of this study was to describe the prevalence of ultrasonographic evidence of lung consolidation in veal calves metaphylactically treated with tildipirosin. The second objective was to quantify the impact of lung consolidation on average daily weight gain (ADG) until weaning. A total of 209 veal calves from a pre‐weaning fattening unit were enrolled. These calves were followed for a 70‐day period in three visits: at day 1 (D1), at day 12 (D12) and at day 30 (D30) after arrival. ADG was calculated at 1 month (ADG1) and 2 month (ADG2) of the pre‐weaning period. The 209 veal calves were randomly allocated to one of two treatment groups (Tildipirosin, TILD, n = 109; Placebo, PLAC, n = 100). The label dose of Tildipirosin (ZUPREVO) or the same amount of saline was blindly administered SQ once at D12. At D1, calves were bled (jugular vein punction) to quantify total solids in order to identify cases of failure of passive transfer (FPT). A systematic thoracic ultrasound examination (using 7.5 MHz linear probe) focusing on the maximal depth of consolidation (DEPTH) was performed at each visit. Lung consolidation was defined as DEPTH ≥3 cm. Prevalence of FPT was 55% (114/209) in calves. At D12, 13% (26/209) of calves had lung consolidation. The presence of consolidation at D30 was observed in 27% (29/109) of calves in the TILD group and 27% (27/100) in PLAC group. Calves with lung consolidation at D12 had greater odds of lung consolidation at D30 (OR = 4.35; 95% CI: 1.79–10.6; P < 0.01) than calves without consolidation at D12. Calves with FPT had greater odds of consolidation (OR = 2.35; 95% CI: 1.19–4.65; P = 0.02) than calves without FPT. Treatment (TILD vs PLAC) was not associated with the odds of lung consolidation at day D30 (P = 0.91). Least square means of ADG2 in calves with lung consolidation was greater than in calves without consolidation (consolidation: 2.12 lbs/d, SEM = 0.06; no consolidation: 1.97 lbs/d, SEM = 0.08; P = 0.03). In conclusions, this study showed that ultrasonographic evidence of lung consolidation in veal calves 1 month after arrival can be predicted by the assessment of the FPT and by the presence of lung consolidation 12 days after arrival. Metaphylatic use of Tildipirosin at D12 doesn't decrease the prevalence of lung consolidation at D30. Surprisingly, the presence of lung consolidation was associated with an increased ADG during the second month of feeding which can appear as counter‐intuitive. Further research should be performed to understand this finding. F02 SYSTEMATIC REVIEW OF THE DIAGNOSTIC ACCURACY OF HAPTOGLOBIN, SERUM AMYLOID A AND FIBRINOGEN VERSUS CLINICAL REFERENCE STANDARD FOR THE DIAGNOSIS OF BOVINE RESPIRATORY DISEASE (BRD) AbdElMonem AbdAllah 1, Joanne Hewson2, David Francoz1, Hatem Selim3, Sébastien Buczinski1 1Faculté de médecine vétérinaire, Université de Montréal, St‐Hyacinthe, QC, Canada, 2Ontario Veterinary College, University of Guelph, Guelph, ON, Canada, 3Faculty of Veterinary Medicine, Zagazig University, Zagazig, Egypt The objective of this study was to assess, using a systematic review of the literature, the accuracy of acute phase proteins (AAP): Haptoglobin (Hp), Serum amyloid A (SAA), and Fibrinogen (Fib) as diagnostic tools for cattle with naturally occurring bovine respiratory disease (BRD) when compared with clinical reference standards. A systematic review was performed with eligible studies selected from CAB and MEDLINE from 1946 to 2014. Results were subsequently screened for inclusion and exclusion criteria. The methodological quality of these studies was assessed (by 2 or 3 independent reviewers) using the Quality Assessment of Diagnostic Accuracy Studies‐2 (QUADAS‐2) to evaluate any risk of bias and applicability concerns for each study. The definition of the reference standards (which always included clinical parameters) was noted. Measures of accuracy (sensitivity (Se), specificity (Sp)) were obtained from the articles or through contact with the authors. A total of 312 studies were identified after removal of duplicates. Based on the title and abstract, 188 studies were non‐relevant and 66 studies were experimental BRD, leaving 58 studies selected for full review; from them 23 met our inclusion criteria as diagnostic studies for naturally occurring BRD using one or more of the mentioned APP. Each study used a different BRD/non‐BRD definition. Results of QUADAS‐2 assessment of the selected studies are summarized in the following table. 1‐ Risk of bias: High Low Unclear A‐ Patient selection 18 3 2 B‐ Index test 5 3 15 C‐Reference standard 5 4 14 D‐Flow and timing 4 2 17 2‐Applicabiliy: A‐ Patient selection 15 2 6 B‐ Index test 1 19 3 C‐Reference standard 13 3 7 From the 18 authors contacted to obtain data concerning accuracy between the APP and BRD status (i.e. those papers where an email contact for the corresponding author was provided), data from only 5 studies were obtained. Additionally, 1 study contained accuracy information (a 2‐by‐2 table) for Hp and SAA. Consequently, up to 6 studies (depending on the APP) were included in the meta‐analysis. The Se/Sp for each of the selected APP varied from study to study. For Hp, the Se varied from 0.70 to 0.94 and Sp from 0.75 to 1.0 (n = 6). For SAA, Se varied from 0.59 to 1.0 and Sp from 0.43 to 0.87 (n = 4), and for Fib, Se varied between 0.43 and 0.89 and Sp varied between 0.68 and 0.85 (n = 4). Unfortunately due to the low number of studies, we could not assess publication bias or the impact of study design or methodological flaws on test accuracy. In conclusion, the standards for reporting the accuracy or clinical utility of APP in diagnosing naturally‐occurring BRD are inconsistent. The absence of a consistent case/non‐case definition was frequently observed during this systematic review. The most common types of biases noted during the review process – spectrum (patient selection), verification (inclusion of non‐consecutive cases) or differential reference (different reference standard for BRD and non BRD cases) – could potentially overestimate the apparent accuracy in the studies. Even if several authors have mentioned that the APP could be helpful in BRD diagnosis, it is not possible to conclude this in terms of accuracy from the present study (especially the incremental value of the tests based on the pre‐test probability using Bayes' rule). Reporting studies on diagnostic tests following the Standard for the Reporting of Diagnostic Accuracy studies (STARD) is encouraged to improve capability for future meta‐analyses. F03 AGREEMENT BETWEEN SAMPLE COLLECTION METHODS USED FOR THE DIAGNOSIS OF BOVINE RESPIRATORY DISEASE IN DAIRY CALVES Danielle Doyle 1, Brent Credille1, Roy Berghaus1, Terry Lehenbauer2, Amelia Woolums1 1University of Georgia, Athens, GA, USA, 2University of California at Davis Veterinary Medical Teaching and Research Center, Tulare, CA, USA Four antemortem sample collection methods are commonly used to determine the causative agents of bovine respiratory disease (BRD)in dairy calves: nasal swabs (NS), deep nasopharyngeal swabs (DNP), bronchoalveolar lavage (BAL), and transtracheal wash (TTW). The objective of this study was to evaluate the agreement between these methods with respect to the isolation of pathogens commonly associated with BRD in dairy calves. Pre‐weaned dairy calves identified as having acute BRD were sampled with all four methods. Each sample was submitted for aerobic culture and viral multiplex PCR. The pathogens isolated by each method were compared to the TTW, our antemortem gold standard. Agreement between methods and significance of the results were determined by calculation of the kappa statistic and McNemar's chi‐square test, respectively. For the NS, DNP, and BAL, very good agreement was seen with respect to the isolation of P. multocida and M. haemolytica relative to the TTW. For bovine respiratory syncytial virus (BRSV), agreement of the NS relative to the TTW was moderate, good for the DNP, and very good for the BAL. For bovine coronavirus (BCV), agreement of the NS and DNP relative to the TTW was moderate, and good for the BAL. The McNemar's test was only significant for NS and DNP compared to TTW when isolating BCV, indicating that the proportions of positive results were not equal for those methods. Based on this study, BAL has the best agreement with TTW and gives the most diagnostic utility for sampling dairy calves with BRD. F04 FRONTAL SINUSITIS IN ADULT BEEF BULLS: A RETROSPECTIVE ANALYSIS OF 18 CASES (1999–2014) Katharine Simpson 1, Robert Streeter2, Jared Taylor2, Suzanne Genova3 1The Ohio State University, Columbus, OH, USA, 2Oklahoma State University, Stillwater, OK, USA, 3Mississippi State University, Starkville, MS, USA Acute and chronic frontal sinusitis in cattle may occur in association with dehorning, traumatic horn injuries, or with respiratory tract disease. The disease is more common in young animals than in adults. A limited number of reports describe sinusitis in beef cattle. The objective of this retrospective study was to provide case‐based evidence describing the signalment, history, clinical findings, diagnostic testing, treatment, and outcome of frontal sinusitis in adult beef bulls. A systematic search was performed and medical records were reviewed for 18 adult (≥ 2 years) beef animals diagnosed with frontal sinusitis over a 15‐year period. Information on previously listed parameters was evaluated. All animals diagnosed with frontal sinusitis were bulls. Breeds represented were crossbred (15), Brahman (1), Brangus (1), and Dexter (1). Seventeen (94%) were rodeo bulls. Median age at presentation was 4.5 years (mean 4.7 years). Duration of clinical signs ranged from 3 days to 1 year (mean 61 days, median 22.5 days). History included horn tipping in 56% (10/18). Clinical findings included nasal discharge (44%), dullness on percussion of the affected sinus (44%), abnormal head position (39%), palpable warmth of the horn on the affected side (33%), drainage from the affected horn (28%), and distortion of the frontal bone (11%; 2/18). Temperature was within normal range in 59% of bulls (10/17; range, 99.3–103.5ᵒF). Hyperproteinemia (6/7), hyperglobulinemia (6/6), and hyperfibrinogenemia (4/6) were the most consistent abnormal clinicopathologic findings at presentation. Diagnostic imaging included skull radiographs (13) and CT (1). Bacterial culture of infected tissue or exudate was performed in 3, and revealed Trueperella pyogenes (2), Pseudomonas aeruginosa (2), and Pseudomonas fluorescens (1). Unilateral frontal sinusitis was diagnosed in all animals, and the left frontal sinus was affected in 61% of bulls (11/18). Treatment included trephination (8/18), horn removal (13/18), sinus lavage (15/18), antimicrobials (17/18), NSAIDs (13/18), sodium iodide (6), locally applied antimicrobial‐impregnated plaster of Paris beads (2), vacuum‐assisted closure (1), and medical‐grade maggots (1). Sinusotomy sites were protected via packing with cotton or gauze (11) and maintained under a sterile bandage (8). Seventeen (94%) bulls were discharged from the hospital following diagnosis and treatment. Long‐term follow‐up (≥1 year) was available for 13 bulls (72%), all of which made a full recovery. Twelve of these were rodeo bulls; of these, 8 were reported to make a full recovery and perform well, 3 (25%) made a full recovery but did not perform well again, and 1 developed an abscess of the right papillary muscle and died 2 years later. Frontal sinusitis should be considered as a differential diagnosis in beef cattle presenting with non‐localizing clinical signs. Ancillary diagnostic findings may also be nonspecific. Bucking bulls performing in rodeos are required to have the tips of their horns blunted to a diameter of 3.8 cm (1.5 inches). This procedure may result in opening of or bacterial translocation into the frontal sinus contributing to the development of frontal sinusitis in these animals, particularly in young adults. With appropriate treatment, prognosis for long‐term survival is good even in chronic cases. F05 COMPARATIVE PLASMA PHARMACOKINETICS OF CEFTIOFUR SODIUM AND CEFTIOFU CRYSTALLINE FREE ACID IN NEONATAL CALVES Jane Woodrow 1, Brent Credille2, Melissa Hines1, James Caldwell1 1University of Tennessee, Knoxville, TN, USA, 2University of Georgia, Athens, GA, USA The objective of this study was to compare the plasma pharmacokinetic profile of ceftiofur crystalline free acid (CCFA) and ceftiofur sodium in neonatal calves. In one group, a single dose of CCFA was administered subcutaneously (SQ) at the base of the ear at a dose of 6.6 mg/kg of body weight. In a second group, a single dose of ceftiofur sodium was administered SQ in the neck at a dose of 2.2 mg/kg of body weight. Concentrations of desfuroylceftiofur acetamide were determined in the plasma of all animals. After administration of the products time to maximum DCA concentration was 12 hour for CCFA and 1 hour for ceftiofur sodium. Maximum plasma DCA concentration was significantly higher for calves given ceftiofur sodium (5.62 μg/mL) than for calves given CCFA (3.23 μg/mL). AUC0‐∞ and Vd/F were significantly greater for calves given CCFA than for calves given ceftiofur sodium. The terminal half‐life of DCA in plasma was significantly longer for calves given CCFA (60.6 hour) than calves for calves given ceftiofur sodium (18.1 hour). Cl/F was not significantly different between groups. The duration of time median plasma DCA concentrations remained above above 2.0 μg/mL was significantly longer in calves that received CCFA (84.6 hour) as compared to calves that received ceftiofur sodium (21.7 hour). Based on the results of this study, CCFA administered SQ at a dose of 6.6 mg/kg in neonatal calves provided plasma concentrations above the therapeutic target of 2 μg/mL for at least 3 days following a single dose. F06 CHANGES IN PLASMA CALCIUM CONCENTRATION IN PERIPARTURIENT DAIRY CATTLE FED AN ACIDOGENIC DIET IN THE LATE DRY PERIOD Ameer Megahed 1, Mark Hiew1, Jonathan Townsend1, Peter Constable2 1Purdue University, West Lafayette, IN, USA, 2University of Illinois at Urbana‐Champaign, Urbana‐Champaign, IL, USA Calcium homeostatic mechanisms are challenged in periparturient multiparous dairy cattle due to the rapid transport of large amounts of calcium into the mammary gland, resulting in decreased plasma calcium concentration ([Ca]). An unresolved issue is the timing of the decrease in plasma [Ca] relative to the time of parturition, with the consensus view being that plasma [Ca] does not decrease until after parturition. The objective of this study was therefore to characterize the change in plasma [Ca] over time in periparturient dairy cattle. Plasma and mid‐stream urine samples (653) were collected daily starting 3 days before calving from 106 periparturient Holstein Friesian dairy cows fed an acidogenic total mixed ration during the late dry period. Mixed models analysis of variance was conducted and P < 0.05 was considered significant. Plasma [Ca] (corrected for plasma total protein concentration) decreased in multiparous cattle (n = 72) but not primiparous cattle (n = 34). Compared to least squares mean values 72 hour before parturition ([Ca] = 10.4 mg/dL), plasma [Ca] in multiparous cattle was first decreased at 4 hour before parturition (9.1 mg/dL), was lower (8.7 mg/dL) at 4 hour after parturition, increased to 9.3 mg/dL at 16 hour but decreased again to 8.6 mg/dL at 28 hour after parturition, after which time plasma [Ca] gradually increased. The urine [Ca] to [creatinine] ratio in multiparous cattle was first decreased at 4 hour before parturition and remained low from this time onward. We conclude that calcium supplementation should be administered at least 4 hour before the anticipated time of calving in multiparous dairy cattle in order to decrease the incidence of hypocalcemia. F07 EVALUATION OF THE PRECISION XTRA® HAND‐HELD METER FOR THE RAPID POINT OF CARE MEASUREMENT OF BLOOD AND PLASMA GLUCOSE CONCENTRATION IN PERIPARTURIENT DAIRY CATTLE Ameer Megahed 1, Mark Hiew1, Jonathan Townsend1, Peter Constable2 1Purdue University, West Lafayette, IN, USA, 2University of Illinois at Urbana‐Champaign, Urbana‐Champaign, IN, USA Accurate diagnosis of hypoglycemia is helpful in the diagnosis and treatment of ketosis in periparturient dairy cattle. The Precision Xtra® meter uses methodology that assumes intra‐erythrocyte glucose concentration ([glucose]e) and plasma glucose concentration ([glucose]p) are equivalent and that the hematocrit (Hct) is the typical value for human blood (0.43). The objectives of this study were therefore to determine the accuracy of the Precision Xtra® meter for measuring [glucose]p in blood or plasma from dairy cattle, and to determine the influence of Hct on the measured [glucose]b. Blood and plasma samples (1131) were collected from 106 healthy periparturient Holstein‐Friesian cattle. Method comparison was performed using Deming regression and Bland‐Altman plots. The [glucose]e was calculated as: [glucose]e = {[glucose]b‐[glucose]p×(1‐Hct)}/Hct. The reference method [glucose]p‐hexokinase on the day of calving (n = 102) ranged from 33 to 284 mg/dL (median, 76 mg/dL). Hematocrit ranged from 0.23 to 0.44 (median, 0.33) and plasma protein concentration ranged from 40 to 90 g/L (median, 63 g/L). The median [glucose]e to [glucose]p ratio was 0.18, demonstrating that [glucose]e and [glucose]p are not equivalent. We conclude that the algorithm in the Precision Xtra® meter that calculates [glucose]p from [glucose]b is optimized for use in human blood and cannot be accurately applied to bovine blood. We therefore recommend using plasma as the analyte instead of blood, and calculation of the [glucose]p in mg/dL using the following equation: [glucose]p = 0.65 × [glucose]p‐PrecisionXtra+14. If blood is measured using the Precision Xtra®, then we recommend using the following equation: [glucose]p = 0.61/{(Hct×0.18)+(1–Hct)×0.94)}×[glucose]b‐PrecisionXtra+10. If Hct values are not available then we recommend the following equation: [glucose]p = 0.89 × [glucose]b‐PrecisionXtra+10. F08 HEALTH BENEFITS OF IL‐10 EGG YOLK ANTIBODIES ADMINISTERED TO MILK‐FED DAIRY CALVES Sarah Raabis, Sheila McGuirk, Jane Rieman, Jordan Sand, Mark Cook University of Wisconsin, Madison, WI, USA Interleukin‐10 (IL‐10) is expressed by intraepithelial lymphocytes of calves following infection with Cryptosporidium parvum but its role in the pathogenesis of diarrhea is unknown. Studies have shown that IL‐10 knockout mice become resistant to C. parvum infection. Oral administration of chicken egg yolk antibodies as a means of controlling enteric disease in calves is of significant interest to the dairy industry. Egg yolk IL‐10 antibodies have been shown to survive gastrointestinal transit, are accessible in the intestinal lumen and remain locally active, without interference of natural immunity or persistence of systemic residues. A double‐blinded, randomized, prospective clinical trial was conducted to investigate the effects of feeding IL‐10 antibodies on calf health, C. parvum oocyst shedding, days of diarrhea, fecal pH, antibiotic use and average daily gain in dairy calves. The study included 134 calves randomly assigned into a treatment group (fed 0.96 g egg yolk powder with IL‐10 antibody per day for 11 days) or a control group (fed 0.96 g of egg yolk powder without IL‐10 antibody per day for 11 days). Each calf received a health score based on attitude, appetite, temperature, ocular and nasal discharge, cough, ear position, navel and joint assessment daily for 14 days. Hydration status, fecal pH, and fecal pathogen (C. parvum, Coronavirus, Rotavirus, and Salmonella) testing were conducted two or three times in the first 14 days and growth, respiratory health and antibiotic use were assessed at the conclusion of the trial on day 56. Results showed 8% of calves in the treatment group had respiratory disease compared to 21% of calves in the control group at day 56 (P < 0.04). In addition, 11.1% of calves in the treatment group were administered antibiotics by farm personnel compared to 25.8% of control calves (P < 0.03). Fecal pH was significantly increased in the treatment group on day 14 compared to enrollment day (P < 0.0001). The average of the days of diarrhea was not significantly different between the two groups except during the first 4 days. The treatment group had 0.49 days of diarrhea compared to the control group that had 0.36 days of diarrhea (P < 0.02). There was no difference between groups in fecal shedding of C. parvum at day 5 or day 14 (as measured by RT‐PCR CT values). Although, no significant difference in average daily gain was noted between groups, the hipometer estimated weight at day 56 was significantly greater for calves in the treatment group compared to the control group (P < 0.02). We suggest that the increased fecal pH in the treatment group may be an indication of improved gastrointestinal health. Using the common mucosal immune hypothesis, we surmise that enhanced mucosal immunity and host defenses in the gastrointestinal tract resulted in less respiratory disease in treated calves. Future directions include dose response and duration of treatment studies with more detailed focus on parameters of gastrointestinal and respiratory tract health. F09 PRESUMPTIVE FENBENDAZOLE TOXICOSIS IN THREE HOLSTEIN CALVES Emily Barrell, Rebecca Ruby, Belinda Thompson, Theresa Southard, Daryl Nydam, Gillian Perkins Cornell University, Ithaca, NY, USA Fenbendazole (FBZ) is a widely used anthelmintic that interferes with microtubule formation and is considered safe in ruminants. Studies have evaluated a 3× dose (15 mg/kg, PO, q24h for 3 days) for treatment of Giardia‐induced diarrhea without complications. Other benzimidazoles have been toxic in a variety of species, resulting in bone marrow hypoplasia and intestinal crypt epithelial necrosis. This abstract describes the clinical findings and necropsy examinations of three 2‐month‐old Holstein heifer calves with presumptive FBZ toxicosis. The calves had received two courses of greater than 3× the labeled dose of FBZ for 3 days, no FBZ for 3 days, then another 3 days of treatment for presumptive Giardia‐induced diarrhea. Clinical signs included fever, tachycardia, dyspnea, profuse diarrhea, and recumbency. All three animals had a persistent and absolute neutropenia. Despite treatment, one calf was euthanized due to a poor response and one calf died. Histologic examination of liver, lung, and intestinal epithelial cells revealed hyperplasia, cell atypia, irregular mitotic figures, and/or ring mitosis and abnormally clumped chromatin. The bone marrow was hypoplastic with an overall cellularity <5%. The feces were negative for Salmonella spp., bovine viral diarrhea virus, rotavirus, and parasites, and positive for bovine coronavirus. The third calf was treated on‐farm and recovered over several weeks. Based on histopathology and history of multiple animals affected, a toxic etiology was suspected, with the only common exposure being repeated administration of high‐dose FBZ. This suggests that judicious use of FBZ in calves is warranted. F10 THE RELATIONSHIP BETWEEN BODY SITE COLONIZATION AND INTRAMAMMARY INFECTION WITH NON‐AUREUS STAPHYLOCOCCAL SPECIES IN DAIRY HEIFERS Pamela Adkins, John Middleton University of Missouri, Columbia, MO, USA The purpose of this study was to use molecular techniques to determine if dairy heifers are colonized with the same coagulase negative staphylococci (CNS) species before calving that are causing CNS intramammary infections (IMI) at parturition. Holstein and Guernsey heifers at the University of Missouri Foremost Dairy were longitudinally enrolled in the study. Samples including pre‐partum mammary quarter secretions and swabbing samples of the teats, muzzle, perineum, and inguinal region were collected at approximately 14 day from expected calving date. Swabbing samples were collected using gas‐sterilized electrostatic dusters (Swiffers®). At calving, mammary quarter foremilk samples were collected twice, at approximately 3–5 and 7–9 days post‐partum, for bacterial culture and somatic cell counting. All body site swabbing samples were chilled on ice for transport and processed the same day. Swabbing samples were mixed with 10 mL of sterile saline, agitated, and plated on Mannitol Salt Agar (MSA). At 24 hours, plates were read and up to 10 morphologically distinct staphylococcal colonies were sub‐cultured onto Columbia Blood Agar (CBA). All pre‐partum secretions and milk samples were cultured according to the National Mastitis Council guidelines and these samples were considered to be culture positive if one or more colonies were observed (≥100 cfu/mL). All staphylococcal isolates were banked in phosphate buffered glycerol at −20°C. All staphylococci isolates were speciated using polymerase chain reaction and DNA sequencing of the rpoB housekeeping gene. To date, 55 heifers have completed the study. Sample processing and bacterial speciation has been completed for 21 heifers (84 quarters). Overall, S. chromogenes was the most common species identified in pre‐partum secretions (17.8%, 15/84) and post‐partum milk samples (16.7%, 14/84). Other non‐aureus staphylococcal species identified in mammary secretion samples included S. agnetis (2/84 pre‐partum quarters), S. devriesei (1/84 pre‐partum quarters), S. pasteuri (1/84 post‐partum quarters), and S. xylosus (1/84 post‐partum quarters). S. chromogenes was the only species found present in both pre‐partum mammary secretion and post‐partum milk samples of the same heifer. A total of 9.5% (8/84) of quarters had a S. chromogenes isolate pre‐partum and at the first post‐partum sampling, and 50% (4/8) of those quarters were still positive for S. chromogenes by the second post‐partum sampling. Eighty‐four body site samples from these 21 heifers yielded 346 staphylococcal isolates, 298 of these isolates have been speciated. A total of 10 different species have been identified, the most common species included S. devriesei (34%, 100/298), S. chromogenes (32%, 96/298), and S. haemolyticus (23%, 68/298). All three of these CNS species were isolated from all four body sites. S. chromogenes, the species found to be most prevalent in mammary secretions and milk, was isolated primarily from the inguinal region (75%, 72/96), followed by the perineum (13%, 12/96), teats (10%, 10/96), and then muzzle (1.0%, 1/96). CNS is a common cause of subclinical mastitis in dairy heifers, and in this study, S. chromogenes was the most common cause of CNS IMI. This species was also commonly found in the inguinal area and perineum of pre‐partum heifers. These body sites could be important sources of S. chromogenes for heifer IMI. Strain typing will be performed using pulsed‐field gel electrophoresis to determine whether the same strain of S. chromogenes is present at these different body sites and in the mammary gland of these heifers. Final results of this project will further the understanding of the epidemiology of CNS infections in dairy heifers and allow for the development and study of methods to control and prevent these infections. F11 VISUAL ESTIMATION OF GOAT LIVE‐WEIGHT BY VETERINARY AND ANIMAL SCIENCE STUDENTS Philippa Gibbons, Meredyth Jones, Virginia Fajt, Brandon Dominguez, David Forrest Texas A&M University, College Station, TX, USA Accurate bodyweight is necessary to correctly dose drugs in goats. Under‐dosing anthelmintics is correlated with development of anthelmintic resistance, and toxicity is a risk of over‐dosing. Many goat producers lack a scale; therefore, visual estimation of body weight is routinely performed by producers and veterinarians. Final year Doctor of Veterinary Medicine (DVM) students and Animal Science undergraduate students participated in the study. Students completed a questionnaire about their prior livestock experience. Seventy female goats of Spanish/Boer breeding were digitally recorded as they moved through an alley. The footage was consolidated into a 3‐second segment with each animal's identification followed by 15 seconds of the animal standing in the alley. Students viewed the videos and estimated weights to the nearest pound. Seventy‐one students participated, 30 Animal Science and 41 DVM students. Goats ranged from 78–208 lb (mean 121.9 lbs). Overall, students underestimated weights 65.2% of the time. Weight estimations were within 10% of the actual weight only 36.6% of the time. Prior livestock experience and student classification had no significant effect on the number of estimates within 10% of the actual weight. The weight of the goat was also not significantly associated with accuracy of live‐weight estimation. Results indicated that underestimation of goat live‐weight is frequent, and that livestock experience prior to and during veterinary education does not have a significant effect on weight estimation accuracy. This study highlights deficiencies in live‐weight estimation of goats in animal science and veterinary trainees, which may be a contributor to anthelmintic resistance. F12 CHRONIC (VAGAL) INDIGESTION IN ADULT DAIRY CATTLE: A RETROSPECTIVE STUDY OF 52 CASES (2004–2013) Thibaud Kuca 1, André Desrochers2, Marie Babkine2, David Francoz2, Sylvain Nichols2, François Schelcher3, Gilles Fecteau1 1Auburn University, Auburn, AL, USA, 2Université de Montréal, Saint‐Hyacinthe, QC, Canada, 3Ecole Nationale Vétérinaire de Toulouse, Toulouse, France Chronic or vagal indigestion refers to a group of functional and/or mechanical disturbances of the stomachs that result in impairment of the gastrointestinal transit. The term chronic indigestion (CI) is generally used to describe a ruminant presented with the following clinical signs: poor appetite, weight loss, abdominal and ruminal distension, dehydration, reduced fecal output and bradycardia. Many other terms have been used over the years, vagal indigestion being the one bringing the most confusion since it implies that the vagus nerve is necessarily involved in the pathophysiology of the disease. The purpose of this study was to retrospectively describe the clinical findings, etiology and outcome of adult dairy cattle diagnosed with CI at a referral teaching hospital. Medical archives were searched to include all dairy cattle older than 1 year with a diagnosis of CI presented to the Centre Hospitalier Universitaire Vétérinaire of the Université de Montréal between January 2004 and November 2013. Patients with history of clinical signs lasting <12 hours or with neither ruminal nor abdominal distension were excluded. Follow‐up information including longevity in the herd was obtained by accessing the free online database of Canadian Dairy Network. Eight records could not be retrieved from the medical archives. Two patients suffering from an acute illness and one patient with neither ruminal nor abdominal distension were excluded. Fifty‐two adult dairy cattle were finally included in this study: 1 bull, 4 heifers, 9 dry cows and 38 lactating cows. Age of patients ranged from 1.2 to 11.5 years old (median 4.1 years). Holstein (49) was the most represented breed. At the time of admission, tachycardia (HR >80 bpm) was present in 20 cases (38%) and bradycardia (HR <60 bpm) in only 4 cases (8%). CBC and biochemistry profiles performed <24 hour after admission were available for 45 cattle. Clinicopathologic findings of six cattle diagnosed with an abomasal volvulus upon arrival were not included. Lymphopenia (range, 1.53–6.75 cells × 109/L; median, 2.8 cells × 109/L; reference range, 4–10 cells × 109/L), neutrophilia (range, 1.61–11.85 cells × 109/L; median, 4.50 cells × 109/L; reference range 1.1–3.6 cells × 109/L), and hyperfibrinogenemia (range, 2.25 to 10 g/L; median, 5 g/L; reference range, 2–5 g/L) were present in 37, 31 and 20 cattle, respectively. Hypocalcemia (range, 1.73–2.46 mmol/L; median, 2.12 mmol/L; reference range, 2.22–2.7 mmol/L), hypochloremia (range, 62.5–105 mmol/L; median, 95 mmol/L; reference range, 96.4–109.2 mmol/L), and hypokaliema (range, 2.44–4.83 mmol/L; median, 3.84 mmol/L; reference range, 3.86–5.28 mmol/L) were present in 33, 27 and 23 cattle, respectively. Clinical signs were thought to result from a forestomach dysfunction in 22 cases (42%) and from an abomasal dysfunction in 15 cases (29%). Clinical diagnosis was confirmed by abdominal surgery or postmortem examination in 45 cases (87%). An exploratory laparotomy was performed in 25 cases (48%) and a rumenotomy in 19 cases (37%). Perireticular abscesses and complications after surgical correction of right abomasal displacement or abomasal volvulus were considered as causative factors in 10 cases (19%) and eight cases (15%), respectively. Liver abscesses, omasal dilatation, perforating abomasal ulcers and advanced pregnancy were also identified as primary cause of CI in three cases (6%) each. However, the precise etiology remained undetermined in 12 cases (23%). Necropsy was performed in 24 cases (46%) and vagal nerve lesions detected in only one case. Short‐term prognosis was good with 30 cases (58%) being discharged. Long‐term prognosis was fair with 13 cases (25%) remaining in the herd for at least one subsequent lactation. Three out of 10 (30%) cattle with perireticular abscesses remained in the herd for at least one subsequent lactation. 9 out of 15 (60%) cattle with an abomasal dysfunction were euthanized or died during hospitalization. The two most frequently identified etiologies of CI in adult dairy cattle were perireticular abscesses and complications after surgical correction of right abomasal displacement or abomasal volvulus. Adult dairy cattle with perireticular abscesses can be treated with a fair prognosis, while abomasal lesions appear associated with a worse outcome. F13 EVALUATION OF ENDOTOXIN IN PLASMA OF DIARRHEIC CALVES AND ITS ASSOCIATION WITH DEMEANOR, HYPER‐L‐LACTATEMIA AND HYPOGLYCEMIA Diego Gomez 1, Juan Rodriguez‐Lecompte1, Jeanne Lofstedt1, Luis Arroyo2, J. McClure1 1Atlantic Veterinary College, University of Prince Edward Island, Charlottetown, PE, Canada, 2Ontario Veterinary College, University of Guelph, Guelph, ON, Canada Information regarding the presence of lipopolysaccharide (LPS) in plasma of calves with diarrhea is lacking. We hypothesized that LPS can be detected in diarrheic calves and it's concentration is higher in nonsurviving than surviving calves. We also hypothesized that plasma LPS is associated with altered demeanour (attitude, suckling reflex and posture) and biochemical variables. This prospective study included 34 calves ≤28 days of age with diarrhea and 5 healthy age‐matched controls. Comparisons between groups were performed using t‐ or Mann‐Whitney U tests. ANOVA test was used to determine association between LPS and posture (standing, sternal or lateral recumbence), attitude (bright, depressed and comatose) and suckle reflex (strong, weak and absent). Spearman rank (r s) test was used to evaluate correlations. Plasma LPS was detected in both healthy and diarrheic calves. LPS was significantly higher in diarrheic than healthy (0.997 ng/mL; 0.976–1.044 and 0.966 ng/mL, 0.961–0.981, respectively; P = 0.008) and in nonsurviving than surviving calves (1.036 ng/mL; 1.004–1073 and 0.97 ng/mL, 0.974–0.996, respectively; P < 0.001). LPS was not associated with calf demeanor (P > 0.05), however, it was positively associated with plasma L‐lactate (r s = 0.47; P = 0.005) and negatively associated with glucose (r s  = −0.45; P = 0.008). LPS was higher in calves with hyper‐L‐lactatemia (>2.5 mmol/L) versus those without hyper‐L‐lactatemia (1.023 ng/mL; 0.995–1.069 and 0.985 ng/mL; 0.974–0.995, respectively; P = 0.01), and in hypoglycemic (<3.5 mmol/L) calves compared with normo/hyperglycemic (1.046 ng/mL; 0.976–1.32 and 0.994 ng/mL; 0.996–1.235, respectively; P = 0.02). Plasma LPS was higher in diarrheic and nonsurviving calves. LPS was higher in hyper‐L‐lactatemic and hypoglycemic calves suggesting that diarrheic calves with endotoxemia may have altered metabolic pathways. F14 EPIDEMIOLLOGIC EVALUATION OF SILICA UROLITHIASIS IN GOATS (106 Cases): PERSPECTIVE FROM THE MINNESOTA UROLITH CENTER Eugene Nwaokorie, Carl Osborne, Jody Lulich, Vachira Humpravit, Thomas Fletcher, Lisa Ulrich, Lori Koehler University of Minnesota, Saint Paul, MN, USA To (1) determine risk factors for silica urolithiasis in goats; and (2) to evaluate changes in silica urolith submission rate over the past 28 years by comparing January 1, 1984 to December 31, 1998 (period 1) and January 1, 1999 to December 31, 2012 (period 2). Study population included 832 goats of which 106 had silica uroliths and 16,366 control goats. Information about age, breed, sex, reproductive status, geographic location, season of the year, and anatomic location, were used to identify risk factors. Pygmy, Nubian, Nigerian dwarf and mixed breeds were more likely to develop silica urolithiasis than other breeds. Breeds of African origin (including Pygmy and, Nigerian Dwarf) comprised 56% of cases. Neutered male goats had significantly increased risk of developing silica uroliths. A significant association was found between breed, sex, reproductive status, geographical location, season, and anatomic location and detection of silica uroliths in goats. The proportion of silica urolith submissions had decreased from 20% in period 1 to 12% in period 2. Mean age of goats with silica increased from 2.5 ± 3.2 years in period 1 to 3 ± 2.5 years in period 2. Males comprised 96% and females comprised 4% in period 1. In period two males comprised 100% of the submissions. Results suggest that the prototypical goat with silica urocystoliths is a neutered male, 2.5–3 years and of African descent. While results of this study indicate risk factors for silica uroliths, these associations do not prove a cause and effect relationship. F15 EVALUATION OF THE CARDIOMYOTOXIC EFFECTS OF DOXYCYCLINE OVERDOSE IN CALVES USING 2‐DIMENSIONAL SPECKLE TRACKING Laureline Lecoq 1, Aurélia Leroux1, Mounir Brihoum1, Frédérick Rollin2, Alexandra Salciccia1, Geoffroy De La Rebiere De Pouyade1, Nassim Moula3, Hélène Amory1 1Department of Companion Animals and Equids, Faculty of Veterinary Medicine, University of Liege, Liège, Belgium, 2Department of Animal Production, Faculty of Veterinary Medicine, University of Liege, Liège, Belgium, 3Biostatistics, Bioinformatics and Animal Selection, Department of Animal Production, Faculty of Veterinary Medicine, University of Liege, Liège, Belgium Doxycycline (DOXY) is associated with left ventricular (LV) dysfunction in calves in accidental overdose but not in experimental models when evaluated with classical and Doppler echocardiography. Two‐dimensional‐speckle tracking (2DST) is used to eveluate LV dysfunction in numerous species but not in cattle. The aim of this study was to evaluate the cardiotoxic effects of an experimental overdose of DOXY using 2DST in calves. Ten healthy male Holstein calves. Group 1: 5 calves (mean age 58.0 ± 16.3 days; mean body weight 72.2 ± 13.0 kg) received 25 mg/kg of DOXY orally for 5 days. Group 2: 5 calves (mean age 56.4 ± 15.7 days; mean body weight 73.4 ± 7.0 kg) received a placebo. Electrocardiography (ECG) and 2DST echocardiography were performed at day 0 and day 8. ECG tracings were analysed for occurrence of arrhythmias. 2DST measurements included global and segmental, peak values for radial and circumferential strains (SR, SC), strain rates (SrR, SrC), rotation (Rot), rotation rates (RotR) and radial displacement (DR).. All calves completed the study. ECG recordings were unremarkable in both groups. Heart rate was neither significantly different between groups nor before and after treatment. LV systolic function was affected in calves receiving an overdose of DOXY as shown by a significant decrease of segmental SR (P < 0.05), SC (P < 0.05) and DR (P < 0.05) in treated calves compared to the placebo group in several segments. The SrC in early diastole was also significantly decreased in 1 segment (<0.05). In calves, DOXY overdose induces a LV dysfunction in systole, and to a lesser extent, in diastole. A better comprehension of the pathophysiology involved in the DOXY overdose will help in the treatment of accidental cases. F16 TWO‐DIMENSIONAL SPECKLE TRACKING ECHOCARDIOGRAPHY IN CALVES: FEASIBILITY, REPEATABILITY AND VARIABILITY STUDY Laureline Lecoq, Hélène Amory, Aurélia Leroux Department of Companion Animals and Equids, Faculty of Veterinary Medicine, University of Liege, Liège, Belgium Two‐dimensional speckle tracking (2DST) is a non‐invasive technique used in many species to evaluate global and regional left ventricular (LV) function; however it received little attention in the bovine species. The aim of this study was to assess the feasibility and reliability of 2DST for the evaluation of circumferential and radial LV wall motions in calves. Fourteen Holstein black calves (age: 62 ± 11.6 days; body weight: 75.25 ± 5.4 kg) were used in this observational study. Right parasternal short axis views at the level of the papillary muscles were recorded and subsequently analysed by 2DST for global and regional radial and circumferential strains and strain rates, radial displacement, rotation and rotation rate. Echocardiographic examinations were performed in unsedated, standing calves by two different observers to evaluate intra‐ and interobserver repeatability and variability. 2DST was feasible in all calves but 2 were excluded from analyses (ventricular septum defect and resting heart rate above 120 bpm, respectively). Automated tracking was better in systole than in diastole. Intraobserver repeatability was good to moderate for most systolic global and segmental peak values. Systolic peak values for radial strain and strain rate were more repeatable than for circumferential strain, circumferential strain rate and diastolic measurements. Variability of the interobserver measurements was greater than the intraobserver measurements. Two‐dimensional speckle tracking is feasible in calves and as in other species systolic radial function can be more reliably evaluated than circumferential and diastolic function. F17 VENOUS LACTATE, PH AND PCO2 LEVELS AS MORTALITY INDICATORS IN PREMATURE CALVES Hasan Guzelbektes 1, Ramazan Yildiz2, Ugur Aydogdu3, Alparslan Coskun3, Ismail Sen1 1Faculty of Veterinary Medicine, Selcuk University, Selcuklu, Konya, Turkey, 2Faculty of Veterinary Medicine, Mehmet Akif Ersoy University, Burdur, Turkey, 3Faculty of Veterinary Medicine, Cumhuriyet University, Sivas, Turkey Hyperlactatemia, hypercapnia, low pH and low O2 sat are commonly observed in premature calves. Effect on mortality of venous Lactate, pH and pCO2 Levels in 110 premature calves with Respiratory Distress Syndrome (RSD) was investigated in the study. The study was performed between 2010 and 2014. All premature calves were admitted to the clinic within 24 hours after birth. In addition to short gestational age, clinical signs of low birth weight, including an inability to stand, short and silky hair, and incomplete eruption of the incisor teeth were also present in the premature animals. Ninety‐four of the premature calves were Holsteins, 16 were Swiss‐Browns; and 42 were bull calves. Blood samples were collected anaerobically into heparinized tubes from the jugular vein and measured immediately using a blood gas analyzer. Data were expressed as means and standard deviation (Mean ± sd). Independent samples T test was used to compare the venous blood gas indicators of surviving and non‐surviving premature calves. All premature calves had low venous pH and low pO2, high pCO2 and high lactate. Total of 81 calves survived and 29 calves died within 48 hours after birth. Venous pH in non‐surviving premature calves was decreased compared to surviving premature calves, while pCO2, O2sat and lactate concentration in non‐surviving premature calves were increased (Table 1). In conclusion, the results indicate that there were positive relationship between mortality and decreasing venous pH, increasing lactate concentrations and pCO2 levels in premature calves with RSD and concurrent evaluations of these parameters could be useful in prognosis evaluation of premature calves with RSD. Table 1 Parameters in non‐surviving and surviving premature calves with RSD Parameters Non‐surviving Mean±SD (n = 29) Surviving Mean ± sd (n = 81) P pH 7.05 ± 0.16 7.29 ± 0.10 0.000 pCO2(mmHg) 78.9 ± 21.0 56.39 ± 11.4 0.000 Lac(mmol/L) 9.50 ± 3.97 5.18 ± 3.22 0.000 pO2(mmHg) 18.8 ± 6.60 19.1 ± 6.09 0.851 O2 SAT(%) 16.2 ± 9.19 25.54 ± 13.20 0.000 F18 COMPARE EFFECT OF COMBINATIONS OF INTRAVENOUS AND ORAL ELECTROLYTE SOLUTION ON TREATMENT OF CALF DIARRHEA WITH MILD–MODERATE DEHYDRATION AND METABOLIC ACIDOSIS Ismail Sen 1, Hasan Guzelbektes1, Ugur Aydogdu2, Ramazan Yildiz3, Amir Nasiri1 1Selcuk University, Faculty of Veterinary Medicine, Selcuklu/Konya, Turkey, 2Cumhuriyet University, Faculty of Veterinary Medicine, Sivas, Turkey, 3Mehmet Akif Ersoy University, Faculty of Veterinary Medicine, Burdur, Turkey The objective of this study was to compare effect of combinations of intravenous isotonic sodium bicarbonate, sodium lactate, sodium acetate and hypertonic sodium chloride solutions with oral electrolyte solution on treatment of calf diarrhea with mild –moderate dehydration and metabolic acidosis. Thirty two calves (1–30 days old, with diarrhea, 5 ‐8% dehydration and venous pH ≥7.2, base excess (BE); −2.6 to −12.8 mEq/L) were used in the study. There were four different treatment groups in the study. Treatment group 1 (n:8); Isotonic sodium bicarbonate (1.3% NaHCO3 [13 mg of NaHCO3/mL] was given via intravenous (IV) at rate 20 mL/kg/h and followed oral electrolyte solution (60 mL/kg) was administrated. Treatment group 2 (n:8) Isotonic sodium acetate (80 mL/kg) was given via IV at rate 30 mL/kg/h. and followed oral electrolyte solution (60 mL/kg) was administrated. Treatment group 3 (n:8) Isotonic sodium lactate (80 mL/kg) was given via IV at rate 30 mL/kg/h. and followed oral electrolyte solution (60 mL/kg) was administrated. Treatment group 4 (n:8) 7.2% hypertonic saline (4 mL/kg) was given via IV and followed oral electrolyte solution (60 mL/kg) was administrated. Intravenous solutions was given to all calves by infusion machine. The changes in clinic, hemodynamic, hematologic, blood gas, plasma volume, serum electrolyte and proteins were determined periodically during 24 hours following fluid administration to calves. Calves that received intravenous isotonic sodium bicarbonate, hypertonic 7.2% NaCI, sodium lactate and sodium acetate solutions along with oral electrolyte solution had an increase in venous blood pH, HCO3 concentration, and base excess within 4 hours after the beginning of the administration. Increase in plasma volume and sodium concentration, but decrease in serum total protein were observed within 0.5 hours following administration of hypertonic 7.2% NaCI +oral electrolyte solution combination compared to other solution. The results of the study show that administration of IV hypertonic 7.2% NaCI solution in small volume along with oral electrolyte solution provided fast and effective improvement of dehydration and acid‐base abnormalities within short time in treatment of calf diarrhea with mild –moderate dehydration and metabolic acidosis, compared to other treatment groups. F19 CHOLANGIOHPATITIS IN CATTLE: 10 CASES Diego Gomez 1, Elizabeth Doré2, Gilles Fecteau2, David Francoz2, Desrochers André2, Hélie Pierre2 1Atlantic Veterinary College, University of Prince Edward Island, Charlottetown, PE, Canada, 2Centre Hospitalier Universitaire Vétérinaire, Saint‐Hyacinthe, Faculté de Médecine Vétérinaire, Université de Montréal, Montreal, QC, Canada Information regarding cholangiohepatitis in cattle is lacking. This study aimed to describe the signalment, history, clinicopathologic and ultrasound findings, and outcome of adult cattle with histopathological diagnosis of cholangiohepatits. This is a report of 10 Holstein cows older than 12 weeks of age suffering from cholangiohepatitis. The diagnosis of cholangiohepatitis was based on the presence of a portal inflammatory infiltrate surrounding and/or infiltrating bile ducts, with or without epithelial damage, and extending into the adjacent lobules. Six cases were diagnosed with based on liver biopsy and four on post‐mortem examination. Reasons for veterinary consultation were mostly decreased appetite and obtundation. The median age of the cows was 5 years (range: 3 months–7 years). At admission, five cows had complete anorexia and five had decreased appetite. The median rectal temperature was 38.9°C (37.7–40°C). Median heart rate was 98 bpm (78–120 bpm). Icterus was present in three cows. Four cows were mildly dehydrated (5–7%), three moderately (8–10%) and one severely (>10%). Four cows had absence or scant feces, one diarrhea with melena. Comorbidity included photosensibilization (n = 3), abomasal dilation/displacement (n = 2), salmonellosis (n = 2), peritonitis (n = 2), glomerulonephritis (n = 1) obstruction of the bile duct (n = 1). The most remarkable hematology findings include leucocytosis (>12.000 cells/μL) in six cases, marked neutophilia (>4.000 cells/μL) in eight cases, four cows had toxic changes identified in the neutrophils and hyperfibrinogenemia (>500 mg/dL) was determined in seven cases. Alteration in the serum biochemistry profile included increased globulin levels (>4.9 mg/dL) in six cases, hypoalbuminemia (<3 g/dL) in six cases, elevated AST (>127 IU/L) in six cases, elevated GGT (>39 IU/L) in six cases, and increased total bilirubin (>75 μmol/L) was detected in four cows. Ultrasonographic examination of the liver and the gall bladder may reveal dilation of the extrahepatic and intrahepatic bile ducts, dilation of gallbladder and increased thickness of the wall. Two out six cows diagnosed by biopsy did not survive and 4 were discharged. The median time of hospitalization for the surviving cows was 9.5 days (range: 8–12). On post‐mortem examination E. coli was isolated from hepatic/gallbladder tissue from two cases. The results of this study suggested that cholangiohepatitis is a rare condition in adult cattle. Ante mortem diagnosis may be challenging because clinical signs are unspecific and laboratory results resemble those of a cow suffering of hepatic lipidosis in combination with an inflammatory process (e.g acute metritis or acute mastitis). Although the number of animals included in this study was small, it appeared that the diagnosis of cholangioheaptitis carries guarded prognosis. F20 COMPARISON OF THE PHARMACOKINETIC VALUES OF CEFQUINOME FOLLOWING INTRAMAMMARY INFUSION IN HEALTHY LACTATING GOATS AND GOATS WITH EXPERIMENTALLY INDUCED STAPHYLOCOCCAL AUREUS MASTITIS Shymaa El Badawy 1, Aziza Amer1, Gehan Kamel1, Kamal Eldeeb2, Peter Constable3 1Pharmacology Department, Faculty of Veterinary Medicine, Cairo University, Giza, Egypt, 2Molecular Chemistry Department of Drug Control Authority, Giza, Egypt, 3Department of Veterinary Clinical Medicine, College of Veterinary Medicine, University of Illinois, Urbana‐Champaign, IL, USA Values for pharmacokinetic variables are usually obtained in healthy animals, whereas drugs are frequently administered to diseased animals. There is therefore an urgent need to determine whether dosage protocols need to be altered in sick animals. Consequently, we compared pharmacokinetic values of intramammary cefquinome in healthy goats and goats with experimentally induced mastitis. Five lactating goats received two intramammary treatments of cefquinome sulfate (75 mg/udder half) at 12 hour intervals before and after induction of clinical mastitis by infusion of Staphlococcus aureus ATCC 29213 (100 cfu). Goats were milked at 12 hour intervals and milk and jugular venous blood samples were obtained periodically after cefquinome administration. Skimmed milk and plasma concentrations of cefquinome were determined using high performance liquid chromatography. Pharmacokinetic values were determined using noncompartmental methods; P < 0.05 was significant. Experimentally induced mastitis decreased the maximal cefquinome concentration and elimination half‐time in milk when compared to healthy goats. The mean residence time of cefquinome in milk was markedly lower in mastitic goats (9.2 hour) than healthy goats (17.1 hour). Cefquinome concentrations in milk exceeded 0.25 μg/mL (effective MIC for the test organism) for more than 72 hour in healthy goats and 48 hour in mastitic goats. Cefquinome was detected in milk until 120 hour in healthy animals and until 96 hour in infected animals. Plasma cefquinome concentrations were low in healthy and mastitic goats and did not exceed 0.25 μg/mL. Mastitis had no effect on the mean transient time of cefquinome in plasma. We conclude that experimentally induced clinical mastitis facilitated the absorption of cefquinome from the mammary gland of lactating goats. F21 COMPARISON OF CEFQUINOME CONCENTRATIONS AND PHARMACOKINETIC VALUES IN PLASMA AND MILK FROM HEALTHY AND MASTITIC GOATS USING A MICROBIOLOGICAL ASSAY AND HIGH PERFORMANCE LIQUID CHROMATOGRAPHY Shymaa El Badawy 1, Aziza Amer1, Gehan Kamel1, Kamal Eldeeb2, Peter Constable3 1Pharmacology Department, Faculty of Veterinary Medicine, Cairo University, Giza, Egypt, 2Molecular Chemistry Department of Drug Control Authority, Giza, Egypt, 3Department of Veterinary Clinical Medicine, College of Veterinary Medicine, University of Illinois, Urbana‐Champaign, IL, USA High performance liquid chromatographic (HPLC) techniques are currently preferred for determining withdrawal times for meat and milk after treatment; however, HPLC methodologies can be complicated and expensive and are susceptible to interference from matrix effects. Development of an alternative analytical methodology that is simple and inexpensive, such as an agar diffusion microbiological assay (MA), would be advantageous. The objective of this study was therefore to compare MA with HPLC in determining the cefquinome concentration in plasma and milk. Five adult lactating goats received each of the following two treatments of intramammary cefquinome sulfate infusion (75 mg/udder half) and intravenous injection of the same dose of cefquinome before and after induction of clinical mastitis. Intramammary infection was induced by infusion of 100 cfu of Staphylococcus aureus ATCC 29213. Jugular venous blood and milk samples were obtained periodically after cefquinome administration and plasma and skimmed milk concentrations of cefquinome was determined using MA and HPLC. The MA had an acceptable mean coefficient of variation (3.8%, skimmed milk; 3.7%, plasma), compared to 1.5% and 1.1% for HPLC. Both MA and HPLC procedures had similar limits of quantification in skimmed milk and plasma. The MA and HPLC methodologies yielded statistically similar results for cefquinome concentration, protein binding percentage, and noncompartmental pharmacokinetic variables in skimmed milk and plasma. We conclude that the MA provides a sensitive, accurate, and inexpensive method for the determination of cefquinome concentrations and pharmacokinetic values in skimmed milk and plasma from goats with or without experimentally induced Staph. aureus mastitis.