2022 ACVIM Forum Research Abstract Program (2024)

2.1.1.1. ¹Ontario Veterinary College; ²Hospital Veterinario UCV y Departamento de Medicina, Universidad Católica de Valencia San Vicente Mártir; ³Atlantic Veterinary College

Background: There is a lack of information regarding the association between canine acquired cardiac disease and the hepatic system.

Objective: Evaluate the relationship between hepatic parameters, survival, and disease stages of dogs with either dilated cardiomyopathy (DCM) or degenerative valvular disease (DVD).

Animals: Ninety‐nine client‐owned dogs consisting of 16 healthy controls and dogs diagnosed with DVD (N=61) and DCM (N=22) in either American College of Veterinary Internal Medicine stage B or C of disease.

Methods: Retrospective study evaluating the association between liver parameters, type and stage of disease, and survival. Univariate and multivariate Cox proportional hazards models were used to determine parameters associated with survival.

Results: Differences in hepatic parameter concentrations between DCM and DVD dogs and disease stages were identified. Alanine aminotransferase (P<0.001), aspartate aminotransferase (P=0.02), and total bilirubin (P=0.005) were significant predictors of mortality in univariate analysis. In the multivariate analysis, total bilirubin was the only hepatic parameter that was an independent predictor of mortality (P=0.029).

Conclusions: Differences in hepatic parameters between DCM, DVD, and disease stages are likely due to disease specific hemodynamics and disease progression. These findings suggest that in dogs with DVD and DCM, the interaction between the cardiac and hepatic systems may be relevant for disease progression and outcome, as is reported for humans with cardiac disease. Further studies into the role of hepatic function in canine cardiac disease are required.

3.1.1.1. ¹The Ohio State University; ²Auburn University; ³Philips Healthcare

Background: Some dogs with myxomatous mitral valve disease (MMVD) develop progressive volume overload and cardiomegaly. Histopathologic studies, but not antemortem evaluations, have demonstrated myocardial fibrosis in some of these dogs. Cardiac magnetic resonance imaging (CMR) aids in antemortem diagnosis of myocardial fibrosis in humans.

Hypothesis/Objectives: The objectives were to evaluate dogs with MMVD and healthy dogs for myocardial ischemia and fibrosis using serum biomarkers and CMR. We hypothesized that myocardial changes would be evident on both modalities in affected dogs.

Animals: Six dogs with MMVD stage B2 and six healthy dogs recruited from a hospital population.

Methods: Prospective case‐control study. Dogs underwent echocardiography and serum cardiac biomarker measurement (cardiac troponin I (cTnI), galectin‐3 (Gal‐3)). Dogs were anesthetized for CMR, where T2 weighted images and pre‐ and post‐gadolinium contrast T1 weighted images were acquired and analyzed on dedicated software. Data were analyzed for normality and expressed as mean±SD or median(interquartile range). Student's t‐tests or Wilcoxon rank sums tests were performed.

Results: No significant differences were observed for pre‐contrast T1 values, post‐contrast T1 values, and T2 values (msec) (MMVD 1171.33±52.77, 728(700–764.75), and 39.2±7.6; controls 1180.83±33.89, 711(697.5–736.5), and 42.38±12.62; p=0.72, p=0.57, p=0.61). Serum cTnI (ng/mL) was greater in MMVD dogs than controls (0.067±0.019 vs. 0.033±0.021; p=0.013). Serum Gal‐3 (pg/mL) was not different between groups (MMVD 178.07±105.59, controls 305.24±123.29; p=0.08).

Conclusions and Clinical Importance: Myocardial fibrosis detected by CMR and Gal‐3 were not more common in MMVD dogs than controls. Dogs with MMVD stage B2 may not have clinically relevant myocardial fibrosis.

4.1.1.1. ¹University of Georgia; ²Assistant Professor, Small Animal Medicine & Surgery, University of Georgia; ³Associate Professor, Small Animal Medicine & Surgery, University of Georgia

Background: Manual compression has been the standard‐of‐care for maintaining hemostasis after percutaneous endovascular intervention, but can be time‐consuming and associated with vascular complications. Alternative closure methods include temporary figure‐of‐eight suture (Z‐stitch) and vascular closure device (VCD) techniques.

Hypothesis: We hypothesized that compared to manual compression, Z‐stitch and VCD would significantly reduce time‐to‐hemostasis, and the proportion of dogs with vascular patency among treatments would not differ significantly.

Animals: Twenty‐one client owned dogs undergoing percutaneous transvenous interventional procedures.

Methods: Dogs with vessel diameter <5 mm were randomized to undergo manual compression (n=6) or Z‐stitch (n=7). Dogs with vessel diameter ≥5 mm were randomized to undergo manual compression (n=2), Z‐stitch (n=2), or VCD (n=4). Time to hemostasis, Bleeding Academic Research Consortium (BARC) score, and presence of vascular patency at 24 hours were recorded. Data are presented as median (95% CI).

Results: Median time to hemostasis was significantly shorter in Z‐stitch (2.9 [1.6–4.0] min) versus manual compression (10.0 [10.0–20.0] min; p<0.001) or VCD (8.5 [8.1–11.8] min); p<0.003) groups, and in VCD versus manual compression (p=0.018) groups. At 5 minutes, BARC scores were significantly higher (consistent with worse bleeding), in the manual compression, versus Z‐stitch (p=0.005), but not VCD (p=0.16) groups. There was no difference in the proportion of dogs with vascular patency at 24 hours among methods (7/8, 4/4 and 9/9 in manual compression, VCD, and Z‐stitch groups, respectively; p=0.57).

Conclusions: Z‐stitch and VCD are effective hemostasis methods, with Z‐stitch providing the most rapid hemostasis.

5.1.1.1. ¹Purdue University; ²Associate Professor, Clinical Sciences, Cummings School of Veterinary Medicine, Tufts University; ³Professor, Hospital for Sick Children (SickKids); ⁴Department of Paediatrics (Cardiology), The Hospital for Sick Children

Background: Arrhythmogenic right ventricular cardiomyopathy (ARVC) is a common entity in Boxer dogs causing syncope or sudden death. As Holter and echocardiographic abnormalities frequently do not develop until after 5–6 years of age, novel biomarkers for identification of affected individuals are needed. Circulating autoantibodies to cardiac desmosomal protein desmoglein‐2 (anti‐DSG2) have been identified in the serum of people and Boxers with ARVC.

Hypothesis/Objectives: Anti‐DSG2 antibodies will discriminate between affected and healthy Boxer dogs and correlate with disease severity as measured by 24‐h burden of ventricular premature contractions (VPCs) in affected dogs.

Animals: Client‐owned Boxer dogs ≥5 years of age were recruited.

Methods: Prospective, cross‐sectional study. The normal group was comprised of Boxers with <100 VPCs on 24‐h Holter and normal echocardiogram. The affected group was comprised of Boxers with >300 VPCs on Holter or documented ventricular tachycardia. Cardiac biomarkers (high sensitivity troponin‐I and NT‐proBNP) were evaluated, and Western blots and direct ELISA were carried out using recombinant DSG2 protein.

Results: Fifty‐one dogs were enrolled (healthy=24; affected=27). Cardiac troponin‐I (Mann‐Whitney, p<0.001) and NT‐proBNP (Mann‐Whitney, p<0.001) were significantly higher in affected dogs, but anti‐DSG2 antibody did not differ between groups (p=0.63). A positive correlation between VPCs over 24 h and anti‐DSG2 was observed (Pearson, p=0.012, r=0.4, Figure 1).

Conclusion: Anti‐DSG2 analysis did not discriminate between normal and affected dogs but may have utility for monitoring arrhythmia burden and possibly progression of ARVC.

6.1.1.1. ¹Purdue University; ²Associate Professor, Department of Clinical Sciences, Tufts University

Background: 2D‐echocardiography evaluation of the RV is limited, as a single‐plane view cannot illustrate the complex morphology of the right ventricle. Since limited criteria are available for arrhythmogenic right ventricular cardiomyopathy (ARVC) in Boxers, further diagnostic investigation is needed.

Objectives: To explore three‐dimensional right ventricular (3DRV) relationships between normal and ARVC Boxers.

Animals: Forty‐nine privately‐owned, aged‐matched (≥5yo) healthy (n=22) and ARVC afflicted Boxer dogs (n=27).

Methods: Prospective cross‐sectional echocardiography study. 3DRV data such as end‐diastolic (EDVRV) and end‐systolic (ESVRV) volumes, fractional area change (FAC), and ejection fraction (EF) were acquired. M‐mode tricuspid annulus plane systolic excursion (TAPSE), pulsed wave velocity of the lateral tricuspid annulus (S′), and body‐weight indices were also assessed.

Results: Age and body weight were similar in both groups. 3DRV analysis was consistently reproducible and data were normally distributed in the healthy group. TAPSE did not correlate to 3DRV functional parameters, such as EF (r=0.13, p=0.3) or FAC (r=0.05; p=0.68). Interestingly, S′ had positive correlation to functional parameters EF (r=0.4, p=0.005) and FAC (r=0.43, p=0.003). No correlation was found between markers of RV function, volume and body weight in this population of Boxer dogs.

Conclusions and Clinical Importance: 3DRV echocardiography might provide reproducible reference values for healthy Boxers and improve assessment of function and volume in afflicted ARVC Boxers.

7.1.1.1. ¹University of Illinois at Urbana‐Champaign; ²Assistant Professor, Veterinary Clinical Medicine, University of Illinois at Urbana‐Champaign; ³Clinical Assistant Professor, Veterinary Clinical Medicine, University of Illinois at Urbana‐Champaign; ⁴VCA Loomis Basin Veterinary Clinic; ⁵Cardiology Intern, Veterinary Clinical Medicine, University of Illinois at Urbana‐Champaign; ⁶Cardiology Resident, University of Illinois at Urbana‐Champaign; ⁷Pacific Northwest Pet ER and Specialty Center

Background: The biologic variability (BV) of a biomarker, or change in analyte concentration independent of disease progression, is required for serial monitoring. In cats, this information has not been determined for the cardiac biomarkers N‐terminal pro‐brain natriuretic peptide (NT‐proBNP) and Galectin‐3 (Gal‐3).

Hypothesis/Objectives: To determine the BV of Gal‐3 and NT‐proBNP in healthy and hypertrophic cardiomyopathy (HCM) cats.

Animals: Twelve healthy and 13 occult HCM cats.

Methods: NT‐proBNP and Gal‐3 were measured using a commercially available assays (IDEXX and RayBiotech). All cats had an examination, echocardiogram, blood pressure, and T4 analysis. Cats were classified as healthy or HCM based on septal or free wall thickness >6 mm and further staged as B1 (n=8) and B2 (n=5) based on left atrial diameter >16 mm. Cardiac biomarkers were evaluated within‐day, daily, and weekly over 6 weeks.

Results: Within‐subject (CV_I) and between‐subject (CV_G) coefficients of variation were calculated. For healthy cats: CV_I of 15.9% for Gal‐3 and 28.1% for NT‐proBNP; CV_G of 33.9% for Gal‐3 and 34.9% for NT‐proBNP. In healthy cats, a required % change of 47.9% for Gal‐3 and 78.9% for NT‐proBNP is required to exceed BV.

For HCM: CV_I of 27.9% for Gal‐3 and 29.2% for NT‐proBNP; CV_G of 67.1% for Gal‐3 and 79.7% for NT‐proBNP. In all HCM cats, a required % change of 83.9% for Gal‐3 and 83.5% for NT‐proBNP is required to exceed BV.

Conclusions and Clinical Importance: The BV of NT‐proBNP and Gal‐3 in cats should be considered when serially monitoring these cardiac biomarkers.

8.1.1.1. ¹College of Veterinary Medicine, Iowa State University; ²Ceva Sante Animale; ³Iowa State University

Background: Spironolactone has been associated with reduced risk of cardiac morbidity and mortality in humans and dogs with congestive heart failure. There is little evidence on the effect of spironolactone on biomarkers of the classical and alternative arms of the renin‐angiotensin‐aldosterone system (RAAS) in dogs.

Hypothesis/Objectives: To characterize the effect of spironolactone on the RAAS in healthy dogs.

Animals: Ten healthy purpose‐bred Beagles.

Methods: Study dogs were randomly allocated to two spironolactone dosing groups (2 mg/kg or 4 mg/kg PO q 24 hr for 7 days). Angiotensin peptides and aldosterone serum levels before and after treatment were quantified by LC‐MS/MS. Wilcoxon sum rank tests were performed to compare study variables between treatments and baseline. P‐values <0.1 were considered statistically significant.

Results: Following spironolactone treatment, dogs showed significant increases in angiotensin I (AngI median 136; IQR 77–167), AngII (median 70; IQR 39–81), and Ang(1‐5) (median 52; IQR 41–62) compared to baseline (AngI median 63; IQR 38–112; p=0.08; AngII median 32; IQR 23–52; p=0.09; Ang(1‐5) median 24; IQR 17–40; p=0.08). Changes from baseline in angiotensin peptides did not differ between study groups for the dosing range investigated.

Conclusions and Clinical Importance: Spironolactone treatment increases the activity of the classical and alternative RAAS pathways in healthy dogs. A greater pharmacological response was not observed for the higher dosing group, suggesting that a 2 mg/kg q 24 hr oral dose of spironolactone already reaches a plateau activity on RAAS.

9.1.1.1. ¹University of Missouri; ²Lamb Consulting LLC

Background: Non‐hemorrhagic ascites (NHA) can be caused by cardiac and non‐cardiac disease. N‐terminal brain natriuretic peptide (NT‐proBNP), cardiac troponin‐I (cTnI) and point‐of‐care ultrasound (POCUS) may aid in differentiating between cardiac and non‐cardiac causes of NHA.

Hypothesis/Objectives: This study compared NT‐proBNP and cTnI concentrations in dogs presenting for cardiac and non‐cardiac causes of NHA. This study also compared POCUS findings, specifically hepatic venous and caudal vena cava (CaVC) distension, and gall bladder wall edema.

Animals: Client‐owned dogs (n=60) were enrolled based on identification of NHA with an effusion packed cell volume <10%.

Methods: Blood samples and POCUS were collected from dogs. Dogs were diagnosed with cardiac (n=28) or non‐cardiac (n=32) NHA based on echocardiography. The cardiac group was subdivided into pericardial disease (n=11) and non‐pericardial cardiac disease (n=17).

Results: A significant difference in NT‐proBNP concentration (mean±SD pmol/L) was noted between cardiac (4869±4078) and non‐cardiac (1882±2627) groups (p=0.022), with a sensitivity of 53.8% and specificity of 85.7% at a cut‐off of 4092 pmol/L. The NT‐proBNP concentrations were also significantly different between pericardial (1179±1398) and non‐pericardial cardiac (7177±3435) groups (p<0.0001). There were no significant differences in cTnI concentrations between groups. Significant differences in POCUS findings between the cardiac and non‐cardiac groups included hepatic venous distension (18/28 vs. 3/29, p<0.0001) and CaVC distension (13/27 vs. 2/29, p<0.001), but not gall bladder wall edema (4/28 vs. 3/29, p=0.7373).

Conclusions and Clinical Importance: NT‐proBNP and POCUS help distinguish between cardiac and non‐cardiac causes of NHA.

10.1.1.1. ¹Purdue University; ²University of Canberra; ³Professor, Veterinary Clinical Sciences, University of Melbourne; ⁴Professor, Veterinary Clinical Sciences, Purdue University

Background: Vetmedin® is commonly prescribed for the treatment of cardiac disease in dogs. Compounded formulations of generic pimobendan are commercially available and may offer more accurate dosing at a more affordable cost.

Hypothesis/Objectives: The aim of this study was to evaluate the pharmacokinetic biosimilarity of pimobendan and its active metabolite, O‐deoxymethyl pimobendan (ODMP) following a single oral dose of Vetmedin® (V), pimobendan capsule (C), and pimobendan suspension (S) in healthy dogs.

Animals: Six healthy, purpose‐bred research beagle dogs.

Methods: This was a randomized cross‐over study. Each dog received one formulation of pimobendan at a mean dose of 0.33 mg/kg (range: 0.3–0.35) and blood was collected for pharmacokinetic analysis; this was repeated for the other two formulations following a 48‐hour washout period after administration of each formulation.

Results: Noncompartmental modeling was used. The pimobendan geometric mean Cmax was 51.9 ng/mL (V), 51.2 ng/mL (S), and 36.1 ng/mL (C) while the ODMP Cmax was 45.3 ng/ml (V), 44.5 ng/ml (S), and 33.2 ng/ml (C). AUC were compared between formulations using a hierarchical mixed model. The ratio of AUC between V and S fell within the a priori acceptable range for bioequivalence (1.01; 0.88–1.16 CrI 90%), while the AUC ratio between V and C was 1.17 (0.93–1.44 CrI 90%).

Conclusions and Clinical Importance: Pimobendan exposure was similar between the generic suspension and Vetmedin®, but lower with the generic capsule. The results support that generic suspension and Vetmedin® are clinically exchangeable. Generic pimobendan capsule may be clinically satisfactory based on available varying dosing strategies or may require dose adjustment.

11.1.1.1. ¹University of California, Davis; ²Assistant Professor, Department of Surgical and Radiological Sciences, University of California, Davis; ³Department of Surgical and Radiological Sciences, University of California, Davis; ⁴Department of Medicine and Epidemiology, University of California, Davis; ⁵Chief Veterinary Medical Officer, Department of Medicine and Epidemiology, University of California, Davis

Background: Congestive heart failure (CHF) secondary to canine myxomatous mitral valve disease (MMVD) is common, and adjunct therapies for acute decompensation are lacking. Afterload reduction enhances cardiac output, but vasodilator options are limited. Clevidipine is a novel intravenous dihydropyridine calcium channel blocker that may aid in canine CHF management.

Hypothesis/Objectives: Clevidipine, compared to standard of care (SOC), will result in a 20% reduction in mean arterial pressure (MAP) in a dose‐dependent manner without adverse effects when used for afterload reduction in CHF secondary to MMVD. A secondary objective was to determine whether oscillometric blood pressure measurement (OSC) provides a reliable assessment of MAP in treated patients.

Animals: Ten client‐owned dogs in acute decompensated CHF secondary to MMVD.

Methods: Prospective, randomized, open‐label, clinical trial. Dogs enrolled received SOC alone or with clevidipine (CLEV) with continuous direct arterial (ART) and OSC monitoring throughout treatment.

Results: Target MAP was achieved in all CLEV patients and MAP was significantly decreased compared to SOC (p=0.0005) at doses 4.5–16.5 mcg/kg/min (5.25, IQR 4.50–9.75 mcg/kg/min). Comparison of ART and OSC methods demonstrated overestimation of MAP in both SOC and CLEV groups (bias −13.99, −15.45). There was no significant difference in BUN or creatinine following treatment in either group, and no notable hypotension was appreciated.

Conclusions and Clinical Importance: Clevidipine was well‐tolerated and resulted in a dose‐dependent reduction of MAP in dogs with MMVD and CHF. Direct arterial blood pressure monitoring is recommended for monitoring response to clevidipine therapy.

12.1.1.1. ¹University of Florida; ²University of Minnesota

Background: Hypertrophic Cardiomyopathy (HCM) is the most common inherited cardiomyopathy in cats, affecting approximately 10–15% of domestic cats. Two separate, breed‐specific genetic mutations have been linked to the development of HCM in Maine Coons and Ragdolls. Previous work has demonstrated the utility of fibroblasts as easily accessible in vitro models for studying inheritance, pathophysiology, and possible novel therapies for cardiomyopathies such as HCM.

Hypothesis/Objective: To establish a novel, non‐invasive sampling technique to create a bank of fibroblast cells for use as a future HCM drug screening platform.

Animals: Maine Coon and Ragdolls with or without history of cardiomyopathy were enrolled from breeding clubs and the University of Florida.

Methods: Cats received physical and echocardiographic examinations. Buccal mucosal swabs were collected for genotyping for the A31P and R820W mutations through North Carolina State University. Vibrissae, including follicle, were collected and fibroblasts cultured.

Results: 21 Maine Coons and 12 Ragdolls were enrolled. All cats were found to be wildtype for the mutations of interest. All Ragdolls were normal on echocardiogram. Two Maine Coons were diagnosed with cardiomyopathy. Following protocol optimization, whisker papilla fibroblast cultures were successfully isolated from 3/3 Maine Coons and 3/3 Ragdolls.

Conclusions and Clinical Importance: Vibrissae serve as a non‐invasive source of fibroblasts from cats. The identification of HCM phenotype in two cats despite wildtype genotype reinforces that development of HCM is multifactorial. Banking of feline fibroblasts has been initiated, and further expansion will facilitate in vitro studies of the physiology of HCM and enable testing of therapies.

13.1.1.1. ¹Veterinary Teaching Hospital, National Chung Hsing University; ²Walk Animal Hospital; ³National Chung Hsing University

Background: Predictive variables and survival time of advanced heart failure (AHF) secondary to myxomatous mitral valve disease (MMVD) in dogs remain unknown.

Hypothesis/Objectives: This retrospective case‐control study investigated predictive variables of AHF in dogs with MMVD.

Animals: The study included 38 dogs diagnosed with AHF and 38 with stable congestive heart failure (CHF) secondary to MMVD. AHF group was determined as dogs with refractory CHF treated with a total daily dosage of 8 mg/kg furosemide or the equivalent dose of torsemide. Dogs were included in the stable group, if there were no changes in the maintenance dose of furosemide (lower than 8 mg/kg/day) and other cardiac medicines above 12 weeks.

Methods: Logistic regression modeling was used to compare and analyze clinical, radiographic, and echocardiographic variables between groups. Survival times were analyzed using the Kaplan‐Meier method.

Results: Left atrium to aortic root ratio, normalized left ventricular dimension at the end‐diastole and end‐systole, isovolumic relaxation time (IVRT) and the ratio of early transmitral inflow velocity to IVRT were associated with AHF progression. The median survival times in the stable group were significantly longer than those in the AHF group. After the diagnosis of AHF, the median survival times of all‐cause mortality and cardiogenic mortality were 194 days (range 0–1,088 days) and 354 days (range 0–1,063 days), respectively.

Conclusions and Clinical Importance: AHF was significantly associated with enlarged left atrium, left ventricle and elevated left atrial pressure in dogs with MMVD; these dogs exhibited a survival time of approximately six months.

14.1.1.1. ¹Colorado State University; ²Unaffiliated; ³Associate Professor, Clinical Sciences, Colorado State University

Background: Deep learning (DL) models have been validated for human electrocardiogram (ECG) analysis, but there is minimal experience with DL models in veterinary ECG interpretation.

Objective: This study aimed to develop a DL model for the interpretation of narrow‐complex ECGs in dogs and to determine model accuracy against the rhythm diagnosis reported in the medical record.

Animals: ECG recordings from 346 dogs were selected from the ECG database of the cardiology service of the Colorado State University Veterinary Teaching Hospital, acquired between October 2020 and November 2021.

Methods: The DL model was programmed to classify canine ECGs into sinus rhythm, second‐degree atrioventricular block, third‐degree atrioventricular block, supraventricular ectopic complexes, and atrial fibrillation/flutter. Wide complex QRS morphologies were excluded due to similarities between ventricular origin and aberrant conduction of supraventricular rhythms. ECGs with multiple concurrent arrhythmias or excessive artifact were excluded. The DL model was trained utilizing ECG metadata, Fourier transformed data, and 6‐lead waveforms. Balanced data perturbation was employed. The model was trained using 271 recordings and tested with 75 additional recordings.

Results: The DL model accuracy was 93% for sinus, 80% for third‐degree atrioventricular block, 80% for supraventricular complexes, 66% for second‐degree atrioventricular block, and 25% for atrial fibrillation/flutter. Negative predictive values for these rhythms varied from 85% to 98%. Model accuracy was 81% using ECG metadata, 72% by Fourier transform, and 37% using hexaxial lead inputs.

Conclusions and Clinical Importance: DL models are a potential future diagnostic aid in canine ECG interpretation with further model refinement ongoing.

15.1.1.1. Veterinary Teaching Hospital, University of California, Davis

Background: The role of platelets (PLT) and their immune function in pulmonary hypertension (PH) remain poorly understood. Evidence in humans suggests vascular changes are secondary to PLT activation and thrombosis.

Hypothesis/Objectives: To assess PLT activation, PLT‐leukocyte interactions and thromboelastography (TEG) in dogs with PH.

Animals: 29 client‐owned dogs. Group 1=5 dogs with PH secondary to myxomatous mitral valve disease (MMVD), and 10 control MMVD dogs without PH. Group 2=5 dogs with PH due to respiratory diseases and 4 dogs with PH due to other causes, and 5 control dogs with respiratory diseases only.

Methods: Prospective observational case‐control study. PH diagnosis was based on ACVIM guidelines. Dogs underwent echocardiogram, thoracic radiographs, heartworm antigen test, TEG, and flow cytometry for PLT P‐selectin expression and PLT‐leukocyte interaction. Platelet activation and PLT‐leukocyte aggregates were measured as P‐selectin median fluorescence intensity (MFI) and percent‐positive events.

Results: Dogs with MMVD and PH had augmented PLT response to ADP (7153±4673 vs. 3110±1328, p=0.021) and thrombin (14,126±4280 vs. 5374±2648, p=0.0003) compared to control MMVD dogs. PH dogs in Group 2 also had augmented thrombin‐mediated activation compared to controls (20588±31693 vs. 4864±1364, p=0.02). Platelet‐leukocyte interaction was higher following ADP‐induced activation in MMVD dogs with PH (25.01%±15.24 vs. 12.92%±4.18, p=0.04). No significant differences were found in any TEG parameters between dogs with and without PH.

Conclusion and Clinical Importance: Platelets are hyper‐reactive in dogs with PH, thereby increasing platelet‐leukocyte aggregate formation. Platelet‐leukocyte interaction likely plays a role in the pathogenesis of PH in dogs.

16.1.1.1. IDEXX Laboratories, Inc.

Background: The vertebral heart score (VHS) is a measurement used to index the heart size to the thoracic vertebra and is a useful tool for identifying the presence of heart disease, staging heart disease, and providing prognostic information. There is increasing development of computer‐aided algorithms to perform specific tasks, which can reduce radiograph reading times and help minimize human error.

Hypothesis/Objectives: The purpose of this study was to validate the use of a VHS algorithm compared to manual VHS scoring by three board‐certified veterinary cardiologists.

Animals: A total of 1,200 canine lateral radiographs were randomly selected from a commercial telemedicine databank.

Methods: A convolutional neural network centered around semantic segmentation of relevant anatomical features was developed to predict heart size and vertebral bodies. These predictions were used to calculate the VHS score. Three board‐certified cardiologists manually scored 400 images each using the traditional Buchanan method. Post‐scoring, the cardiologists evaluated the algorithm for appropriate algorithm landmarks and overall image quality.

Results: The 95th percentile absolute error between the cardiologist VHS score and the algorithm VHS score was 1.05 vertebrae (95% CI: 0.97–1.20 vertebrae) with a mean error of −0.09 vertebrae (95% CI: −0.12–−0.05 vertebrae) and an absolute mean error of 0.3 vertebrae. The model was well‐calibrated across the predictive range.

Conclusions and Clinical Importance: We found the performance of the VHS algorithm comparable to that of three board‐certified cardiologists.

17.1.1.1. ¹Ghent University; ²Department of Internal Medicine, Reproduction and Population Medicine, Ghent University; ³Department of Cardiology, AZ Groeninge, Belgium; ⁴Universitair Ziekenhuis Gent (Ghent University Hospital), Ghent University; ⁵Department of Anesthesiology, Ghent University; ⁶Department of Medical Imaging, Ghent University; ⁷Small Animal Department, Ghent University

Background: Three‐dimensional electroanatomical mapping (3D EAM) has expanded radiofrequency catheter ablation (RFCA) applications in humans to almost all complex arrhythmias and has drastically reduced fluoroscopy use, yet its potential in dogs is poorly investigated.

Objectives: To assess the feasibility and safety of 3D EAM of all four heart chambers in dogs using minimal fluoroscopy.

Animals: Eight Beagles (median weight 12.2 kg (11.2–15.7)).

Methods: Prospective experimental trial. Three‐dimensional EAM was attempted during sinus rhythm under general anesthesia using a 22‐electrode mapping catheter and CARTO 3 system. Left heart catheterization was achieved by retrograde transaortic access (n=4) or transseptal puncture (TSP) (n=4). Primary outcomes included feasibility, safety and fluoroscopy time.

Results: Successful 3D EAM of the right atrium and ventricle was achieved in all dogs. Left atrial and ventricular 3D EAM was achieved in six and seven dogs respectively. Median fluoroscopy time was 6.8 min (0.0–45.1) and almost exclusively associated with TSP. Complications requiring intervention occurred in one dog only and were a transient third‐degree atrioventricular block after interatrial septum engagement, which reversed after intracardiac pacing (10.3 min), and development of pericardial effusion following TSP treated by pericardiocentesis. Peri‐operative self‐limiting arrhythmias were frequently observed. Only two dogs received lidocaine to resolve intra‐operative ventricular tachycardia.

Conclusions and Clinical Importance: Three‐dimensional EAM of all cardiac chambers is feasible, requires minimal fluoroscopic assistance and is promising to guide RFCA of arrhythmias in dogs. Complications were mainly related to TSP and similar to those reported in people.

18.1.1.1. ¹University of Illinois at Urbana‐Champaign; ²Assistant Professor, Veterinary Clinical Medicine, University of Illinois at Urbana‐Champaign; ³Cardiology Resident, Veterinary Clinical Medicine, University of Illinois at Urbana‐Champaign; ⁴Clinical Assistant Professor, Veterinary Clinical Medicine, University of Illinois at Urbana‐Champaign; ⁵Cardiologist, Pacific Northwest Pet ER and Specialty Center

Background: N‐terminal pro‐brain natriuretic peptide (NT‐proBNP) is useful in identifying Doberman Pinschers with echocardiographic evidence of dilated cardiomyopathy (DCM). An independently validated, quantitative point‐of‐care (POC) NT‐proBNP assay (Vcheck, BioNote USA Inc) is now commercially available and uses fresh serum, eliminating the need for specialized sample processing and shipping.

Hypothesis/Objectives: The purpose of this study was to prospectively screen a cohort of adult Dobermans for DCM using the Vcheck NT‐proBNP assay, echocardiography, and 3‐minute electrocardiography.

Animals: One hundred and twenty‐seven adult Doberman Pinschers were screened.

Methods: Tests performed included physical examination, transthoracic echocardiography with simultaneous 3‐lead electrocardiography, and quantitative NT‐proBNP levels using the Vcheck assay (detection range of 500–10,000 pmol/L).

Results: Population median age was 5 years (2–12 years), and there were 74 females and 53 males. Diagnoses included normal (n=103), DCM (n=13), Equivocal/VPC only (n=5), and MMVD (n=6). Median (range) NT‐proBNP for normal=499 pmol/L (499–1585) and DCM=900 pmol/L (511–2604) P=0.001. A cutoff of >509 pmol/L=100% sensitivity and 55% specificity, the highest Youden Index was for >615 pmol/L=85% sensitivity and 81% specificity. To achieve a 95% specificity a cutoff of >892 pmol/L was needed; however, this yielded a sensitivity of 46%.

Conclusions and Clinical Importance: This study indicates that quantitative NT‐proBNP using a novel POC analyzer is useful as a screening test to detect DCM in Doberman Pinschers and compares favorably with previous studies.

19.1.1.1. ¹Swedish University of Agricultural Sciences; ²Professor, Clinical Sciences, Swedish University of Agricultural Sciences; ³Professor, University of California, Davis; ⁴AniCura Albano Small Animal Hospital; ⁵University of California, Davis, ⁶Associate Professor, Clinical Sciences, Swedish University of Agricultural Sciences

Background: Taurine deficiency‐induced dilated cardiomyopathy (DCM) has been reported in Cocker Spaniels (CS).

Objectives: To investigate the occurrence of low blood taurine concentrations in a Swedish population of CS presenting as clinically healthy or with a DCM phenotype and congestive heart failure (CHF), and to assess associations between taurine concentrations and dog characteristics and clinical features.

Animals: One‐hundred seventy privately owned CS (168 English (ECS) and two American (ACS)).

Methods: Prospectively recruited dogs underwent physical and echocardiographic examination, and blood analyses (hematology and biochemistry). Taurine concentrations were analyzed in plasma (EDTA and heparin), and whole blood (WB). Dogs were classified as having low taurine concentrations if results were below reported reference range in either type of plasma and/or WB.

Results: A total of 46 dogs (27%) had low taurine concentrations. Dogs with low taurine concentrations were older compared to dogs with normal concentrations (P=0.03). Other dog characteristics and clinical features did not differ between the groups. Eight of 46 dogs (17%) with low taurine concentrations, of which seven were ECS one ACS, presented with a DCM‐phenotype and CHF. Heparin was the additive that identified the highest number of dogs with low taurine concentrations.

Conclusions and Clinical Importance: Low taurine concentrations were common in the population of CS investigated and were found in both clinically healthy dogs as well as in all dogs with CHF. Besides age, no difference in dog characteristics and clinical features were found between dogs with normal and low taurine concentrations.

20.1.1.1. ¹Swedish University of Agricultural Sciences; ²Professor, Clinical Sciences, Swedish University of Agricultural Sciences; Professor, ³Texas A&M University; ⁴Associate Professor, Anatomy, Physiology and Biochemistry, Swedish University of Agricultural Sciences; ⁵Professor, Companion Animals, University of Prince Edward Island; ⁶AniCura Albano Animal Hospital; ⁷College of Veterinary Medicine, Cornell University; ⁸Chuan Animal Hospital; ⁹Cardiospecial Veterinary Hospital; ¹⁰Associate Professor, Department of Clinical Sciences, Swedish University of Agricultural Sciences

Background: Veterinary echocardiographers’ preferences for left atrial (LA) size assessment in cats have never been systematically investigated.

Hypothesis/Objective: Investigate echocardiographers’ preferences concerning LA size assessment in cats including positioning of the cat, acquisition views, indexing methods and timing; and whether geographic/professional profiles could be associated with echocardiographers’ preferences for assessing LA size in cats.

Animals: None.

Methods: An online survey was distributed globally to veterinary echocardiographers.

Results: A total of 655 veterinary echocardiographers from 6 continents and 54 countries, working in specialty practice (56%) and in general practice (38%), completed the survey. Linear 2D‐based methods were favored by most echocardiographers (n=612). Most commonly, echocardiographers combined linear 2D‐based methods with subjective assessment (n=227), while 209 used linear 2D‐based methods alone. Most echocardiographers (n=464) using linear 2D‐based methods preferred the right parasternal short‐axis view and to index the LA to the aorta. Less than 8% (n=43) used linear 2D obtained images from a right parasternal long‐axis 4‐chamber view without indexing the LA. Approximately one third (n=210/612, 34.3%) assessed LA size using linear 2D echocardiography shared the same preferences regarding cat position, acquisition view, indexing method and time‐point identification for the measurement. The responses were comparably homogeneous across geographic location, level of training, years performing echocardiography, and type of practice.

Conclusion and Clinical Importance: Most veterinary echocardiographers assessed LA size in cats using linear 2D echocardiography from a right parasternal short axis view, indexing the LA to the aorta. The echocardiographers’ demographic and professional profiles had a minor influence on the responses.

21.1.1.1. ¹Cummings School of Veterinary Medicine, Tufts University; ²Tufts University; ³Tufts Medical Center

Background: Toceranib phosphate is a tyrosine kinase inhibitor (TKI) and a non‐selective inhibitor of Vascular Endothelial Growth Factor (VEGF). While hypertension has been documented as a side effect in canine patients receiving TKIs, to date, there have not been studies evaluating left ventricular systolic function and endothelial function in canine patients.

Objectives: To characterize changes in echocardiographic parameters and biomarkers of endothelial function in canine patients treated with toceranib.

Animals: 12 client‐owned dogs diagnosed with neoplasia (mast cell tumors N=5, anal sac adenocarcinoma N=4, thyroid carcinoma, mesothelioma, and multiple N=1) receiving single agent toceranib for cancer treatment.

Methods: This is a prospective study approved by the Tufts University IACUC. Patients had echocardiographic exams and blood pressure measurements. Blood and urine samples were obtained to evaluate levels of VEGF, endothelin‐1, platelet derived growth factor (PDGF), prostacyclin, nitric oxide (nitrite and nitrate) metabolites, and cGMP using commercially available ELISA kits. Examination time‐points include baseline, one‐, three‐, and five‐ months after start of treatment. Statistical analysis was performed using repeated measures mixed models.

Results: Statistically significant decreases in urine nitrate and plasma prostacyclin concentrations were detected over time. Although not reaching statistical significance, VEGF and systolic blood pressure increased. Four of the study dogs required treatment with hypertension medication (BP >170 mm Hg). No significant changes in ejection fraction or fractional shortening were observed.

Conclusions: Biomarkers of endothelial dysfunction associated with hypertension, including prostacyclin and nitrate, decreased during treatment with toceranib. Larger sample size will be required to confirm the findings of this study.

22.1.1.1. ¹Ghent University; ²Professor, Small Animal Department, Ghent University; ³Small Animal Department, Ghent University; ⁴Department of Veterinary and Biosciences, Ghent University; ⁵Professor, Department of Veterinary and Biosciences, Ghent University

Background: Combretastatin A4‐phosphate (CA4P) is a vascular disrupting agent recently described for treatment of solid tumors in dogs. Conventional echocardiography did not show signs of cardiotoxicity. However, the gold standard for assessing myocardial damage in humans receiving cardiotoxic chemotherapeutics is two‐dimensional speckle tracking echocardiography (2D‐STE).

Hypothesis/Objectives: The aim of this study was to evaluate the cardiotoxic effect of CA4P in dogs, based on 2D‐STE measurements.

Animals: Echocardiographic images of healthy dogs and cancer patients that received CA4P were retrospectively reviewed. Seven healthy beagles and five cancer patients were included.

Methods: Peak global longitudinal strain (GLS), peak global circumferential strain (GCS) and peak global radial strain (GRS) were measured before and 24 hours after administration of CA4P. Strain measurements were compared with cardiac troponin I (cTnI) measurements. On seven echocardiographic examinations, each strain parameter was measured by 3 observers on 3 consecutive days to obtain intra‐observer and inter‐observer variability.

Results: Median GCS and GLS values decreased by 17.6% (P=0.002) and 24.0% (P=0.0005), respectively. Median GRS values were not significantly different (P=0.16). The decrease in GLS was correlated with a rise in cTnI (Spearman rho = −0.63, P=0.03). The intra‐observer coefficient of variation was 0.04, 0.13 and 0.09 for GCS, GRS and GLS, respectively, while the corresponding inter‐observer coefficients were 0.12, 0.20 and 0.11.

Conclusions and Clinical Importance: Administration of CA4P has a cardiotoxic effect, which can be detected by 2D‐STE strain measurements and correlates with cTnI concentrations.

23.1.1.1. ¹Federal University of Uberlandia; ²Federal University of Minas Gerais; ³Purdue University

Background: The agreement between systolic arterial pressure (SAP) measured by Doppler ultrasonic (DU) and by high‐definition oscillometric device (HDO) in hypotensive, normotensive, and hypertensive dogs has not yet been evaluated.

Objectives: To assess limits of agreement (LoA) of SAP measured by DU and HDO normotensive (90≤SAP≤160 mm Hg), hypotensive (SAP <90 mm Hg) and hypertensive (SAP >160 mm Hg) dogs.

Animals: Sixty privately‐owned, non‐anesthetized dogs were studied: normotensive (n=34), hypotensive (n=10) and hypertensive (n=16).

Methods: Prospective clinical study. Same cuff sizes were used for both methods, and readings were not recorded simultaneously. Five consecutive SAP readings were obtained from each dog with DU and HDO. All measurements were analyzed according to the criteria outlined by the ACVIM consensus statement.

Results: A moderate correlation was shown in normotensive (r=0,53; p<0.001) and hypertensive (r=0.52; p<0.038) groups, while a strong correlation was documented in the hypotensive (r=0,81; p<0.004) group. The SAP bias for normotension and hypotension groups was 1.00±13.55 mm Hg (LoA: −25.62 to 27.52) and −3.50±7.61 mm Hg (LoA: −18.42 to 11.43 mm Hg), respectively. No significant differences between SAP bias by DU and HDO in normotensive (P=0.93) and hypotensive (P=0.452) were appreciated.

Conclusion: Good agreement between methods was shown in normotensive and hypotensive groups for SAP. Since values of MAP and DAP measured by HDO are more accurate than SAP, and DU cannot provide these values, further studies evaluating agreement with invasive pressure are needed.

24.1.1.1. ¹Friendship Hospital for Animals; ²VCA West LA; ³ACCESS; ⁴Idexx; ⁵Oracle Animal Hospital; ⁶VCA Veterinary Specialists of the Valley

Background: A previous study evaluating medication adherence in dogs with cardiovascular disease reported medication adherence was high. Medication adherence for cats with cardiovascular disease remains unexamined.

Hypothesis/Objectives: To examine chronic medication adherence for the management of feline cardiovascular disease in the referral setting.

Methods: Over a three‐month period, a questionnaire was distributed to owners at each investigator's hospital which employed a board‐certified veterinary cardiologist or a cardiology resident supervised by a board‐certified veterinary cardiologist. For inclusion into the study, owners had to have been medicating their cat with at least one cardiac medication requiring daily administration for at least one month. The cat did not need to have been diagnosed with congestive heart failure to participate in the study.

Results: Fifty‐four questionnaires were available for review. The most common diagnosis was hypertrophic cardiomyopathy (HCM) (31/54, 57.4%). Clopidogrel was reported as the medication most difficult to consistently administer (13/54, 24.0%) although 20 owners (20/54, 37.0%) reported no difficulties chronically administering cardiac medications. Thirty‐six (66.7%) owners reported the highest frequency of medications they could administer was twice‐a‐day dosing. Medication adherence in this study population was high with 50 owners (92.6%) meeting the criteria for medication adherence.

Conclusions and Clinical Importance: Medication adherence was reportedly high in this study. Veterinary cardiologists should be aware of owner concerns when determining treatment protocols for management of feline cardiovascular disease.

26.1.1.1. ¹University of Missouri; ²North Carolina State University; ³Professor and Section Head, Neurology and Neurosurgery, University of Missouri; ⁴Professor, Veterinary Pathobiology, University of Missouri

Background: Canine degenerative myelopathy (DM) is a debilitating condition. The DM‐associated SOD1:c.118G>A genetic variant is the most common disease‐associated variant in dogs. Risk of DM associated with this variant is a controversial topic.

Hypothesis/Objectives: Further characterize age‐ and breed‐related risk for development of DM associated with the SOD1:c.118G>A variant.

Animals: Dogs with DNA voluntarily submitted for SOD1:c.118G>A variant screening were identified for follow‐up, and 2,143 dogs were included following owner response.

Methods: Dogs were selected from a population screened for the SOD1:c.118G>A variant prior to age 6, who were 7 or older at follow‐up. A survey was emailed to assess myelopathic signs. Responses were reviewed independent of breed, sex, or genotype, to assign dogs as having DM compatible signs or not. Kaplan‐Meier survival analysis and Cox proportional hazards regression were evaluated.

Results: Cumulative event curves revealed age‐related risk of disease that varies by genotype (Figure 1) and breed. Rhodesian Ridgebacks, Chesapeake Bay Retrievers, Boxers, German Shepherd Dogs, and Pembroke Welsh Corgis had sufficient numbers for analysis, and 83 breeds were represented by at least one dog. DM hazard ratio was increased over wild‐type in homozygous variant and heterozygous dogs.

Cumulative event curve showing cumulative risk divided by SOD1:c.118G>A genotype. Animals are censored at death or ascertainment. N: A/A=292, G/A=759, G/G=1092. Log‐rank p value = < 0.0001

Conclusions and Clinical Importance: This study illustrates effect of SOD1:c.118G>A genotype on development of DM‐compatible signs in several breeds, including increased risk in heterozygous dogs over wild‐type. Agreement between clinical signs and genotype underscores use of genetic testing as an adjunctive diagnostic tool and for screening of breeding animals.

27.1.1.1. ¹University of Missouri; ²Beth Israel Deaconess Medical Center, Harvard Medical School; ³Veterinary Health Center, University of Missouri

Background: Electrical impedance myography (EIM) is a tool for muscle assessment based on voltages obtained with application of high‐frequency, low‐intensity electrical current. The phase value metric is calculated based on resistance and reactance values.

Hypothesis/Objectives: To assess normal EIM values in healthy dogs and determine an age dependence to the metric.

Animals: 73 healthy dogs (62 males, 49 females) were sampled for a total of 111 muscle observations. Mean age was 5.1 years (0.63–13.48 years), and mean body condition score was 5 (4 to 6).

Methods: EIM measurements were performed on bilateral cranio tibialis, gastrocnemius, and sartorius muscles and assessed at 100 kHz phase. A second‐degree polynomial regression was used to fit for phase with age and extract the lower limit of prediction for sex.

Results: Lower limits of normal were established across a range of ages. Data suggested modest reduction in phase values with age, consistent with expected decline in muscle condition. The average value decreased from 16.6° in 2‐year‐old male dogs to 12.7° in 13‐year‐old male dogs (23.4% reduction) and decreased from 15.4° in 2‐year‐old female dogs to 12.1° in 13‐year‐old female dogs (21.4% reduction).

Conclusions and Clinical Importance: EIM is a non‐invasive, quantitative measure of muscle pathology utilized to track muscle condition in healthy dogs. There is a small but consistent reduction in EIM values with increased age. The normal EIM values established here can serve as reference for future studies in disease.

28.1.1.1. ¹University of Missouri; ²Beth Israel Deaconess Medical Center, Harvard Medical School; ³College of Veterinary Medicine, University of Missouri

Background: Canine degenerative myelopathy (DM) causes disuse and neurogenic muscle atrophy. Electrical impedance myography (EIM) has been used to quantify muscle pathology in neuromuscular disorders.

Hypothesis/Objectives: To compare EIM between DM‐affected and similar aged healthy dogs, and assess EIM changes over time in DM‐affected dogs.

Animals: Eleven DM‐affected (histopathologically confirmed) and fifteen similar aged neurologically normal dogs.

Methods: A prospective observational study evaluated multifrequency EIM performed on DM‐affected dogs at baseline and during disease progression. Muscles evaluated included bilateral cranio tibialis, gastrocnemius, gracilis, sartorius, and biceps femoris. The 100 kHz phase angle (θ) was extracted from the full frequency set for analysis.

Results: As compared to normal dogs, DM‐affected dogs had lower phase values of the gastrocnemius on the left (θ=7.69, 13.06; p=0.002) and right (θ=6.11, 11.72; p=0.001) and lower mean phase values of combined muscles on the left (θ=9.24, 11.62; p=0.012) and right (θ=9.18, 11.72; p=0.021). Other individual muscles measured did not show significant differences, although values were consistently lower in DM‐affected dogs. A significant (p<0.05) decline in phase of the gastrocnemius was seen with disease progression over time in serially monitored DM‐affected dogs.

Conclusions and Clinical Importance: EIM can detect pathologic changes in muscles of DM‐affected dogs and phase progressively declines with disease progression. As a non‐invasive and quantitative measure of muscle pathology, EIM may be useful in future studies for tracking disease progression and response to therapy.

29.1.1.1. Cornell University

Background: A subset of intervertebral disk extrusion (IVDE) cases has compressive, hemorrhagic inflammation within the epidural space in addition to compression from the herniated disk material. We term this hemorrhagic extradural longitudinal parenchymal (HELP) compression. Previous reports have been conflicting in the outcomes of these cases.

Hypothesis/Objectives: The goals are to compare success of surgical decompression in dogs with HELP compression compared to Modified Frankel Score (MFS) matched control dogs with non‐hemorrhagic disk extrusions, evaluate the extent of spinal cord compression on MRI compared to final patient outcomes in HELP compression, and determine the compression to decompression ratio and its relation to outcomes in HELP cases.

Animals: 143 dogs from the neurology service were included: HELP compression dogs (n=65) and IVDE dogs (n=78).

Methods: This is a retrospective case‐control cross‐sectional study. Each case was assigned an MFS on admission and follow‐up. MRIs were reviewed and compression to decompression ratios were calculated based on T2 sagittal MRI images. Outcomes were assigned for each patient based on MFS at follow‐up.

Results: Outcomes of HELP and IVDE did not differ when taken to surgery when stratified by MFS. Dogs with HELP compression had more compressed sites than IVDE (P=0.001) and had more sites decompressed (P<0.001). Compression to decompression ratio did not differ between groups (P=0.52) and their outcomes were not significantly different.

Conclusions and clinical importance: When addressed surgically, dogs with HELP compression have similar outcomes to dogs with non‐hemorrhagic IVDE.

30.1.1.1. ¹Purdue University; ²Associate Professor, Neurology and Neurosurgery, VCS Courtesy Associate Professor, BMS, Veterinary Clinical Sciences, Purdue University; ³Assistant Professor, Neurology, Veterinary Clinical Sciences, Purdue University; ⁴Director, Clinical Trials and Professor of Epidemiology, Department of Veterinary Administration, Purdue University

Background: The prognosis of dogs with meningoencephalomyelitis of unknown etiology (MUE) is difficult to predict. MUE cases with normal magnetic resonance imaging (MRI) occur, but it is unknown whether this finding alters prognosis.

Hypothesis: Normal‐MRI MUE dogs survive longer than MUE dogs with MRI lesions.

Animals: 82 client‐owned dogs with MUE presenting to a university hospital from 2010–2020.

Methods: Retrospective study. Dogs with a clinical diagnosis of MUE were included if they were between 6 months to 10 years at diagnosis, had brain or spine MRI performed, had a CSF white blood cell count >5 cells/μL, were tested for infectious diseases, and had follow‐up for ≥3 months or until death. MRIs were reviewed for the presence or absence of MUE lesions. Deaths due to MUE were compared between dogs with and without MRI lesions using a Kaplan‐Meier curve. A log‐rank test was performed to evaluate statistical significance (p<0.05).

Results: Sixty dogs (73%) had MRI lesions with a median survival of >70.7 months. Twenty‐two dogs (27%) did not have MRI lesions with a median survival of >98.0 months. The death rate was 23% for dogs with MRI lesions and 9% for dogs without (p=0.152).

Conclusions: The majority of MUE dogs had prolonged survival. No statistically significant difference in death rate of dogs with and without MRI lesions was found. Additional outcome measures (e.g., remission rates) with extended follow‐up should be studied to determine whether the presence of MRI lesions influences prognosis in MUE.

31.1.1.1. ¹University of Wisconsin‐Madison; ²Clinical Assistant Professor in Small Animal Neurology, Department of Medical Sciences, University of Wisconsin‐Madison; ³Department of Biostatistics and Medical Informatics, University of Wisconsin‐Madison

Background: Cerebrovascular accidents (CVAs) in dogs are documented with magnetic resonance imaging (MRI). This imaging modality is not always available. Additionally, some CVAs may be too small to be detected or appear similar to other pathology on MRI. Thromboelastography (TEG) and D‐Dimer are utilized in human medicine for detection of CVAs but their use in veterinary patients has not been assessed.

Hypothesis/Objectives: To assess utility of TEG and D‐Dimer testing in diagnosing CVAs in dogs.

Animals: 76 client‐owned dogs with acute onset neurologic signs having a brain MRI and TEG or D‐Dimer testing performed.

Methods: Multicenter, retrospective study. The incidence of abnormal TEG or D‐Dimer results was compared between patients with clinical or MRI evidence of CVA and a control population (neurologic diagnosis other than CVA). Analysis methods included Fisher's or Chi‐square test for association.

Results: In our population, neither TEG or D‐Dimer was significantly associated with a CVA (p=0.374 and 0.425, respectively). The sensitivity of TEG for CVA diagnosis was significantly higher than D‐Dimer in all cases (60.6% and 25.0%, respectively; p=0.042) and for cases with MRI evidence alone (TEG 66.7%, D‐Dimer 23.1%, p=0.025). The specificity of abnormal TEG or D‐Dimer for CVA diagnosis was not significantly different, even in cases with MRI evidence alone (p=0.606, p=0.440). When comparing hemorrhagic and ischemic CVAs, an abnormal TEG or D‐Dimer were not significantly associated with diagnosis (p=0.168 and 0.832, respectively).

Conclusions and Clinical Importance: Abnormal TEG or D‐Dimer alone is not a reliable indicator of CVA in dogs.

32.1.1.1. ¹Medical Teaching Hospital, Texas A&M Veterinary; ²Professor, Department of Small Animal Clinical Sciences, Veterinary Medical Teaching Hospital, Texas A&M University

Background: Surprisingly, there is currently no agreed definition of ‘ambulation’ following recovery after spinal cord injury (SCI) in dogs.

Hypothesis/Objectives: To create and validate a new definition of ambulation in dogs recovering from SCI.

Animals: 206 client‐owned dogs weighing <20 kg: 110 ambulatory dogs were used to determine relationship between limb (ulna) length and distance covered by 50 step cycles; 96 non‐ambulatory dogs that underwent surgical decompression for acute thoracolumbar disc herniation.

Methods: Prospective descriptive study. Ambulatory dogs were leash‐walked and the distance covered by 50 step cycles was measured. Ulna length was measured in each dog and a regression equation relating ulna length to 50‐step cycle distance was generated. The non‐ambulatory dogs underwent decompressive surgery and were followed until they recovered to walk the distance covered by 50 step cycles. Recovery of ability to walk and the time taken to achieve this distance were compared amongst dogs that were paraparetic, paraplegic and paraplegic with absent pain sensation.

Results: 11/25 dogs without pain sensation at presentation recovered the ability to walk the 50‐step distance at 4 months, compared with 69/71 dogs with intact pain sensation. Median time to recovery was 87 days in dogs presenting without pain sensation and 14 days in both paretic and plegic dogs.

Conclusions and Clinical Importance: The 50‐step walking test provides a standard outcome that is easily applied by owners, can reliably detect differences in outcome following different severities of thoracolumbar SCI and could be used in future clinical trials of new therapies.

33.1.1.1. ¹Veterinary Hospital, North Carolina State University; ²University of Missouri; ³The Ohio State University; ⁴Cummings School of Veterinary Medicine at Tufts University; ⁵VIN, Cornell University

Background: Ante‐mortem diagnosis of degenerative myelopathy (DM) is presumptive and there are no accepted guidelines regarding management of this neurodegenerative condition.

Hypothesis/Objectives: To describe current practices of neurologists and rehabilitation specialists in diagnosis and management of DM.

Animals: None.

Methods: Online questionnaires regarding diagnosis and management of DM in dogs were constructed and emailed to neurology and rehabilitation listservs.

Results: 190 neurologists and 79 rehabilitation specialists from 20 countries completed the questionnaires. Most (142/190) neurologists and rehabilitation specialists (23/39) require SOD‐1 genetic testing for presumptive diagnosis and 82/190 neurologists also require spinal magnetic resonance imaging (MRI). Most neurologists recommend regular exercise (187/190) and physical therapy (184/190). Over 50% (102/190) of neurologists perform rechecks on dogs they diagnose with DM, while the remainder do not. Rehabilitation specialists report preservation or improvement of strength (78/79) and coordination (77/79) as therapeutic goals. They recommend at‐home exercises (75/79) as well as underwater treadmill (64/79), gait training (55/79) and strength building exercises (65/79) and report that treatment helps maintain strength (58/79), coordination (56/79), and muscle mass (56/79) with improvement in overall wellbeing (54/79). Neurologists reported that owners elect euthanasia when dogs become non‐ambulatory paraparetic while rehabilitation specialists report euthanasia when paraplegia and incontinence develop.

Conclusion and Clinical Importance: Most clinicians use SOD‐1 genetic testing to presumptively diagnose DM. While many neurologists reevaluate dogs diagnosed with DM, continued care is often referred to rehabilitation specialists or primary veterinarians. Treatment recommendations include exercise and rehabilitation. Many rehabilitation specialists report clinical stabilization and improved wellbeing with treatment.

34.1.1.1. College of Veterinary Medicine, The Ohio State University

Background: Spasticity following spinal cord injury (SCI) occurs commonly, but ways to measure this in dogs are underexplored. Dynamometry is commonly used to quantify spasticity in people.

Objectives: Evaluate a handheld dynamometer for quantifying spasticity in healthy dogs and SCI‐affected dogs.

Animals: Prospective cohort study in 11 healthy dogs and 10 with acute TL‐IVDE (SCI‐affected).

Methods: A handheld dynamometer (The Commander™ console and muscle tester, JTECH Medical) was used to perform measurements at multiple thoracic and pelvic limb joints. Healthy dogs were evaluated at two sessions in a 24‐hour period. SCI‐affected dogs were evaluated 0, 14, 30, and 60 days after surgical decompression. Measurements were compared between sessions in healthy dogs and between healthy and SCI‐affected dogs using a Wilcoxon matched‐pairs signed rank test and a Mann Whitney test, respectively. P<0.05 was considered significant.

Results: Apart from the carpi, measurements were consistent across sessions in healthy dogs. Measurements were higher in SCI‐affected dogs for the carpi at day 0 (P<0.0003); for the carpi and stifles at 14 days (P<0.015); and for the carpi, stifles, and tarsi at days 30 and 60 (P<0.03).

Conclusions: Dynamometry measurements for most joints are consistent across sessions in healthy dogs, but differ between healthy and SCI‐affected dogs for at least 60 days after TL‐IVDE. The inconsistency in carpal measurements require further investigation. Dynamometry might represent a helpful quantitative measure of spasticity of pelvic limbs after SCI in dogs and could be used in future clinical studies.

35.1.1.1. Veterinary Hospital, North Carolina State University

Background: Thoracic vertebral stabilization is challenging due to unique vertebral anatomy and proximity to critical anatomical structures.

Hypothesis/Objectives: To quantify implantation corridor parameters of canine thoracic vertebrae and distances to adjacent critical structures.

Animals: 30 client‐owned dogs with a thoracic computed tomography (CT) study, grouped by weight (n=6/group; <5 kg, 5–10 kg, 10–20 kg, 20–40 kg, >40 kg).

Methods: Retrospective morphometric study. Implantation corridor parameters of thoracic vertebrae (T1‐T13) were measured (corridor length, width, angle from midline, angle of deviation) based on ideal trajectories, and distances to important adjacent structures were quantified.

Results: Implantation corridor (390 vertebrae) widths varied between 1.8 mm (T4, <5 kg) and 12.9 mm (T1, >40 kg), lengths between 8 mm (T13, <5 kg) and 33.8 mm (T5, >40 kg), angle from midline between 3° (T11, 20–40 kg) and 56° (T1, 5–10 kg), and angle of deviation between 3° (T7, <5 kg) and 45° (T13, 20–40 kg). Distances to vital structures were: subclavian artery (n=49; 5.8–40 mm), aorta (n=310; 0.8–43.4 mm), trachea (n=173; 1.3–43.5 mm), lungs (n=362; 0.4–28.3 mm), azygous vein (n=223; 0.3–21.7 mm).

Conclusion and Clinical Importance: Thoracic vertebrae are in close proximity to vital anatomical structures and have narrow implantation corridors with steep angles of ideal trajectory, which vary considerably by vertebra and dog size. Accurate implantation is necessary to avoid serious complications from invading adjacent anatomical structures, spinal canal breech, or ineffective implant placement.

36.1.1.1. ¹Wheat Ridge Animal Hospital; ²Ethos Veterinary Health

Background: Intervertebral disc‐associated epidural hemorrhage (EH) in dogs is a poorly understood neurological condition.

Objective: To compare the signalment, clinical presentation, MRI changes, and clinical outcome of dogs with acute thoracolumbar intervertebral disc herniation (TL‐IVDH) with and without EH.

Animals: 160 client‐owned dogs that underwent MRI and hemilaminectomy for acute TL‐IVDH at a private practice in Colorado, including 63 dogs with EH and 97 dogs without EH.

Methods: Retrospective review of medical record data from 160 dogs presenting sequentially to a single practice with acute TL‐IVDH that underwent MRI and hemilaminectomy surgery.

Results: 63 of 160 (39%) dogs had confirmed EH. French Bulldogs were significantly over‐represented (23/63; OR 4.1; 95% CI: 1.8–9.0; p<0.001) of the EH cases. Dogs with EH were more likely to present with clinical signs under 48 hours than dogs without EH (24–48 hours vs 48–72 hours; OR 2.4; 95% CI: 1.2–4.6; p=0.023) and were more likely to be non‐ambulatory on presentation (OR 2.1; 95% CI: 1.0–4.1; p=0.044). MRI showed that dogs with EH were more likely to have <50% cross‐sectional spinal cord compression than dogs without EH (OR 2.3 vs. 0.4; 95% CI 1.2–4.4 and 0.2–0.9, respectively), longer longitudinal spinal cord compression (3 spaces vs 1 space, p<0.001), and greater intrinsic spinal cord change (grade 3/severe vs grade 1/mild; p<0.001). The location of the intervertebral disc herniation in French Bulldogs with EH was more likely to be thoracolumbar (OR 10.8; 95% CI: 2.1–55.7; p=0.033).

Conclusions and Clinical Importance: French Bulldogs have a high prevalence of intervertebral disc‐associated EH. Dogs with EH have a shorter clinical course, are more likely to present non‐ambulatory, and have milder compression with more severe intrinsic spinal cord changes on MRI.

37.1.1.1. ¹College of Veterinary Medicine and Biomedical Sciences, Texas A&M University; ²Postdoctoral Research Associate, Arizona Canine Cognition Center, School of Anthropology, University of Arizona; ³Professor of Biological Sciences and Anthropology, Human and Evolutionary Biology Section, Department of Biological Sciences, University of Southern California; ⁴Professor, Department of Psychology, University of Arizona; ⁵Professor, Co‐director and PI of the Dog Aging Project, Department of Laboratory Medicine and Pathology, School of Medicine, University of Washington; ⁶Director, Arizona Canine Cognition Center, School of Anthropology, University of Arizona

Background: Canine Cognitive Dysfunction (CCD) is a form of dementia that shares many similarities with Alzheimer's Disease. In humans, physical activity is believed to reduce risk of Alzheimer's Disease. We explored the association between physical activity and cognitive health in a population of companion dogs, aged 6–18 years.

Hypothesis/Objectives: We hypothesized that physical activity would be associated with improved cognitive function and decreased risk of dementia, and that this association would remain robust when controlling for age, comorbidities, and potential confounders.

Animals: Our sample included 11,574 companion dogs enrolled through the Dog Aging Project. Of these, 287 had scores over the clinical threshold for CCD.

Methods: In this observational, cross‐sectional study, we used owner‐reported questionnaire data to quantify dog cognitive health (via the Canine Cognitive Dysfunction Rating scale), physical activity levels, health conditions, training history, and dietary supplements. We fit linear regression models with measures of cognitive health as the outcome, and physical activity—with several important covariates—as predictors.

Results: We found a significant negative relationship between physical activity and current severity of cognitive symptoms (estimate =‐0.12, p<0.001), magnitude of worsening over a 6‐month interval (estimate =‐0.09, p<0.001), and whether a dog reached a clinical level of CCD (odds ratio =‐0.67, p<0.001).

Conclusions and Clinical Importance: Physical activity was robustly associated with higher cognitive functioning in dogs; to help determine causality, future studies can follow the same dogs longitudinally or administer physical‐activity based interventions.

38.1.1.1. ¹MedVet Pittsburgh; ²BluePearl Specialty and Emergency Hospitals; ³Department of Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University; ⁴Small Animal Medicine and Surgery, College of Veterinary Medicine, University of Georgia

Background: Frequency of findings in canine brain magnetic resonance imaging (MRI) after referral to neurologists has not been reported. Multiple studies have focused on radiologist interpretations without association to confirmed neurological dysfunction.

Hypothesis/Objectives: Study objectives were to determine frequency of brain MRI findings in dogs with signs of intracranial disease from two referral institutions and agreement between neurologists reviewing MRIs. We hypothesized there would be no difference in frequency of MRI findings between institutions and neoplastic lesions would be the most frequent MRI diagnosis.

Animals: A cohort of 300 randomly selected dogs undergoing brain MRI to determine underlying cause of primary neurological presentations.

Methods: Medical records of dogs from two tertiary‐care institutions were collected retrospectively from 2014–2019. Brain MRIs were reviewed blindly/independently by two ACVIM board‐certified neurologists. Lesion characteristics and suspected MRI diagnosis based on published criteria were described.

Results: An intracranial MRI abnormality was determined in 64–71% of cases. Neoplastic (38–42%) and inflammatory (26–33%) lesions were the most common differentials. Seizures were noted in 52% of cases; 31% of dogs with seizures had normal MRIs. Interobserver agreement on abnormal scans was considered very good (Κ=0.764)

Conclusions and Clinical Importance: At least 29% of dogs with intracranial disease have normal MRIs. The ability of neurologists to differentiate and agree on normal versus abnormal MRIs is very good. Neoplastic and inflammatory were the most common MRI diagnoses. This data may help guide discussion with clients on expectations before imaging, and aid in empiric therapy decisions.

39.1.1.1. ¹Department of Clinical Sciences, College of Veterinary Medicine, North Carolina State University; ²Vet Tech, Department of Clinical Sciences, North Carolina State University; ³Assistant Professor, Department of Psychology, Universita Di Bologna; ⁴NCSU; ⁵Assistant Professor, Behavioral Medicine, NCSU; ⁶Professor, Clinical Sciences, NCSU; ⁷Professor, Clinical Sciences, North Carolina State University

Background: Senior dogs undergo cognitive decline and suffer increasing pain, potentially altering activity. Additionally, this decline in quality of life could alter motivation to exercise. Changes in activity and sleep‐wake cycle have health consequences and influence the human‐animal bond but data on this topic are scarce.

Objective: Determine the relationship between activity patterns and cognition, pain, and motivation in senior dogs.

Animals: 32 senior dogs (12.48±1.50 years old) participating in the Longitudinal Study of Neuro‐Aging in Dogs were included in this study.

Methods: Weekday activity was evaluated over a 2‐week period using accelerometers. Cognitive status was assessed with the CAnine DEmentia Scale (CADES). Total pain (joint + spinal pain) was scored by physical examination. A motivation score was generated using a ratio of on‐leash/off‐leash gait speed moving towards a treat. The interaction between these parameters and activity was evaluated using functional linear modeling.

Results: Median (range) CADES and pain scores were 10 (0–70) and 8 (1–19) respectively; mean ± SD motivation score was 1.72±0.64. Dogs with lower CADES scores were more active between 5:30 p.m. and 8:00 p.m. Greater pain was associated with higher activity during sleep time (1:30–3:00 a.m.), and with lower activity in the early‐morning (4:00–5:30 a.m.). Finally, motivated dogs were more active between 10:00 a.m. to 2:00 p.m. and 5:00 p.m. to 8:00 p.m. (Figure 1).

Conclusion: Strategies to reduce cognitive dysfunction and pain and increase motivation could improve activity patterns including quieter rest at night and greater diurnal activity.

40.1.1.1. ¹University Animal Hospital, Swedish University of Agricultural Sciences; ²Department of Medicinal Chemistry, Analytical Pharmaceutical Chemistry; ³Swedish National Veterinary Institute

Background: Alpha‐chloralose (AC) is an anesthetic compound approved as a neurotoxic rodenticide in the USA since 2019. Reports on AC poisoning in companion animals are scarce. Alpha‐chloralose was previously approved in Sweden but following a surge of suspected cases of accidental AC poisoning of cats in 2019 this study was initiated, and layman use of AC was subsequently banned.

Objectives: To describe the clinical presentation of AC poisoning in cats.

Animals: Sixty‐four client‐owned, outdoor cats with suspected AC poisoning presenting to a teaching hospital in October 2019–February 2020.

Methods: Of 64 cats presenting with a strong clinical suspicion of AC poisoning, 24 cats were included in the case series, based on availability of surplus serum. A novel, quantitative UHPLC‐MS/MS analysis was developed and used to detect AC. Clinical signs described in the literature were tallied for each cat and disease severity was scored using set criteria.

Results: Alpha‐chloralose poisoning was confirmed in all 24 cases. The most common signs of intoxication were ataxia (100%), tremor (100%), cranial nerve deficits (96%), hyperesthesia (88%), bradycardia and somnolence (79% respectively). Hypothermia, previously reported in >90% of AC poisoned cats, was seen in only 38%. Disease severity differed from previous studies, with our population presenting with less severe disease, and no mortalities.

Conclusions and Clinical Importance: The present study describes the clinical presentation of AC poisoning in cats. Recognizing AC poisoning as a differential diagnosis in acute onset neurological signs may influence clinical decisions, and ultimately the outcome for the patient.

41.1.1.1. ¹Department of Veterinary Clinical Sciences, College of Veterinary Medicine, Washington State University; ²Department of Clinical Sciences, College of Veterinary Medicine, Cornell University

Background: The use of cannabidiol (CBD) has shown promise in the treatment of pain in dogs.

Hypothesis/Objectives: CBD‐rich hemp extract provided with gabapentin following surgery will result in improved analgesia and decreased opioid requirements compared to patients receiving gabapentin alone.

1. No major adverse effects will be observed with CBD rich hemp and gabapentin.

Animals: Dogs, with intact deep pain, presenting for acute intervertebral disc herniation.

Methods: Double‐blinded, randomized, placebo‐controlled study.

5 mg/kg CBD‐rich hemp provided as an equal mix of CBD and CBDA in sesame oil or placebo sesame oil were provided every 8 hrs in conjunction with oral gabapentin.

After surgery, all patients were given injectable methadone and started on the oral treatments. Patients were pain scored with a validated system every 2 hours, 24 hours, and every 4 hours from 24–48 hours. Rescue analgesia (methadone) was given if pain score was >5/20 and recorded each time.

Results: Twenty dogs completed the study without adverse events.

Average pain scores over 48 hours for the CBD group were2.9±1.1(median 2.7). Placebo average pain scores were 3.7± 4.3, median 4.3). Mann Whitney U analysis found no significant difference (p=0.054); yet a trend towards significance with CBD rich hemp was observed.

Average number of methadone rescues over 48‐hours in the CBD group was 2.8±2.5 (median 2); placebo group mean was3.66±2.45 (median 5). Mann Whitney U Test found no significant difference (p=0.535).

Conclusion and Clinical Importance: Further investigation of CBD as an adjunctive analgesic following spinal surgery in dogs is warranted.

42.1.1.1. ¹Ontario Veterinary College; ²Convergent Medical Technologies, Inc.

Background: Transcutaneous vagus nerve stimulation (tcVNS) has been used to treat epilepsy in people and dogs. Measurable changes in the electroencephalogram and heart rate variability (HRV) are reported in people using tcVNS.

Hypothesis/Objectives: Objective changes in brain activity and HRV will be detected by electroencephalography (EEG) and electrocardiography (ECG), respectively, after cervical tcVNS in dogs.

Animals: 6 client‐owned healthy dogs.

Methods: Prospective cohort feasibility study comparing the differences pre‐ and post‐tcVNS in frequency band power analysis (EEG) and HRV using Holter monitoring. tcVNS was applied at the maximal tolerable intensity; the animals’ reactions were noted.

Results: In the post‐stimulation EEG, the average power per channel was found to be significantly decreased in the theta (p=0.02) and alpha bands (p=0.04). Alpha (p<0.01), theta (p=0.01) and beta (p=0.035) frequencies post‐stimulation decreased significantly in the pooled power spectral analysis. The multivariate model did not detect any significant interaction between dog, attitude, and stimulation, neither within the same dog nor between individuals.

Holter monitoring revealed a significant increase in the HRV measured by the deviation of the inter‐beat (SDNN) index (p<0.01) and a decrease in mean heart rate (p<0.01) after tcVNS.

tcVNS was well tolerated by the animals without significant adverse effects reported.

Conclusions and Clinical Importance: This study suggests that tcVNS results in measurable changes in brain activity and HRV as detected by EEG and Holter monitoring, respectively, in dogs. Future investigations could confirm these findings in a larger cohort and objectively assess the potential clinical relevance of tcVNS.

43.1.1.1. ¹Chungbuk National University; ²CBNU

Background: Tau protein is a microtubule‐associated protein involved in the assembly and stabilization of the microtubules. The hyperphosphorylation of tau implicates microtubule stability, is demonstrated in the progression of multiple sclerosis (MS) in human medicine. MS is an autoimmune neurological disease and has many similarities with the meningoencephalitis of unknown etiology (MUE) in dogs, including pathological mechanisms.

Objectives: This study investigated the presence of hyperphosphorylated tau in dogs with MUE, and experimental autoimmune encephalomyelitis (EAE).

Animals: Total eight brain samples of the dogs were examined, including two neurological normal dogs, three dogs with MUE, and three canine EAE models.

Methods: Anti‐tau (phospho‐S396) antibody was used for immunohistochemistry (IHC), which stains hyperphosphorylated tau.

Results: In normal brain tissues, the expression of hyperphosphorylated tau was not identified. In all EAE dogs and one MUE dog, immunoreactivity for S396+ p‐tau was observed at glial cell cytoplasm and background in the periphery of the inflammatory lesion.

Conclusions and Clinical Importance: These result suggests that the first evidence that tau pathology may be involving neuroinflammation in the dogs, as in the human MS.

44.1.1.1. ¹Tierärztliche Hochschule Hannover; ²Royal Veterinary College

Background: Nausea and emesis can be associated with vestibular system dysfunctions in dogs. Up to date, antiemetic drugs, such as maropitant and metoclopramide, are commonly used as therapeutic options but fail to address nausea.

Hypothesis/Objectives: The objective of the study was to show the efficacy of ondansetron, a 5‐HT receptor antagonist, against nausea in dogs with an acute vestibular syndrome.

Animals: Fourteen client‐owned dogs presented at the neurology service with clinical signs of a vestibular syndrome and nausea were included in the study.

Methods: A placebo‐controlled, double‐blinded, randomized, crossover‐study was designed to observe nausea before and after the administration of ondansetron. Dogs either received ondansetron or placebo (group 1). After the last behavioural assessment (2 hours later) dogs received then placebo or ondansetron respectively (group 2). The degree of nausea was scored with a former validated nausea score. In addition, arginine‐vasopressin‐serum levels, a biomarker of nausea, were determined.

Results: A reduction of the nausea score in the behavioral assessment was observed in all dogs. In group 1, the average score decreased significantly from 13.625 to 1.68 two hours after administration of ondansetron (p<0,0001) and in group 2 from 13.17 to 2.08 (p<0,001). The arginine‐vasopressin‐level decreased significantly four hours after administration (p<0,05).

Conclusion and Clinical Importance: Ondansetron showed a clinical resolution of nausea one hour after administration, while the arginine‐vasopressin level decreased four hours after administration. Ondansetron is likely beneficial for improving vestibular nausea in dogs.

45.1.1.1. ¹Tierärztliche Hochschule Hannover; ²University of Haifa

Background: Objective pain assessment in cats is not routinely performed in veterinary practice. Many veterinarians lack the skill to objectively assess and, therefore, do not provide an adequate level of analgesia to painful cats. Automated tools based on Artificial Intelligence (AI) could provide a promising avenue to overcome these issues.

Objective: Our goal is to investigate, evaluate, and develop methods for automated facial analysis for pain assessment in cats that can be integrated in clinician's workflow.

Methods: An AI model based on Dino‐Resnet50 backbone transfer learning was trained on image dataset with 68 cats with pain and 68 cats without pain. All cats were scored using a validated pain score for cats.

We evaluated the model using a 10‐fold cross validation with 20% of train set for validation. Model was trained during 100 EPOCs with Adam optimizer and learning rate 0.0005.

Results: Model prediction reached accuracy of 0.64, precision of 0.64, and recall of 0.70.

Conclusion and Clinical Importance: The deep learning model shows promising preliminary results for binary classification of pain/no pain in cats based on images. To increase accuracy, more quality data is planned to be collected. More sophisticated pre‐processing methods, such as centering the cat's face, will be developed.

Assessing pain via an AI could clinically be beneficial as it guarantees objective pain scored independently of the examiner. This could lead to easier adjustments of the analgesic therapy to the patient's needs.

46.1.1.1. Laboratory of Veterinary Internal Medicine, College of Veterinary Medicine, Chungbuk National University

Background: The neutrophil‐to‐lymphocyte ratio (NLR) has been identified as a biomarker in several inflammatory and autoimmune diseases. It has been investigated that multiple sclerosis is associated with NLR in human patients.

Hypothesis/Objectives: To examine the diagnostic and prognostic value of the NLR in meningoencephalitis of unknown etiology (MUE) dogs.

Animals: Thirty‐eight MUE dogs, 20 hydrocephalus dogs, 10 brain tumor (BT) dogs, 32 idiopathic epilepsy (IE) dogs, and 41 healthy dogs.

Methods: This study was performed as case control study. Medical records were reviewed to identify dogs diagnosed with neurologic disease.

Results: The NLR was significantly higher in MUE dogs (median [interquartile range] = 6.08 [3.48–8.84]) than in healthy dogs (1.78 [1.36–2.20]; P<0.001). The NLR of MUE dogs was higher than that of IE (2.50 [1.58–3.43]; P<0.001) and hydrocephalus dogs (1.79 [1.32–2.83]; P<0.001), but not differed with that of BT dogs (4.27 [2.85–4.86]; P>0.99). The NLR of MUE dogs was significantly associated with brain lesion volume (P=0.02, r=0.37), but not with the survival time (P=0.31, r=0.17).

Conclusions and Clinical Importance: The NLR could be a biomarker for diagnosing MUE and distinguishing it from other intracranial diseases such as IE and hydrocephalus.

47.1.1.1. ¹Department of Medical Sciences, School of Veterinary Medicine, University of Wisconsin‐Madison; ²Clinical Assistant Professor, Department of Medical Sciences, Madison School of Veterinary Medicine, University of Wisconsin

Background: Trazodone is an oral anxiolytic used to reduce anxiety in dogs. Whether trazodone affects the neurologic examination in dogs has not been previously reported.

Objectives: To investigate whether trazodone administration is associated with changes in the neurologic examination in healthy dogs.

Animals: 32 healthy dogs between 1 and 6 years old, perceived by their owners as neurologically normal, with no previously diagnosed medical conditions.

Methods: Baseline sedation and anxiety assessments and neurologic examination were performed for each dog, followed by trazodone administration (6.25–8.60 mg/kg PO). The sedation and anxiety assessments and neurologic examination were repeated 1–2 hours after trazodone administration. The examinations were performed by a single board‐certified veterinary neurologist and were video‐recorded. The videos were randomized and reviewed by a different, masked neurologist who scored the exams.

Results: Seven of 32 (22%) dogs experienced worsening scores on neurologic exam after receiving trazodone, manifesting as new or progressive postural reaction deficits. Although not clinically significant, 18.75% of dogs had consciousness levels changed from bright, alert, responsive to quiet, alert, responsive after trazodone administration. There were no other changes present on neurologic exam. Sedation and anxiety scores were significantly changed post‐trazodone compared to pre‐trazodone (p=0.0001 and p=0.0001, respectively).

Conclusions: Most dogs did not have changes on neurologic examination after trazodone administration. However, approximately 1 out of 5 dogs have new or worsening postural reaction deficits after receiving trazodone. Ideally, trazodone should not be given prior to neurologic examination in dogs.

48.1.1.1. ¹North Carolina State University; ²Duke University

Background: Neurofilament light chain (NF‐L) is a neuronal structural protein that is released into the cerebrospinal fluid (CSF) and blood with neuronal injury and might be a useful biomarker of CNS disease in dogs.

Hypotheses: NF‐L concentrations are higher in dogs with meningoencephalomyelitis than in healthy dogs and are associated with clinical outcome.

Animals: 36 dogs with meningoencephalomyelitis from a hospital population and 24 healthy control dogs.

Methods: Retrospective case‐control study. Serum and CSF samples were obtained from a biobank. An ultra‐sensitive immunoassay was used to determine NF‐L concentrations. Survival times and information regarding neurological outcomes were obtained from the medical records.

Results: NF‐L concentrations were significantly higher in meningoencephalomyelitis dogs than in healthy controls for both serum (median 94.7 vs. 10.6 pg/ml; P<0.0001) and CSF (median 7,178.0 vs. 268.1 pg/ml; P<0.0001). However, matched serum and CSF concentrations in meningoencephalomyelitis dogs were not well correlated (Spearman's r=0.49, p=0.0589). We found no differences in CSF or serum concentrations between dogs that did or did not survive to hospital discharge (P=0.12 and 0.44 respectively) or between those surviving greater than or less than 3 months (P=0.31 and 0.66 respectively). Finally, we found no differences in CSF or serum concentrations between dogs that did or did not have persistent neurological deficits after treatment (P=0.31 and 0.99 respectively).

Conclusion: Dogs with meningoencephalomyelitis have markedly elevated CSF and blood concentrations of NF‐L compared to healthy controls. NF‐L was not associated with outcomes, although further study is required to adequately answer these questions.

50.1.1.1. ¹Texas A&M University; ²Clinical Associate Professor, Small Animal Clinical Sciences, Texas A&M University; ³Professor, Texas A&M University; ⁴Professor, Small Animal Clinical Sciences, Texas A&M University; ⁵Associate Professor, Department of Veterinary Pathobiology, Texas A&M University

Background: Fasting is often recommended prior to the collection of blood for routine biochemical analysis despite a paucity of evidence to support this requirement.

Objectives: To compare measurements of selected biochemical analytes collected before and after feeding in clinically healthy dogs.

Animals: 100 clinically healthy dogs weighing >15 kg.

Methods: Prospective observational study. Food was withheld for at least 10 hours. Preprandial serum was collected, and then dogs were fed their usual food at an amount equivalent to at least 2/3 resting energy requirement. Serum was collected at 2‐, 4‐, 6‐, and 8‐hours postprandially. The proportion of postprandial values that exceeded either the reported total allowable error (TEa), or for SDMA, the reference change value (RCV), was determined.

Results: The proportion of dogs with postprandial measurements that exceeded the TEa or RCV was 92/100 for triglycerides, 66/100 for BUN, 17/100 for glucose, 46/100 for phosphorus, 9/100 for bilirubin, 5/100 for SDMA, 2/100 for creatinine, and 0/100 for cholesterol and albumin. Of dogs with a postprandial result above the upper end of the reference interval, the proportion that had a fasted measurement above the reference interval was 3/36 for triglycerides, 7/17 for SDMA, 2/3 for creatinine, 8/10 for cholesterol, and 0/6 for glucose. Albumin and BUN never exceeded the reference interval, nor did postprandial lipase in dogs with normal fasted lipase.

Conclusions and Clinical Importance: Certain biochemical analytes may be impacted by feeding in healthy dogs. The impact in patients with illness requires investigation.

51.1.1.1. ¹The Ohio State University; ²Colorado Animal Specialty & Emergency

Background: Several alternative tests to measure 25‐hydroxyvitamin D (25D) concentrations are commercially available, but they have not been assessed for agreement to standard methodology in cats.

Hypothesis/Objective: To compare three tests (two dried blood spot[DBS]tests and one lateral flow assay)to the gold‐standard measurement of serum 25D via liquid chromatography/tandem mass spectrometry: LC/MS/MS. We hypothesized that these tests would show good agreement with LC/MS/MS.

Animals: Six healthy purpose‐bred 2‐year‐old research cats (3 male neutered, 3 spayed female).

Methods: This was a prospective observational pilot study.Blood was collected at 6 timepoints over 6 weeks.Using two DBS tests and a lateral flow assay, whole blood 25D was measured and compared to serum 25D concentrations via LC/MS/MS. Agreement between methods was evaluated via Bland Altman analysis.The predetermined criterion for clinically significant variance in results was set at a difference of +/‐ 10 ng/mL or greater between the test being compared to the reference method.

Results: Both DBS tests and the lateral flow assay demonstrated a mean difference greater than the pre‐determined criterion for clinical significance.

Bland‐Altman analyses:Difference in 25D concentrations between methods.

Conclusions and Clinical Importance: All three tests demonstrated poor agreement with serum LC/MS/MS concentrations.These novel methods cannot be recommended as an alternative to LC/MS/MS.

52.1.1.1. ¹University of Wisconsin‐Madison Veterinary Care; ²University Distinguished Professor (with Tenure), Small Animal Internal Medicine Dr. Mark Morris Chair in Small Animal Gastroenterology & Nutrition Director, Gastrointestinal Laboratory (GI Lab), Department of Small Animal Clinical Sciences, Texas A&M University; ³Melita Grunow Family Professor in Companion Animal Health Professor and Assistant Dean for Clinical and Translational Research, Department of Medical Sciences, University of Wisconsin‐Madison Veterinary Care

Background: Greyhounds have reported hyperhomocysteinemia (HHC) relative to other dog breeds. However, the underlying mechanisms for HHC in greyhounds, and its clinical implications, are not understood.

Hypothesis/Objectives

• 2

  To determine whether HHC in clinically healthy greyhounds is associated with increased oxidative stress, as measured by plasma isoprostanes, compared to healthy non‐sighthound dogs.

• 3

  To assess whether plasma homocysteine in healthy greyhounds correlates with low serum folate or cobalamin, which are possible mechanistic drivers of HHC.

• 4

  To evaluate whether healthy greyhounds with HHC have decreased plasma concentrations of methionine, cysteine, or glutathione, which would support specific pathway defects.

Animals: 31 healthy, client‐owned greyhound dogs and 15 healthy, client‐owned non‐sighthound control dogs.

Methods: Prospective case‐control observational study. All dogs were screened with a physical exam, CBC, and chemistry profile. Homocysteine, folate, cobalamin, methionine, cysteine, reduced glutathione (GSH), and 8‐isoprostane‐F2 concentrations were measured in all dogs.

Results: Cysteine concentrations between greyhound (median 6.7 uM, range 1.9–14.2) and non‐sighthound dogs (7.9 uM, range 3.5–9.3) were not significantly different (P=0.36). Plasma GSH concentrations between greyhound (3.0 uM, range 0.6–7.1) and non‐sighthound dogs (2.6 uM, range 0.4–3.8) were also not significantly different (P=0.12). Data on folate, cobalamin, homocysteine, methionine, and 8‐isoprostane‐F2 concentrations are underway.

Conclusions and Clinical Importance: Greyhounds do not have significant differences in plasma cysteine and GSH concentrations compared to non‐sighthound dogs, suggesting that greyhounds do not have a defect in conversion of homocysteine to cysteine or GSH. Pending data will help elucidate other pathways of HHC and a possible association with oxidative stress.

53.1.1.1. ¹University of Guelph; ²Assistant Professor, Clinical Studies, Ontario Veterinary College; ³PhD Candidate, Clinical Studies, University of Guelph; ⁴Associate Professor, Clinical Studies, Ontario Veterinary College

Background: Canine fecal microbiota profiling provides insight into host health and disease. Standardization of methods for handling, collection, storage, and analysis of fecal samples for microbiomics is currently inconclusive, however.

Hypothesis/Objectives: To investigate the effects of storage temperature and addition of RNAlater on the microbiota profile of canine fecal samples. We hypothesized that storage temperature and the addition of RNAlater would alter the alpha diversity and relative abundances of the microbial population.

Animals: Four intact female and two intact male healthy colony beagles, 1 year of age.

Methods: Fecal samples were homogenized and aliquoted within 15 minutes of voiding. Aliquots were subjected to four treatments: fridge (4°C), fridge + RNAlater, freezer (‐20°C), and freezer + RNAlater. DNA extraction, PCR amplification, then sequencing (Illumina MiSeq) were completed on all samples. Inverse Simpon, Shannon evenness, and Chao1 indexes calculated alpha diversity. Wilcoxon rank sum tests compared these indexes and relative abundances between groups.

Results: Samples with RNAlater added had higher relative abundance of 5 genera (Prevotella, Fusobacterium, Phocaeicola, Bacteroidales, Anaerobiospirillum) and lower relative abundance in 4 genera (Peptacetobacter, Steptococcus, Romboutsia, Faecalibacterium) (P<0.05). No differences in relative abundance were detected between fridge and freezer samples, and alpha diversity did not differ between any treatments.

Conclusions: The addition of RNAlater to fresh samples affect the relative abundance of some taxa present in the canine fecal microbiota.

54.1.1.1. ¹The J.M. Smucker Co.; ²Consultant, Whole Pet Provisions

Background: Choline, an essential nutrient for cats, is a precursor of phosphatidylcholine, acetylcholine, and betaine. The Association of American Feed Control Officials (AAFCO) and the National Research Council (NRC) dietary recommendations for adult cats’ are 2,400 mg/kg and 2,550 mg/kg choline dry matter (DM) respectively without suggested maximums. The European Food Safety Authority (ESFA) previously deemed it virtually non‐toxic.

Objective: This case series describes the signalment, presentations, laboratory changes and outcomes of cats with choline chloride toxicity.

Animals: Medical records from 19 adult pet cats with confirmed exposure to recalled canned cat food.

Methods: Record reviews from January 1, 2020–January 15, 2021, report signalment, examinations, treatment, and outcomes. Data analysis was performed with JMP®, Version 16. SAS Institute Inc., Cary, NC, 1989–2021.

Results: Signalment: 9 neutered males, 7 ovariohysterectomized females, 3 unreported; weight (mean; SD): 5.0; ±1.5 kg; ages 1–12 years old.Observations: tachycardia (100%), vomiting (95%), ataxia (58%), fishy odor (47%), tachypnea (42%), ptyalism (42%), inappetence (26%), lethargy (21%), diarrhea (21%), abdominal pain (21%), tremors (10%) and mydriasis (5%). Onset occurred 2–96 hours post‐exposure; duration 12 hours to 5 days (excluding death/euthanized). Biochemical abnormalities: table 1. Radiographs: (n=4) gas‐distended intestines. Food analysis: 66,157–164,010 mg/kg choline DM.

Treatments: supportive IV fluids, heat, oxygen, antiemetics, hepato‐protectants, antibiotics, neurotransmitter inhibitors, muscle relaxants and enemas.

Outcomes: 15/19 recovered within 5 days. 4 died/euthanized within 24 hrs.

Conclusions and Clinical Importance: Recognizing choline and probable acetylcholine toxicity may improve treatment and outcomes. Choline's safe limit <66,000 ppm DM.

55.1.1.1. ¹College of Veterinary Medicine, Oklahoma State University; ²Associate Professor, Department of Animal and Food Sciences, Oklahoma State University; ³Assistant Professor, Department of Statistics, Oklahoma State University; ⁴Department Chair and Professor, Large Animal Clinical Sciences, College of Veterinary Medicine, University of Florida

Background: Non‐traditional dog foods (raw meat‐based diets (RMBD), dehydrated, freeze‐dried) are becoming popular feeding choices. Limited data exists characterizing the impacts on gastrointestinal health and systemic inflammation in dogs fed a non‐traditional diet versus a traditional kibble diet (KD).

Hypothesis/Objectives: Serum and inflammatory markers were evaluated in dogs fed either KD or RMBD for >1 year. We hypothesized that RMBD fed dogs would have an improved intestinal and systemic inflammatory profile.

Animals: 55 healthy, adult client‐owned dogs fed either KD (n=27) or RMBD (n=28) for >1 year.

Methods: Dogs were grouped according to current feeding method (raw or kibble). Each group received a standardized diet (single brand of kibble or RMBD, with single ingredient snacks) for an additional 30 days following inclusion. Serum and feces were collected on day 30. Serum IgG, CRP, galactin‐3, soluble receptor for advanced end glycation end products (sRAGE), and fecal intestinal alkaline phosphatase (IAP), IgA and IgG were evaluated. Linear regression and correlation analysis were used to statistically analyze the data and log‐transformation was used when necessary.

Results: Fecal IAP, IgA and IgG levels were significantly higher in the RMBD group compared to the KF group (p<0.0004, p<0.0006, p<0.0002, respectively). No difference was observed in serum galactin‐3, IgG, CRP or sRAGE between groups.

Conclusions and Clinical Importance: Dogs fed RMBD for 1 year have increased fecal IAP, IgA and IgG which may reflect improved gastrointestinal health and local immune function.

56.1.1.1. University of Montreal

Background: Feeding raw food diets is popular among dog and cat owners. In Quebec, Canada, small animal raw food industry has not been explored.

Hypothesis/Objectives: This study was performed to draw a portrait of the small animal raw food industry in Quebec, Canada.

Animals: No animals were used in this study.

Methods: All the companies producing raw food products for dogs and cats registered in Quebec were contacted by phone or email to answer the following questions: which species (dog or cat) and which life stage the raw food diets are for? Is the product formulated to meet the nutritional levels established by the AAFCO (dog/cat)? Who is responsible for the diet formulation? Has the product undergone a feeding trial? Are bacteriological tests carried out on a regular basis? How many employees are working in the company? Does the company export its products abroad?

Results: 23 companies producing raw food diets for small animals are registered in Quebec. Among them 18 answered the questions. Most companies produce raw food for dogs and cats (14/18), most companies follow AAFCO guidelines (14/18), “All life stages” was the most represented (12/18), agronomists were often responsible for product formulation (11/18), the number of employees varies from 2 to 70, a few companies export outside of Quebec (6/14), feeding trials are rarely done (2/16), and bacteriological controls are often performed (10/13).

Conclusions and Clinical Importance: These results contribute to a better understanding of the small animal raw food industry in Quebec, Canada.

57.1.1.1. ¹Texas A&M University; ²Small Animal Clinical Sciences, Texas A&M University; ³Epidemiology, Family Medicine, University of Washington; ⁴Population Health Sciences, Virginia Tech; ⁵Biostatistics, University of Washington

Background: Canine obesity is a complex multifactorial disease. While recent data suggest that eating behavior in dogs influences the development of obesity, little is known about the particular factors that may affect a dog's eating behavior.

Objectives: To evaluate the impact of dog food motivation and owner food management, as scored by the Dog Obesity Risk Assessment (DORA) questionnaire, on canine obesity, and to investigate the impact of demographic and environmental factors on dog food motivation.

Animals: USA dogs (n=410) whose owners elected to participate in the Dog Aging Project.

Methods: Owners (n=410) completed the DORA questionnaire and assigned their dog a body status (overweight or not overweight).

Results: Owner‐assigned overweight body status was positively associated with owner management score (OMS) (p<0.001) and food motivation score (FMS) (p=0.016) with an odds ratio of 1.05 (95%CI: 1.027–1.074) and 1.02 (95%CI: 1.003–1.03), respectively. Dog FMS varied among breed groups (p<0.001) with the highest average FMS in the Sporting group. None of the other evaluated demographic or environmental factors had a significant association with FMS.

Conclusions and Clinical Importance: Owners of overweight dogs have heightened feeding management practices, which contradicts the supposition that owner laxity in feeding practices contributes to canine obesity. Certain breed groups have an innately higher food motivation. Since food motivation is associated with overweight body status, one factor contributing to obesity development in these dogs may be their heightened drive to eat.

58.1.1.1. ¹Nutramax Laboratories; ²Clinical Research Director, Nutramax Laboratories; ³Business and Scientific Assessment Manager, Nutramax Laboratories; ⁴Clinical Research Manager, Nutramax Laboratories; ⁵Scientific Technical Writer, Nutramax Laboratories

Background: Phosphate is an essential mineral for cats. However, excess can distress renal function in cats with CKD. Phosphate binders have shown reductions in renal phosphate load.

Hypothesis/Objectives: This study was a proof‐of‐concept assessing the ability of a novel mineral‐based (iron and calcium) phosphate binder to bind phosphate through gastrointestinal absorption in healthy cats.

Animals: This study was comprised of a laboratory population of 21 healthy, neutered male cats.

Methods: Cats were randomly distributed into three groups and fed standard diet twice daily for 29‐days.Two groups received 1 (Group 1) or 2 (Group 3) phosphate binder capsules with each meal, and an additional group (Group 2) served as a control. Blood was drawn on days ‐5, 6, 13, 20, and 27. Feces and urine were pooled for analysis every 72 hours beginning on day 6. Two‐way repeated‐measures ANOVA was performed. All relevant blood, urine, and fecal parameters were grouped by cat, dosing group, and time factors.

Results: Urinalysis showed significant reductions of phosphate in groups 1 and 3 (p=0.020; Figure 1) compared to pre‐study levels. There were no significant changes in CBC and chemistry measures. Food amounts and body weight were measured throughout. There were no significant differences in the cats’ body‐weight or food consumed, indicating acceptance of material with food.

Pooled urine showed a significant decrease (p=0.020) in phosphorus for Groups 1 and 3 in weeks 1 and 4 when compared to the pre‐study

Conclusions and Clinical Importance: The study demonstrated the safety of the phosphate binder even at larger amounts. Nevertheless, the study supports, one capsule with a meal is a recommended starting point for administering to cats with CKD.

59.1.1.1. ¹Hill's Pet Nutrition; ²Associate Scientist, Hill's Pet Nutrition; ³Principal Scientist, Hill's Pet Nutrition; ⁴Manager, Life Science Laboratory, Hill's Pet Nutrition

Background: Companion dogs are susceptible to chronic gastroenteritis (CGE) and inflammation drives its pathogenesis. Restriction of dietary digestible carbohydrate increases ketone production which has a direct anti‐inflammatory effect.

Hypothesis: A diet low in digestible carbohydrate (LoCHO) will reduce inflammatory markers in dogs with CGE.

Animals: Canines with CGE (n=18; diagnosed by histopathological assessment) and healthy controls matched for age, gender, and weight (n=19) were housed in Hills’ animal colony.

Methods: The study was approved by the institutional animal care and use committee (protocol CP864.0). Healthy and CGE dogs participated in a crossover study, being fed high carbohydrate (HiCHO) then LoCHO foods for 5 weeks each in random order. Macronutrients (P/F/C; % energy): HiCHO (26/21/43); LoCHO (26/62/4). Blood was collected at the end of each feeding phase. Gene expression was assessed by NanoString nCounter and data analyzed using JMP Pro and Ingenuity Pathway Analysis.

Results: Feeding LoCHO food decreased inflammatory gene expression compared to HiCHO, including pathways responsible for phagocytosis, recruitment, and cellular infiltration (p=5.7E 03). LoCHO also inhibited immune cell trafficking. Food and health status interacted to influence the expression of 95 genes. Thus, JAM3, a tight junction protein, increased in CGE dogs when fed a LoCHO diet but did not in healthy dogs.

Conclusions and Clinical Importance: A reduction in inflammatory gene expression, including that for immune cell trafficking, in CGE dogs fed LoCHO is indicative that reduction in dietary carbohydrate may provide clinical benefit to CGE and could improve quality of life.

60.1.1.1. ¹School of Veterinary Medicine and Animal Science of University of Sao Paulo; ²Anclivepa's Veterinary College; ³School of Veterinary Medicine and Animal Science, University of Sao Paulo/Nutricare Vet, São Paulo, Brazil; ⁴Nutricare Vet

Background: The outbreak of COVID‐19 has impacted the whole world.

Objectives: To evaluate the implications of pandemic on feeding habits of dogs.

Methods: An online questionnaire was completed by dog owners of São Paulo, Brazil. Data were analyzed by descriptive statistics and McNemar test was used to compare profiles before and during the pandemic. Associations between variables were verified using chi‐square test.

Participants: 1188 dog owners.

Results: Weight gain occurred in 19.4% and weight loss in 5.5% of dogs. Physical activity decreased for 34.8% and increased for 14.3% of dogs (p<0.001). Daily food intake increased in 13.4% of dogs and decreased in 4.8%. 8% of owners changed their dog's diet type to reduce costs and 13.1% to make dogs lose weight. The number of animals that were fed homemade food defined by owner increased by 65.6% (p<0.001) and prescribed by a professional reduced by 43.5% (p<0.001). The number of owners who provided treats several times a day increased by 34.6% (p=0.004). Dogs ate more foods intended for human consumption, such as cookies (20% increase) and cake or bread (18% increase). There was an association between the health score attributed by owners to their own routine and to that of dogs before and during the pandemic (p<0.001).

Conclusion: There were changes in the feeding management of dogs during the pandemic, probably reflecting changes in the owners’ dietary profiles, which may have caused damage to pet's health. Dog's lifestyle is related to the environment and it is influenced by the historical context.

62.1.1.1. ¹Carlson College of Veterinary Medicine, Oregon State University; ²College of Veterinary Medicine, Texas A&M University

Background: The standard of care for treatment‐naive canine lymphoma is a multi‐agent protocol consisting of prednisone, vincristine, cyclophosphamide and doxorubicin. However, most patients relapse and combination chemotherapy can be an effective option for treating resistant lymphoma.

Hypothesis/Objectives: We hypothesized the combination of melphalan, vincristine and cytarabine (MOC) would be well tolerated and an effective treatment for dogs with relapsed or refractory lymphoma.

Animals: Twenty‐three dogs met the inclusion criteria for response and toxicity assessment. Two dogs were excluded from progression‐free survival (PFS) analysis. Both dogs had experienced clinical benefit after one cycle and were switched to a different protocol due gastrointestinal adverse events and owner preference.

Methods: Data was collected retrospectively. Dogs were treated with cytarabine and vincristine on day one. On day seven, dogs were treated with melphalan. This two‐week protocol was repeated for at least three cycles or until treatment failure.

Results: The overall response rate was 42%. The median PFS was 33 days. The overall clinical benefit was 74% for a median of 36 days. The majority of adverse events were mild and were hematologic or gastrointestinal in nature.

Conclusions and Clinical Importance: The combination of melphalan, vincristine and cytarabine is a safe treatment option for relapsed or refractory lymphoma in dogs. Additional studies are warranted to validate the potential benefit of the MOC protocol in a larger cohort and explore optimal dosing.

63.1.1.1. ¹College of Veterinary Medicine, Iowa State University; ²Associate Professor, Department of Veterinary Clinical Sciences, Iowa State University; ³Assistant Professor, Department of Veterinary Clinical Sciences, College of Veterinary Medicine, Iowa State University; ⁴Associate Professor, Hospital Director, Department of Clinical Sciences, College of Veterinary Medicine, Colorado State University; ⁵Associate Veterinarian, Abraham's Equine Clinic; ⁶Associate Veterinarian, Department of Veterinary Clinical Sciences, College of Veterinary Medicine, University of Georgia; ⁷Associate Professor, Department of Veterinary Pathology, College of Veterinary Medicine, Iowa State University

Background: Multicentric canine lymphoma is frequently treated with the modified Madison‐Wisconsin CHOP protocol. Serial CBCs are performed to monitor patients for lymphoma‐associated disease and adverse chemotherapy‐associated hematologic effects. While certain chemotherapeutics are known to induce morphologic alterations in leukocytes and erythrocytes, chemotherapy‐induced abnormalities that may occur in dogs receiving CHOP have not been described. Identifying the leukocyte and erythrocyte cytomorphologic changes by lymphoma dogs throughout CHOP may help differentiate expected changes from adverse chemotherapy side‐effects or disease progression.

Objectives: The objectives of our study were to 1) assess differences in leukocyte and erythrocyte morphology between healthy matched‐control dogs and lymphoma dogs at diagnosis and 2) evaluate changes in cellular cytomorphology as lymphoma dogs progress throughout CHOP.

Animals: 21 dogs with lymphoma and 21 healthy, age‐breed matched control dogs from a university hospital population.

Methods: Retrospective descriptive study. Quantitative and qualitative leukocyte and erythrocyte data was collected from serial CBCs from dogs with lymphoma that completed 26‐week CHOP and matched controls. Differences in leukocyte and erythrocyte metrics at diagnosis, along with differences during CHOP therapy, were compared.

Results: No differences in leukocyte or erythrocyte cytomorphologic abnormalities were found between controls and lymphoma dogs at diagnosis. Changes in RBC count (p<0.0001), hemoglobin (p=0.00031), mean cell volume (p=0.0005), red blood cell distribution width (p<0.0001), and magnitude of polychromasia (p=0.0229) were found throughout CHOP therapy.

Conclusion: Our data suggests that RBC metrics and cytomorphologic changes consistent with regenerative erythropoiesis develop in lymphoma dogs being treated with CHOP.

64.1.1.1. ¹PetDx; ²Veterinary Specialty Hospital of San Diego; ³Ontario Veterinary College, University of Guelph; ⁴Bridge Animal Referral Center; ⁵College of Veterinary Medicine, University of Minnesota; ⁶Veterinary Specialty Hospital of San Diego; ⁷Veterinary Specialty Hospital of North County

Background: Guidelines‐driven screening protocols for early cancer detection in dogs are lacking, and cancer is often detected at advanced stages.

Hypothesis/Objectives: To examine how cancer is typically detected in dogs, and whether liquid biopsy could identify genomic abnormalities in patients with incidentally detected cancers.

Animals: Client‐owned dogs with definitive cancer diagnoses, enrolled in a clinical validation study for a novel blood‐based liquid biopsy test.

Methods: Medical records were reviewed for history and presenting complaint that led to definitive cancer diagnosis. Blood samples from these dogs were subjected to DNA extraction, proprietary library preparation, and next‐generation sequencing. Sequencing data were analyzed using an internally developed bioinformatics pipeline to detect genomic alterations associated with the presence of cancer.

Results: In a cohort of 358 cancer‐diagnosed subjects, 3% were detected during a wellness exam, 8% during other routine care (including pre‐dental evaluations and grooming), and 89% of cases were diagnosed after the owner reported clinical signs suggestive of cancer. Liquid biopsy results were available for at least 35 of the 41 subjects in which cancer was detected incidentally during a wellness exam or other routine care. The test returned a Cancer Signal Detected (positive) result in approximately 55% of subjects.

Conclusions and Clinical Importance: The vast majority of canine cancer is diagnosed when the patient is presented for evaluation following the onset of clinical signs. In the majority of patients with incidentally detected cancers, liquid biopsy delivered positive results, demonstrating its potential to serve as a novel cancer screening tool for dogs.

65.1.1.1. ¹College of Veterinary Medicine, The Ohio State University; ²MedVet Cincinnati; ³School of Veterinary Medicine, University of Pennsylvania; ⁴School of Veterinary Medicine, University of California; ⁵College of Veterinary Medicine, University of Minnesota; ⁶College of Veterinary Medicine and Biomedical Sciences, Colorado State University; ⁷College of Veterinary Medicine, Texas A&M University; ⁸School of Veterinary Medicine, University of Wisconsin‐Madison

Background: Feline renal carcinomas (RC) are uncommonly encountered. Limited information regarding clinical presentation, post‐surgical outcomes, and survival times are available.

Hypothesis/Objectives: The purpose of this multi‐institutional, retrospective study was to describe the presenting features and clinical outcomes of cats with RC treated with nephrectomy.

Animals: Thirty‐six client‐owned cats were included.

Methods: Medical records from participating institutions were searched to identify cats that underwent nephrectomy and had a histopathologic diagnosis of RC.

Results: The most common presenting complaints were weight loss (n=13) and hyporexia (n=11). Twenty‐eight cats survived to discharge (77.8%). Median progression free interval (PFI) could not be determined, as only 6 cats had suspected recurrence (16.7%) and 7 cats had suspected metastasis (19.4%). The median survival time (MST) for all cats was 203 days (95% CI: 84, 1379 days). When cases that died prior to discharge were excluded, MST increased to 1217 days (95% CI: 127, 1641 days). Preoperative erythrocytosis had a protective effect, as these patients had significantly longer MST (p=0.006). Pre‐surgical azotemia, renal tumor size, mitotic index, and tumor histologic subtype were not statistically prognostic.

Conclusion and Clinical Importance: For cats surviving to discharge, prolonged survival times were possible. Due to varied post‐operative treatment approaches, conclusions regarding the impact of chemotherapy could not be made. Further studies are necessary to elucidate other potential prognostic factors, the utility of adjuvant treatment, and to identify patients at‐risk in the perioperative period.

66.1.1.1. ¹National Taiwan University; ²Department of Veterinary Medicine, National Taiwan University; ³Uni Pharma Co., Ltd

Background: Fibrinogen degradation product (FDP, DR‐70) has been used as a biomarker for cancer progression in humans for decades; however, few studies describing the clinical values of DR‐70 are available in veterinary medicine.

Animals: The purpose of this study is to investigate the correlation between DR‐70 values in the plasma and the development of various types of canine neoplasms. The plasma samples were collected from both normal (n=53) and tumor dogs (n=262) in National Taiwan University Veterinary Hospital.

Methods: The plasma DR‐70 concentrations were determined through a commercial DR‐70 enzyme‐linked immunosorbent assay (ELISA) assay.

Results: Expression levels of DR‐70 were significantly higher in cancer dogs (n=262, 2.140 μg/ml ± 0.6388 μg/ml) than those in tumor‐free individuals (n=53, 1.233 μg/ml ± 0.5426 μg/ml, p<0.0001). The area under curve of DR‐70 is 0.9089 under a 95% confidence interval from 0.8730 to 0.9466 with significant difference (p<0.0001). Furthermore, DR‐70 levels were positively correlated to the tumor progression whereas the values decreased with the responses to the treatment in numerous cases. These findings suggested that the canine plasma DR‐70 levels could be used as a biomarker for cancer patients. In conclusion, this study presents evidence that plasma DR‐70 levels are associated with cancer formation, therefore the increased levels of DR‐70 could be an effective cancer biomarker or canine tumor detection. This is the first study to show the clinical significance of DR70 in veterinary cancer medicine.

67.1.1.1. PetDx

Background: Lymphoma is a common and aggressive cancer in dogs.

Hypothesis/Objectives: To assess the performance of next‐generation sequencing‐based liquid biopsy testing in a cohort of lymphoma‐diagnosed dogs, with sub‐analyses by phenotype, stage, and substage.

Animals: Client‐owned dogs enrolled in a clinical validation study for a novel blood‐based liquid biopsy test.

Methods: Blood samples from 99 dogs with confirmed diagnoses of lymphoma were subjected to DNA extraction, proprietary library preparation, and next‐generation sequencing. Sequencing data were analyzed using an internally developed bioinformatics pipeline to detect genomic alterations associated with the presence of cancer.

Results: Liquid biopsy returned a positive result in 90 of 99 lymphoma‐diagnosed dogs, for an overall sensitivity of 90.9%. The test showed a 96.2% detection rate in patients with B‐cell (50/52), 88.9% in T‐cell (16/18), and 60% in indolent lymphoma (3/5); 24 patients had an “undefined” or “other” phenotype, and the detection rate was 87.5% (21/24) in these patients. There was no statistically significant difference in detection rate by cancer stage or substage across phenotypes. In 35 of the 99 lymphoma‐diagnosed dogs, a “Cancer Signal Origin” prediction of “lymphoma” was correctly provided.

Conclusions and Clinical Importance: Liquid biopsy was successful at detecting genomic alterations associated with cancer in over 90% of dogs with a diagnosis of lymphoma, and was further able to predict the presence of lymphoma (vs any other cancer type) in over one‐third of these patients. This test offers veterinarians a novel, noninvasive tool for the detection and characterization of lymphoma in dogs.

68.1.1.1. ¹University of Minnesota; ²Associate Professor, Veterinary Clinical Sciences, University of Minnesota; ³Veterinarian, Small Animal Medicine, Veterinary Teaching Hospital, Federal University of Santa Maria

Background: Distinguishing primary and secondary pulmonary neoplasia can be challenging via cytology. No rapid, inexpensive diagnostic tool to aid in this distinction is currently available. Alkaline phosphatase cytochemistry (ALP‐CC) has been used in people to identify primary pulmonary carcinomas. We hypothesized this approach could be applied to canine and feline lung aspirates.

Objectives: The primary objectives of this study were to evaluate the utility of ALP‐CC in aiding in the diagnosis of primary pulmonary epithelial neoplasia in dogs. Secondary objectives including reporting ALP staining characteristics of canine mesothelium and feline pulmonary lesions.

Animals: Hospital population and archived samples.

Methods: A retrospective case search was conducted to identify canine and feline cases with contemporaneous cytology and histopathology reports. Neoplastic and non‐neoplastic etiologies were included. Slides were prospectively stained for ALP activity. Sensitivity and specificity were calculated using the histopathologic diagnosis as the gold standard. Mesothelial cells from cavitary lavage were previously collected as part of a different study and ALP activity was evaluated.

Results: Fifty canine cases and six feline cases met inclusion criteria. Ninety percent of canine primary pulmonary epithelial neoplasms were ALP‐positive. The sensitivity and specificity for assignment of pulmonary epithelial origin among all samples and among all neoplastic lung lesions were 90% and 80%, and 90% and 78% respectively. Canine mesothelial cells did not display ALP positivity. Feline pulmonary carcinomas displayed ALP positivity.

Conclusions: ALP‐CC has good sensitivity and specificity for aiding in the diagnosis of canine primary pulmonary epithelial neoplasia, and mesothelial cells do not display ALP‐positivity.

69.1.1.1. ¹Faculty of Veterinary Science, University of Pretoria; ²Professor, Department of Production Animal Studies, Faculty of Veterinary Science, University of Pretoria; ³Dr, Department Companion Animal Clinical Studies, Faculty of Veterinary Science, University of Pretoria; ⁴Professor, Department of Veterinary Clinical Sciences, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark; ⁵Professor, Department Companion Animal Clinical Studies, Faculty of Veterinary Science, University of Pretoria

Background: Prevalence of thrombotic disease in canine carcinoma and sarcoma and its association with clinicopathological parameters has not been investigated.

Objective: Estimate the prevalence of thrombotic disease in dogs with carcinomas and sarcomas and estimate the association with clinicopathological parameters.

Animals: 30 dogs with carcinoma, 32 with sarcoma (including 11 hemangiosarcomas) and 20 healthy dogs were prospectively enrolled.

Methods: All dogs had complete blood count, biochemistry, and hemostasis parameters (thromboelastography, fibrinogen, D‐Dimer concentration, Factor X, VII and antithrombin activity) performed. Dogs with cancer underwent complete post‐mortem and histopathological evaluation for intra‐ and extra‐tumoral micro‐ and macro‐thrombotic disease. Measured parameters were compared between healthy dogs, dogs with cancer and dogs with and without microthrombi.

Results: Microthrombi were identified in 31/62 (50%) dogs with cancer; 20/31 (65%) had exclusively intra‐tumoral microthrombi while 5/31 (16%) had exclusively distant microthrombi and 6/31 (19%) had both intra‐tumoral and distant microthrombi. Microthrombi were identified in 18/32 (56%) dogs with a diagnosis of sarcoma and 13/30 (43%) carcinoma. Macrothrombi were identified in three dogs, two with concurrent microthrombi. D‐Dimer concentration was also significantly higher in dogs with microthrombi (2950 ng/ml (interquartile range: 652, 5975)) compared to dogs without microthrombi (581 ng/ml (348, 1107), P<0.001). Hemangiosarcomas had significantly higher D‐Dimer concentrations (6000 ng/mL (3928, 6000)) compared to carcinomas (662 ng/mL (425, 2950)) and healthy dogs (287 ng/mL (200, 424), P<0.001), but not different to other sarcomas (714 ng/mL (473, 3930)).

Conclusion: Intra‐ and extra‐tumoral microthrombi are prevalent in dogs with carcinomas and sarcomas and positively associated with increased D‐Dimer concentration.

70.1.1.1. ¹University of California‐Davis, Davis, CA, USA; ²Department of Surgery, School of Medicine, University of California‐Davis, Davis, CA, USA; ³Department of Surgical and Radiological Sciences, School of Veterinary Medicine, University of California‐Davis, Davis, CA, USA

Background: Natural killer (NK) cells are a promising cellular therapy to treat cancer partly because allogeneic sources can be used for off‐the‐shelf treatment without the risk of graft‐versus‐host disease. We evaluated the combination of palliative radiotherapy (RT) and allogeneic NK cell transfer in a first‐in‐dog feasibility trial to speed translation of novel NK modalities in both dogs and people.

Hypothesis/Objectives: Our objective was to establish feasibility and describe preliminary data for safety and outcomes in dogs with unresectable oral melanoma treated with allogeneic NK cells following palliative RT.

Animals: Five dogs with naturally occurring melanoma were enrolled in this IACUC‐approved pilot trial. Allogeneic NK cells were expanded from blood obtained from five healthy beagles.

Methods: Patients underwent weekly RT×4 followed by infusion of intravenous allogeneic NK cells (7.5×10^6 cells/kg). Peripheral blood was obtained for biochemical and immune monitoring at baseline and day one, seven, and fourteen days post‐treatment.

Results: Adverse events related to NK infusion were classified as grade 1 or 2 and included emesis, fever, lymphopenia, metabolic acidosis, and hypoglycemia. Lymphocyte counts decreased one day post NK infusion but increased and peaked on day seven post‐infusion. Median survival time was 145 days with maximum and minimum survival times of 445 days and 48 days, respectively.

Conclusions and Clinical Importance: Allogeneic NK cell infusions appear well tolerated when administered to dogs with melanoma. This proof‐of‐concept trial provides preliminary data validating the canine model for investigating allogeneic adoptive NK cell transfer alone or in combination with other immunotherapies.

71.1.1.1. FidoCure/One Health Company

Spontaneous tumors in dogs are highly similar to human cancers histologically, genetically, molecularly and clinically, but the similarities across species correlating biomarkers and outcomes are under investigation. In this study, we identified interesting findings validating concordance across the cancer journeys. This is important to deepen the field of comparative oncology and affirm spontaneous cancer in dogs as an ideal ‘model’ for precision oncology, building upon recent work in tumor biology concordance. Using the Next‐Generation Sequencing (NGS) panel from FidoCure® Precision Medicine Platform associated with real‐world clinical information and treatment outcomes from dogs, we systematically evaluated the prognostic effect of genomic alterations of 1108 dogs from 2702 dogs enrolled in the platform. Our analysis identified mutations in TP53 and PIK3CA as prognostic markers for poor survival in dogs with an overall survival hazard ratio (OS HR) of 1.56 and 1.34, respectively (P<0.01, P=0.03). On the other hand, KIT mutations were associated with good prognosis (OS HR 0.44, P=0.02). These results are aligned with human findings where TP53 is associated with poor prognosis across several types of cancer. Among tumor types, hemangiosarcoma, histiocytic sarcoma and lymphoma showed the worst survival outcomes, OS HR=2.06 (P<0.01), HR=1.54 (P=0.01) and HR=1.47 (P=0.01), respectively. Interestingly, we also identified that tumors carrying somatic BRAF mutant tumors had a better prognosis when treated with lapatinib (OS HR 0.10, P=0.02), ATM with vorinostat (OS HR 0.01 P=0.03), and FLT3 with trametinib (OS HR 0.03, P=0.02). To the best of our knowledge, this is the largest canine tumor clinical genomic dataset and analysis. Applying real‐world evidence and data tools in the FidoCure® dataset, we successfully identified associations between gene mutations and survival and response to target therapy treatment. This study highlights how one health approach can improve not just R&D by using real‐time data from pet dogs being treated for cancer but also improving canine oncology outcomes through precision medicine and target therapy.

72.1.1.1. ¹VCA Pet CancerCare Alliance; ²Treasurer, Collaborative Care Coalition; ³Trone Research + Consulting; ⁴Head of Research and Data Science, Trone Research + Consulting; ⁵Center for Veterinary Business and Entrepreneurship, Cornell University; ⁶Director, Collaborative Care Coalition

Background: Collaboration between primary‐care veterinarians (pcVets) and veterinary oncologists is common for pets diagnosed with cancer, but no data exists that explores client perceptions of collaborative care.

Hypothesis/Objectives: The objectives of the study were to 1) describe client perceptions of the value of collaborative veterinary care when treating a pet for cancer, and 2) identify predictors of client satisfaction related to collaborative care between the pcVet and a veterinary oncologist.

Sample: The sample consisted of 890 U.S. dog owners who had pets diagnosed with cancer in the past three years.

Methods: Research was conducted using an online contextual survey. Scores related to client advocacy were derived from the net promoter scale, and were analyzed using predictive modeling.

Results: Seventy percent of clients across all income brackets rated seeing a specialist as a high value, based on money spent and patient outcome. Client perceptions of pcVets were six times more likely to improve with a referral than worsen. Delayed referral resulted in lower client satisfaction scores for pcVets. Top predictors of client satisfaction with pcVets were: being responsive to questions, staying involved with their dog's care, and willing to work with other veterinarians and specialists. For specialists, top predictors were: providing accurate cost estimates, cancer knowledge, and effectiveness in treatment and care. All were significant predictors of client advocacy (p<0.0001).

Conclusions and Clinical Importance: Clients view early referral and collaboration between pcVets and specialists positively, fostering client satisfaction and a positive value‐based care model for dogs diagnosed with cancer.

73.1.1.1. Vidium Animal Health

Genomic diagnostics are routinely used in human cancer medicine to guide diagnosis, treatment and prognostication. Growing evidence supports that human and canine cancers share genomic and clinical similarities. However, mutation‐based biomarkers from canine data are less well‐established than those utilized in human precision medicine.

Our objective was to bridge the canine precision medicine knowledge gap via “caninization” of the largest database of human cancer mutations, the Catalogue of Somatic Mutations in Cancer (COSMIC), and consume these data into a canine cancer knowledgebase, Vidium Insight.

To identify canine equivalents of human COSMIC mutations, cross‐species conversion was performed via both genomic liftover and protein alignment, considering evolution conservation score and sequence homology. Known/predicted pathogenic mutations in COSMIC Cancer Gene Census (CGC) genes were selected for “caninization” as a proof‐of‐concept.

Of 1763970 total mutation records in 717 CGC genes in COSMIC (v95), 295439 unique, pathogenic mutations in 707 CGC genes across ~200 tumor types were initially converted. Histologically similar canine cancers exist for most primary human cancer types, thus these inferences can be informative for dogs. Of 707 human CGC genes, 95% have canine orthologs and are highly conserved among species. Canine equivalents were identified for ~80% of human mutations.

Most of the nearly 3 million documented pathogenic human cancer mutations have canine counterparts. Thus, systematic “caninization” of these human cancer mutations with established biomarker associations will not only expand genomic knowledge of canine cancers in research settings, but will also improve utilities of genomic diagnostic tools in canine cancer patient care.

74.1.1.1. University of Cambridge

Background: Urinary extracellular vesicles (UEVs) are membrane bound particles that contain microRNAs (miRNAs) from cells in the urinary tract. Evaluation of UEV‐associated miRNAs in canine urothelial carcinoma (UC) could provide novel biomarkers for UC and indicate pathways involved in tumorigenesis.

Objectives: Compare UEV‐associated miRNAs of dogs with UC with those of healthy dogs and dogs with urinary tract infection (UTI—primary differential diagnosis for UC).

Animals: Discovery cohort: 12 UC dogs, 6 healthy dogs and 5 UTI dogs. Validation cohort: 8 UC dogs, 5 healthy dogs and 8 UTI dogs.

Methods: UEVs were isolated from 1–5 mL of urine by ultrafiltration combined with size exclusion chromatography, prior to RNA extraction. Small RNAs were identified using next generation sequencing in the discovery cohort. Differential expression (DE) analysis of small RNA sequencing data identified candidate miRNAs, the expression of which was evaluated in both discovery and validation cohorts using digital droplet PCR (ddPCR).

Results: DE analysis identified 11 miRNAs that were upregulated in UEVs from UC dogs (vs. healthy), six of which (miR‐143, miR‐145, miR‐150, miR‐199, miR‐451, miR‐486) were selected for further evaluation. Increased expression of most candidate miRNAs (after normalisation to urine creatinine concentration) in UEVs of UC dogs (vs. UTI and healthy dogs) was confirmed by ddPCR in the discovery cohort, but these candidate miRNAs were not significantly over‐expressed in the validation cohort.

Conclusions and Clinical Importance: Differentially expressed miRNAs associated with UEVs may provide insights into the pathogenesis of canine UC, however their utility as biomarkers requires further investigation.

75.1.1.1. ¹College of Veterinary Medicine and Biomedical Sciences, Colorado State University; ²Small Animal Internal Medicine, Colorado State University; ³Oncology, Colorado State University; ⁴Urgent Care Faculty, Emergency and Critical Care, Colorado State University

Background: The association between thromboembolic events and cancer in humans has been described over the course of almost two centuries, with the first observation taking place in the mid 1800’s. Little research exists in veterinary medicine on cancer‐associated thrombosis.

Objective: Describe the prevalence of neoplasia in dogs suffering from thromboembolic events and determine associated factors.

Animals: 154 client‐owned dogs where a thrombus or thromboembolic event was documented in the medical record.

Methods: Medical records were reviewed from 2014–2019. Cases were excluded if thrombi were not definitively identified or if the medical record was incomplete such that critical data was not available. Cases were examined for the presence of neoplasia, presence of metastasis, concurrent morbidities, exogenous steroid administration, recent surgery, and clinicopathological parameters including hematocrit, platelet count, nucleated cell count, and albumin concentration.

Results: Fifty‐five breeds were represented with Labrador Retrievers making up the majority (23/154, 14.9%). Of the 154 cases, 78 had confirmed neoplasia (50.6%) and 76 did not have evidence of neoplasia (49.4%). Of the 78 cases with neoplasia, 38 had metastatic disease (48.7%) and 40 did not have definitive evidence of metastasis (51.3%). The majority of cases were not receiving steroids (84.6%) and did not have recent surgery (88.4%).

Conclusions and Clinical Importance: Preliminary data from this retrospective study reveal that half of dogs with thrombosis had concurrent neoplasia and support the need for further investigation into predictive factors and preventative therapies in this patient population.

76.1.1.1. ¹College of Veterinary Medicine, Seoul National University; ²Professor, Laboratory of Veterinary Internal Medicine, Seoul National University; ³Administrative Assistant, College of Veterinary Medicine, Seoul National University

Background: Canine mammary gland cancer (CMGC) is one of the most common neoplasms in intact female dogs. However, adjuvant chemotherapy has not been demonstrated to have a benefit in dogs with CMGC.

Objectives: The objective of this study was to investigate the anti‐cancer effects of recently developed oral paclitaxel against canine mammary gland cancer in vitro and in vivo.

Animals: Twenty‐four female athymic nude mice xenografted with canine mammary gland carcinoma were used in this study.

Methods: This was a prospectively designed experimental study. Cell cycle arrest, apoptosis and angiogenesis of mouse xenograft tumor tissues were investigated by flow cytometry, western blot analysis, TUNEL assay and immunostaining using anti‐cluster of differentiation‐31 antibody.

Results: Oral paclitaxel inhibited both cell viability and induced G2/M phase cell arrest and apoptosis in a concentration‐proportional manner (p<0.01). In animal experiment, the average tumor volume was significantly decreased in proportion to the administered oral paclitaxel dose (p<0.01). Histologically, oral paclitaxel induced apoptosis and showed anti‐angiogenic effect in tumor tissues (p<0.01). It was also downregulated the expression of cyclin‐D1 in tumor tissues (p<0.01).

Conclusions and Clinical Importance: Our results suggest that oral paclitaxel may have anti‐cancer effects on CMGC through cell cycle arrest and induction of apoptosis. This study will provide a novel approach to the treatment of dogs with CMGC.

77.1.1.1. ¹Seoul National University; ²Future Medicine Co.

Background: The interleukin‐1 receptor‐related kinase 4 (IRAK4), downstream of Myd88, plays an essential role in hyperactive TLR signaling seen in some B‐cell lymphomas. In particular, efficient IRAK4 inhibitors of activated B‐cell subtype of human diffuse large B‐Cell lymphoma (DLBCL) are being developed. However, the anticancer effect of IRAK‐4 inhibitors in veterinary medicine has not been elucidated. It is, therefore, explored in this study involving the GL‐1 and CL‐1 canine lymphoma cell lines in vitro.

Hypothesis/Objectives: Treatment with IRAK4 inhibitor in canine lymphoid cell lines will decrease viability and increase apoptosis according to drug concentration.

Animals: Healthy Beagle dogs were included as donors for canine peripheral blood mononuclear cells (cPBMCs).

Methods: Pre‐clinical study. MyD88 expression was analyzed using polymerase chain reaction. GL‐1 and CL‐1 cells were subjected to concentration‐ and time‐dependent treatment with an IRAK‐4 inhibitor and assessed for viability, TLR signaling association, and apoptosis using a cell counting Kit‐8 assay, Western blotting, and flow cytometry.

Results: The GL‐1 and CL‐1 cells exhibited enhanced MyD88 expression, however, cPBMCs did not. The IRAK‐4 inhibitor reduced cell viability in a dose‐ and time‐dependent manner, significantly reduced the phosphorylation of molecules associated with TLR signaling at IC₅₀ such as IRAK1, IRAK4, NF‐κB and STAT3, and induced apoptosis in GL‐1 and CL‐1 cells.

Conclusions and Clinical Importance: The anticancer effect of the IRAK‐4 inhibitor on canine lymphoma cells is mediated by apoptosis via downregulation of TLR signaling. The potential of IRAK4 inhibitor as a treatment option for canine lymphoma is suggested.

78.1.1.1. PetDx

Background: Certain canine cancers occur in anatomical locations that are difficult or impossible to access by fine needle aspiration or surgical biopsy for confirmation of malignancy. Alternative means of assessing these “difficult to diagnose” (DTD) cases may offer utility to clinicians, patients, and their families.

Hypothesis/Objectives: To evaluate performance of a next‐generation sequencing‐based liquid biopsy test for the detection of genomic alterations associated with cancer in a population of dogs with a “difficult to diagnose” cancer.

Animals: Client‐owned dogs enrolled in a clinical validation study for a novel blood‐based liquid biopsy test.

Methods: Blood samples from an all‐comers cohort of 351 dogs with a variety of cancer diagnoses were subjected to DNA extraction, proprietary library preparation, and next‐generation sequencing. Sequencing data were analyzed using an internally developed bioinformatics pipeline to detect genomic alterations associated with the presence of cancer.

Results: Of 351 cancer‐diagnosed dogs, 33% (n=115) had malignant masses in locations designated as DTD, including bone and joint, central nervous system, head and neck, intra‐abdominal, and intra‐thoracic. The median age of DTD subjects was 10 years; median weight was 31.1 kg; 57% of subjects were male and 43% female; there was an equal distribution of purebred to mixed‐breed subjects. A Cancer Signal Detection (positive) result was issued in 54 cases, for an overall detection rate of 47% in the “DTD” cohort.

Conclusions and Clinical Importance: Liquid biopsy offers clinicians and families a new, noninvasive option for cancer evaluation in cases where traditional diagnostic approaches may be challenging or impossible.

79.1.1.1. ¹Seoul National University; ²National Institute of Animal Science; ³Associated Professor, Internal Medicine, Seoul Nation University

Background: The role of several genes is not well defined in the pathway of canine high‐grade multicentric lymphoma progression but may provide important information regarding the development of canine lymphoma.

Hypothesis/Objectives: The goal of this study is to explore the important genes and pathways involved in the progression of canine high‐grade multicentric lymphoma and to understand the underlying molecular mechanisms by using RNA sequencing.

Animals: Fourteen dogs (7 dogs with high‐grade multicentric lymphoma and 7 dogs without lymphoma) were included.

Methods: RNAs acquired from seven pairs of lymphoma patients and non‐lymphoma blood samples from different breeds of dogs were sequenced. Sequencing reads were preprocessed, aligned with the reference genome, and assembled. The expressions were estimated through bioinformatics approaches.

Results: At the false discovery rate (FDR) <0.05 and fold change (FC) ≥1.5, a total of 625 differentially expressed genes (DEGs) were identified between lymphoma and non‐lymphoma samples, including 347 up‐regulated DEGs such as SLC38A11, SCN3A, ZIC5, etc. and 278 down‐regulated DEGs such as LOC475937, CSMD1, KRT14, etc. Especially, the pathway of PI3K‐Akt signaling, which was directly related to the lymphoma, was enriched in KEGG analysis. In the protein‐protein interaction network, CDK1 was found to be the top hub gene with the highest degree of connectivity in lymphoma samples.

Conclusions and Clinical Importance: These findings were highlighted as a new candidate marker for the development of canine high‐grade multicentric lymphomas and facilitated the understanding of the molecular mechanisms of lymphoma progression in dogs.

80.1.1.1. ¹Virginia‐Maryland College of Veterinary Medicine; ²Virginia Polytechnic Institute and State University of Biomedical Engineering and Mechanics

Introduction: Histotripsy is a non‐thermal and non‐invasive high‐intensity focused ultrasound ablative technique that causes mechanical fragmentation of tissue resulting in liquefied cellular debris with histologically clear demarcated boundaries between treated and non‐treated tissue. Histotripsy has never been evaluated in a spontaneous cancer model. Our objective was to evaluate the safety and feasibility of histotripsy to achieve tumour ablation in dogs with soft tissue sarcoma (STS).

Methods: Dogs diagnosed with STS were recruited. CT of the chest, abdomen, and the tumour was performed for staging and treatment‐planning, and pre‐treatment biopsy was obtained. Safety was monitored with exams, owner reports and CBC/serum biochemistry. Partial tumour ablation was performed using a prototype Histotripsy system. Anatomical ablation characteristics were evaluated with contrast CT at 1‐ and 4‐days post‐treatment, with tumour resection 4‐days post‐treatment. Tumour ablation effectiveness was evaluated with H&E.

Results: Ten dogs were recruited and treated. Tumour histologies included three grade III STS, 4 grade II STS, 2 grade I STS and 1 malignant mesenchymoma. Currently, seven dogs are alive, two dogs were euthanized from recurrence or suspected metastasis and one dog was lost to follow‐up. There were no changes in bloodwork values. The mean planned ablated volume was 8.41±5.01 cm3. The mean duration of treatment was 30.43±12.67 min. Histotripsy‐related complications were generally self‐limiting and included various degrees of cutaneous injury. Post‐treatment histopathology indicated complete ablation of targeted tumour with no intact cells identified.

Conclusion: Histotripsy can achieve safe, rapid, and effective tumour ablation in dogs diagnosed with STS.

82.1.1.1. ¹University of Georgia; ²Department of Small Animal and Surgery, College of Veterinary Medicine, University of Georgia; ³Department of Population Health, College of Veterinary Medicine, University of Georgia

Background: The relative effects of angiotensin‐converting enzyme inhibitors (ACEi) and angiotensin receptor blockers (ARB) on the circulating renin‐angiotensin‐aldosterone system (RAAS) in dogs with proteinuric chronic kidney disease (CKD) are undescribed.

Objective: To characterize the RAAS in dogs with CKD treated with an ACEi (enalapril) or ARB (telmisartan).

Samples: Serum and urine from dogs with naturally occurring proteinuric CKD (n=36).

Methods: Serum concentrations of angiotensin (Ang) peptides (Ang I, II, III, IV, 1‐5, and 1‐7) and aldosterone, and urinary aldosterone‐to‐creatinine ratio (UACR) were determined using samples collected prior to and after 30 days of treatment with enalapril (n=17) or telmisartan (n=19). Liquid chromatography and mass spectrometry were used to determine serum concentrations of Ang peptides and aldosterone. Data were analyzed using linear mixed models controlling for concurrent treatment with amlodipine, presence of heart disease, and age.

Results: Relative to pre‐treatment values, serum concentrations of Ang II were significantly increased and decreased in telmisartan‐ and enalapril‐treated dogs, respectively (both P<0.001). Mean [95% CI] percentage change from pre‐treatment value in serum Ang1‐7 concentration was significantly greater in telmisartan‐ (753% [489% to 1134%]) versus enalapril‐treated (149% [69% to 268%]) dogs (P<0.001). Differences in percentage change in serum aldosterone or UACR, and significant effects of concurrent amlodipine treatment, presence of heart disease, or age, were not observed.

Conclusions and Clinical Importance: Compared to enalapril, treatment with telmisartan was associated with significantly greater increases in the beneficial Ang peptide Ang1‐7, representing a potential treatment advantage of the latter in dogs with CKD.

83.1.1.1. ¹Cornell University; ²UC Davis; ³Assistant Professor, Clinical Sciences, Cornell University

Background: Most dogs with aminoaciduric canine hypoaminoacidemic hepatopathy syndrome (ACHES) have superficial necrolytic dermatitis (SND) lesions (i.e., ACHES‐SND or hepatocutaneous syndrome). In people, SND is most commonly caused by glucagonomas that over‐produce glucagon. Dogs with ACHES are also predisposed to diabetes mellitus (DM). Glucagon‐like peptide‐1 (GLP‐1), derived from proglucagon, is important for glucose regulation, as it potentiates insulin secretion from pancreatic beta‐cells and may be dysregulated in ACHES.

Hypothesis/Objectives: To compare plasma concentrations of GLP‐1 between heathy and ACHES dogs, hypothesizing lower concentrations in ACHES.

Animals: Healthy dogs (n=5) and dogs with ACHES (n=8; ACHES‐DM n=1).

Methods: Preprandial and 15‐minute postprandial plasma total GLP‐1 concentrations were measured with a commercial assay (MesoScale Discovery). We compared groups by two‐way ANOVA with repeated measures or Mann‐Whitney test.

Results: Dogs with ACHES had significantly (P=0.02) lower plasma GLP‐1 concentrations than healthy dogs. The mean postprandial change in GLP‐1 was lower, although not significantly, in ACHES dogs (mean 1.85 pg/ml) than healthy dogs (3.5 pg/ml).

Conclusions and Clinical Importance: Lower plasma GLP‐1 concentrations suggest a contributing role to the propensity of DM in ACHES dogs. If lower GLP‐1 concentrations in ACHES are primarily due to reduced excretion, this may also be a surrogate marker of reduced GLP‐2 and concomitant reductions in amino acid transport function. The observed trends in GLP‐1 could reflect a unifying mechanism explaining hypoaminoacidemia, aminoaciduria, and predisposition to DM in ACHES, and an additional diagnostic or therapeutic target.

84.1.1.1. ¹Veterinary Teaching Hospital, Auburn University; ²Veterinary Information Network; ³Auburn University

Background: Endogenous ACTH is commonly measured using a validated chemiluminescent assay (Immulite 1000). Underfilling of blood tubes can result in high EDTA concentrations, which may interfere with hormone measurement. Magnesium chloride (MgCl₂) may reverse the effect.

Hypothesis/Objectives: The objectives of the study were 1) to investigate the effect of EDTA on ACTH measurement; 2) to determine if MgCl₂ addition could overcome the effect of EDTA.

Animals: Purchased canine serum was used.

Methods: EDTA was added to samples to concentrations of 4.1 mmol/L (simulating a completely filled collection tube), 8.2 mmol/L (50% filled), and 16.4 mmol/L (25% filled). Human ACTH was added to the desired concentration creating low and high concentration pools. The effect of 5 mmol/L or 6.7 mmol/L MgCl₂ in the presence of each EDTA concentration was evaluated. For each condition, 8 samples were created, and all samples were assayed in duplicate.

Results: A one‐way repeated measures ANOVA was performed to compare apparent concentrations between conditions with post hoc analysis via the Holm‐Sidak method. Significance was set at p<0.05. Addition of 8.2 mmol/L or 16.4 mmol/L EDTA significantly lowered apparent ACTH concentration compared to 4.1 mmol/L EDTA in both low and high ACTH pools. Addition of 5 mmol/L or 6.7 mmol/L MgCl₂ overcame the effect of EDTA.

Conclusions and Clinical Importance: Elevated EDTA concentrations in plasma that could result from underfilling of tubes causes a significant decrease in apparent ACTH concentration. This effect can be overcome with the addition of 5 or 6.7 mmol/L MgCl₂.

85.1.1.1. ¹Ontario Veterinary College; ²Animal Cancer Center, Ontario Veterinary College, University of Guelph; ³Department of Animal Biosciences, Ontario Agricultural College, University of Guelph; ⁴Department of Veterinary Medicine and Epidemiology, University of California, Davis; ⁵Veterinary Nutrition Group; ⁶Department of Pathobiology, Ontario Veterinary College, University of Guelph

Background: Exogenous iodine sources interfere with uptake of radioactive iodine (¹³¹I) by the thyroid gland. This has implications for the treatment of cats with hyperthyroidism that have recently undergone computed tomography (CT) with intravenous (IV) iodinated contrast medium (ICM). In human medicine, a 4‐week delay is required between CT and ¹³¹I.

Hypothesis/Objectives: It will require 4 weeks for urinary iodine concentration (UIC) to decrease to baseline after IV ICM administration.

Animals: 10 healthy adult neutered male cats living in a group colony.

Methods: Cats were sedated and received IV ICM (600 mg iodine/kg). On samples collected prior to injection (n=3) and on days 1, 2, 3, 7, 10 and weeks 2–6 post injection, UIC and urinary creatinine concentration (UCC) were measured, and the urinary iodine creatinine ratio (ICR) was calculated. Urinary outcome variables were modelled using a linear mixed effects model with time as a fixed effect and individual cats as a random effect.

Results: UIC increased 37‐ to 884‐fold on day 1 post ICM injection and returned to baseline during week 2 of the study. Compared to baseline, UIC was significantly increased for days 1–7, (all p<0.001); UCC was significantly lower for days 1–10 (all p<0.029); and ICR was significantly increased from day 1–10 (all p<0.001, except day 10 p=0.049).

Conclusions: Urinary clearance of iodine after IV ICM administration requires between 7–14 days to return to baseline in healthy cats. This suggests that a 2‐week delay between the iodinated contrast study and ¹³¹I treatment is appropriate.

86.1.1.1. ¹University of Florida Small Animal Hospital; ²University of Florida

Background: Continuous glucose monitoring systems that measure interstitial glucose (IG) have been validated previously for use in cats. In people, the FreeStyle Libre 2 (FSL2) is considered sufficiently accurate in the hypoglycemic range to make critical treatment decisions without the need to confirm its results with a blood glucose (BG) measurement. There are currently no studies on the accuracy of FSL2 in the hypoglycemic range in cats.

Objectives: To determine FSL2 accuracy during hypoglycemia in cats.

Animals: Six healthy, purpose‐bred cats.

Methods: Hyperinsulinemic‐hypoglycemic clamps were performed by concurrent IV infusion of regular insulin (constant rate) and dextrose (variable rate). To account for blood‐interstitium lag time, paired BG (AlphaTrak®) and FSL2 readings were considered only when BG was stable (change over 10 min ≤0.5 mg/dL/min). Pearson's r and Bland‐Altman tests were used for correlation and bias, respectively (p<0.05 considered significant).

Results: Overall, BG and IG correlated strongly (r=0.85, p<0.001). IG underestimated BG by 19.3±11.4, 9.9±7.4, and 2.5±5.6 mg/dL in the 80–120 (n=47), 60–79 (n=16), and 50–59 mg/dL (n=20) ranges respectively. IG overestimated BG by 6.0±6.8 mg/dL in the 38–49 mg/dL range (n=17).

Conclusions: Clinicians should be cautioned that while IG underestimates BG throughout most of the eu‐hypoglycemic range, IG generally overestimates BG in severe hypoglycemia (<50 mg/dL). It is therefore recommended to measure BG in cats with low IG readings before critical treatment decisions are made.

87.1.1.1. ¹Small Animal Hospital, University of Florida; ²Associate Professor, Director of Diagnostic Laboratories, North Carolina State University; ³Research Specialist, North Carolina State University; ⁴Microbiologist, North Carolina State University; ⁵Associate Professor, Small Animal Internal Medicine, North Carolina State University

Background: Most small animal diabetic patients are considered insulin‐dependent. Treatment can be challenging for cost‐conscious owners, particularly when insulins are discarded after 28–60 days of use, according to manufacturer recommendations. Veterinarians commonly recommend off‐label use of a single insulin vial for up to 6 months; however, limited data exist to support the safety of this practice.

Objectives: To evaluate sterility in refrigerated, multidose insulin vials through 6 months of routine aspiration.

Animals: None

Methods: Six vials each of Lantus (glargine, 10‐mL multidose vial, Sanofi, Bridgewater, containing metacresol preservative) and ProZinc (PZI, 10‐mL multidose vial, Boehringer Ingelheim, containing phenol preservative) insulin were refrigerated and aspirated twice daily for six months, using a new insulin syringe each time. Three of each insulin type were prepared with a single‐use alcohol swab prior to sampling. Three times weekly, aspirated samples were inoculated in Tryptic Soy Broth enrichment media (Fisher Scientific). Samples with evidence of microbial growth were cultured and speciated. Endpoints were microbial vial contamination (defined as three consecutive positive cultures of the same organism) and completion of the six‐month study period.

Results: Microbial contamination was not identified in any vial throughout the study period. A total of 454 aspirated samples were cultured, one of which exhibited non‐repeatable positive growth of Staphylococcus epidermidis. This vial was prematurely lost to breakage after 59 culture samples (29 after the positive growth).

Conclusions: Refrigerated phenol‐ and metacresol‐containing multidose insulin products carry minimal risk for iatrogenic infection through 6 months of use, regardless of alcohol swab preparation.

88.1.1.1. ¹Colorado State University; ²Animal Endocrine Clinic

Background: A thyrotropin‐releasing hormone (TRH) stimulation test can be used to evaluate the integrity of the pituitary‐thyroid gland axis. Administration of exogenous TRH (0.1 mg/kg) to cats with normal thyroid function will increase thyroid stimulating hormone (TSH) and serum total thyroxine (tT₄) concentrations. However, adverse side effects (vomiting, panting, hypersalivation) are common with that protocol.

Hypothesis/Objectives: To determine the lowest intravenous dose of TRH that adequately increases serum TSH and tT₄ concentrations in normal cats. Our hypothesis was that TRH administered at doses lower than 0.1 mg/kg will adequately stimulate TSH and tT₄ production in healthy cats and cause less side effects.

Animals: Six adult healthy research cats.

Methods: Cats were randomly administered 3 different doses of TRH (0.01, 0.05, or 0.1 mg/kg) intravenously. Serum TSH and tT₄ concentrations were measured using Immulite® 1000 at time 0, 30, and 60 minutes after TRH administration.

Results: All 3 TRH doses induced a significant TSH response (0.01 mg/kg, P=0.0055; 0.05 mg/kg, P=0.0017; 0.1 mg/kg, P=0.0001). There was no statistical difference among the different TRH doses in stimulation of TSH (Figure 1). Only the lowest dose of TRH (0.01 mg/kg) induced an increase in serum T₄ concentration (P=0.03). Administration of lower TRH doses (0.01 and 0.05 mg/kg) caused less side effects (1/6 cats) than did the highest concentration (0.1 mg/kg; 3/6 cats).

Conclusions and Clinical Importance: TRH stimulation test induced a significant TSH response at all concentrations in adult healthy cats. The results suggest using lower TRH doses to decrease the side effects.

89.1.1.1. ¹Department of Veterinary Medical Science, University of Parma; ²Department of Veterinary Medical Sciences, University of Bologna

Background: Non‐thyroidal illnesses (NTI) can affect serum total thyroxine (tT4) and thyrotropin (TSH) concentrations in dogs.

Hypothesis and Objective: To longitudinally evaluate tT4 and TSH in dogs in course of acute NTI and during the recovery phase.

Animals: Ten euthyroid client‐owned dogs from a hospital population.

Methods: Prospective, longitudinal, observational study. Dogs hospitalized due to acute NTI were enrolled. Serum tT4 and TSH concentrations were measured at the hospital admission (T‐1), every 72 h during the hospitalization, at the discharge (T0), at 3, 7, 14, and 21 days after discharge (T1, T2, T3, T4, respectively). Hypothyroidism was excluded based on tT4 within reference interval (WRI) and absence of clinical signs at T4.

Results: Mean (±SD) tT4 was 22.9 (±6.7), 28 (±9.3), 31.3 (±9.8) mmol/L, and mean (±SD) TSH was 0.22 (±0.17), 0.15 (±0.07), 0.31 (±0.19) ng/dL at T‐1, T0, T4, respectively. At T‐1, tT4 was WRI in all dogs and TSH was increased in 1/10 (10%) dogs. One out of 10 (10%) dogs showed low tT4 during the hospitalization. Both tT4 and TSH were WRI in all dogs at T0. Four out of 10 (40%) dogs showed high TSH at least once during the follow‐up period. TSH was increased in 2/10 (20%) dogs at T4. tT4 and TSH comparison between different time‐points showed no difference.

Conclusions and Clinical Importance: Serum TSH concentration increases above the upper reference limit in some euthyroid dogs during the recovery phase from NTI. This increase is detectable up to at least 21 days after the resolution of the acute phase of the disease.

90.1.1.1. ¹Department of Veterinary Clinical Medicine, University of Illinois, Urbana‐Champaign; ²PhD Candidate, Veterinary Clinical Medicine, University of Illinois, Urbana‐Champaign; ³DVM Student, Veterinary Clinical Medicine, University of Illinois, Urbana‐Champaign; ⁴DVM Student, Veterinary Clinical Medicine, University of Illinois, Urbana‐Champaign; ⁵Gastrointestinal Laboratory, Texas A&M; ⁶Professor, Small Animal Internal Medicine, Veterinary Clinical Medicine, University of Illinois, Urbana‐Champaign

Background: Gastrointestinal microbiota participate in regulating glucose homeostasis. Enteric microbiota dysbiosis has been associated with feline diabetes mellitus (DM), and with metabolic syndrome and DM in humans and other animals, but the functional association of microbial dysbiosis with feline DM has not been investigated.

Objectives: Compare gut microbial gene‐networks and metabolic pathways between diabetic and nondiabetic cats.

Animals: Ten diabetic and 21 nondiabetic age‐matched cats.

Methods: Whole‐genome shotgun sequencing (Illumina NovaSeq) and HMP Unified Metabolic Analysis Network (HUMAnN v3.0) were used to generate fecal metagenomic microbiome profiles. Microbiome Multivariable Associations with Linear Models (MaAsLin v2) was used to compare the abundance of microbial genes and metabolic pathways between groups.

Results: The relative abundances of 849 microbial genes from 37 distinct bacterial species belonging to the phyla Firmicutes and Bacteroidetes were significantly decreased in diabetic cats (FDR<0.05). Faecalibacterium prausnitzii (404/849; 47.6%) was the most common species associated with genes significantly under‐represented in diabetic cats. Metabolic pathway analysis identified six distinct microbial metabolic pathways significantly under‐represented in diabetic cats; five generate pyruvate and acetyl‐CoA, and the sixth generates the urea cycle intermediates ornithine and agmatine.

Conclusions: In health, intestinal gluconeogenesis and ureagenesis critically downregulate hepatic gluconeogenesis and satiety that are deregulated in DM. We speculate that bacterial pyruvate and acetyl‐CoA participate in the feline host intestinal gluconeogenesis, whereas ornithine and agmatine in the host's hepatic ureagenesis. Hence, deficiency in these bacterial metabolic pathways may contribute to the development and maintenance of feline DM through increased hepatic gluconeogenesis and decreased satiety.

91.1.1.1. ¹Department of Veterinary Clinical Medicine, University of Illinois, Urbana‐Champaign; ²University of Illinois, Urbana‐Champaign; ³School of Agriculture and Environment, Massey University

Background: Endocrine‐disrupting compounds (EDCs) are prevalent in the environment and have been detected in cat food and blood.

Hypothesis/Objectives: Determine the association between EDCs, altered metabolism, and hyperthyroidism.

Animals: Sixteen hyperthyroid (10 females, 6 males) and 19 euthyroid (11 females, 8 males) cats.

Methods: We prospectively collected urine and blood samples in which we measured creatinine‐normalized urinary EDCs (parabens and phthalates metabolites) and serum lipid metabolites by liquid chromatography‐mass spectrometry, and serum total T4 (tT4). We analyzed the data using a linear mixed effect models and used weighted correlation network analysis to identify subnetworks of serum lipids associated with thyroid status and urinary EDC concentrations.

Results: Hyperthyroid cats had significantly (p<0.05) increased ethyl‐paraben and decreased sum‐phthalate metabolites concentrations in their urine, compared with healthy cats. One hundred sixty‐two lipid metabolites significantly differed between groups. We identified four subnetworks of serum lipid metabolites significantly correlated with urinary EDCs and serum tT4. Two subnetworks, one containing triglycerides and another containing sphingomyelins, had positive correlations with ethyl‐, butyl‐, propyl‐ and sum‐paraben metabolites and mono‐methyl phthalate. A third phosphatidylcholines‐dominant subnetwork had positive correlations with methyl‐, propyl‐ and sum‐paraben metabolites. A fourth cardiolipins‐dominant subnetwork had negative correlations with ethyl‐, butyl‐, propyl‐ and sum‐paraben metabolites and mono‐isobutyl phthalate. Serum tT4 positively correlated with the triglycerides, sphingomyelins, and phosphatidylcholines subnetworks and negatively correlated with the cardiolipins subnetwork.

Conclusions: Feline hyperthyroidism is associated with differences in urine EDCs and serum lipid derivatives that warrant further investigation as agents that can affect thyroid gland transcription factors and growth.

92.1.1.1. ¹Department of Veterinary Clinical Medicine, University of Illinois at Urbana‐Champaign; ²University of Illinois at Urbana‐Champaign; ³AnimalBiome; ⁴Texas A&M; ⁵Massey University

Background: Gastrointestinal microbiota participate in regulating glucose homeostasis. Enteric microbiota dysbiosis has been associated with canine diabetes mellitus, and with metabolic syndrome and diabetes in humans and other animals. The effect of fecal microbial transplantation (FMT) on clinical glycemic control in diabetic dogs has not been evaluated.

Hypothesis/Objectives: FMT will improve clinical glycemic control in diabetic dogs.

Animals: Nine diabetic dogs (4 control; 5 FMT) with median (IQR) age 114 (26) months.

Methods: Prospective, double‐blind, placebo‐controlled randomized study. Dogs received daily 1 g of healthy‐donor lyophilized feces or corn‐starch in acid‐resistant capsule for 2 months and were treated with insulin Toujeo per a sliding‐scale insulin dosing protocol that was based on the 60‐minute postprandial interstitial glucose (IG) levels. 24‐hour water intake (24WI) adjusted to bodyweight, and serum fructosamine were analyzed with a linear mixed effect model for repeated measures with age, bodyweight, and baseline values of dependent variables as covariates; period, group and their interaction as fixed effects; and dog as random effect to account for repeated measures on the same dog. The area under the curve (AUC) of IG divided by AUC of daily insulin dose (AUC_IG/ins) was analyzed by a t‐test, and the percent change from baseline of HbA1C was analyzed by the Mann‐Whitney nonparametric test. P value was set <0.05.

Results: See figure for results.

Conclusions: Our interim analysis cautiously indicates a trend for improved glycemic control in diabetic dogs that were treated with oral lyophilized fecal capsules.

93.1.1.1. ¹University of Bologna; ²University of Parma; ³Associate Professor, University of Bologna

Background: In dogs, duration of hypothalamic‐pituitary‐adrenal axis (HPA‐axis) suppression after intermediate‐acting systemic glucocorticoids (IAGC)therapy is reported to vary from a few days up to 7 weeks after GC discontinuation. However, this data derives mainly from experimental studies on healthy dogs and not from animals with spontaneous disease.

Hypothesis and Objective: To determine the timeline for recovery of the HPA axis in a group of chronically ill dogs treated with IAGC.

Animals: Seventeen client‐owned dogs that received IAGC for at least one week.

Methods: Single‐center prospective observational study. ACTH stimulation test, endogenous ACTH (eACTH), serum biochemistry and urinalysis were performed at T0 (2–6 days after IAGC discontinuation)and then every two weeks until the documentation of HPA‐axis recovery (post‐ACTH cortisol >6 μg/dl).

Results: Median time of the HPA axis recovery was 3 days (2–133 days). 11/17 dogs (65%) showed recovery of the HPA‐axis at T0, 3/17 (17%) after 15‐17 days and one dog (6%) at 31, 64, 133 days, respectively. Cumulative dose, maximum dose, median daily dose, duration of treatment with IACG and body weight were not correlated with the timing of HPA‐axis recovery. ALT, GGT, ALP and eACTH were significantly correlated with the post‐ACTH cortisol (rs=−0.45, P=0.0059; rs=−0.33, P=0.047; r=−0.30, P=0.045, rs=0.44, P=0.008, respectively). Timing of HPA‐axis recovery in dogs underwent alternate‐day tapering process was not different compared to dogs who did not (3.5 vs. 3 days, P=0.98).

Conclusion and Clinical Importance: The majority of dogs showed recovery of the HPA‐axis within 4 days after IAGC discontinuation. However, 12% of dogs required more than 8 weeks.

94.1.1.1. University of Florida

Background: The oral glucose tolerance test (OGTT) is the most sensitive test for detecting prediabetes in people; however, it cannot be used in cats because of the stress associated with force feeding glucose. In non‐diabetic cats, capromorelin blunts insulin secretion and causes glucose intolerance.

Objectives: To develop a modified, stress‐free OGTT and assess its ability to detect mild dysglycemia caused by capromorelin in cats.

Animals: Ten purpose‐bred healthy cats.

Methods: Repeated‐measure study. Blood samples were collected via previously implanted vascular access ports. After 16 h fasting, capromorelin (3 mg/kg) or placebo were administered at ‐60 min, before offering at time zero 4 g/Kg of a highly palatable diet (Hill's a/d), mixed with 0.5 g/Kg of glucose. Blood glucose concentrations (BG) were measured at ‐70, ‐61, ‐10 and ‐1 min before, and at 15, 30, 45, 60, 120, and 180 min post‐feeding. A mixed‐model analysis was used to compare BG between time points and treatments with Bonferroni adjusted p‐values for multiple comparisons.

Results: All cats voluntarily consumed the glucose‐rich meal in its entirety in <2 min. Compared to placebo, BG was increased after capromorelin at ‐1, 30, 45, 60 and 180 min (adjusted p<0.05, Figure 1). Post‐feeding, BG best discriminated capromorelin and placebo at 45 min, differing by (median [range]; 95% CI) 40% (12–69%); 29.6–50.1%.

Conclusions: This study demonstrates a simple protocol for a stress‐free modified OGTT. Future studies should examine the utility of a one‐sample (45 min post‐feeding) ear‐prick protocol for diagnosis of prediabetes in cats.

95.1.1.1. ¹Colorado State University; ²Resident, Small Animal Surgery, University of California, Davis; ³Assistant Professor, Small Animal Internal Medicine, Colorado State University; ⁴Assitant Professor, Small Animal Internal Medicine, Oregon State University

Background: Historically, routine urine cultures were commonly performed in diabetic dogs regardless of lower urinary tract signs (LUTS). However, recent guidelines do not recommend screening urine cultures in diabetic dogs that are not showing LUTS. No recent studies have evaluated the association between clinical signs and positive urine culture as defined under the current guidelines in dogs with diabetes mellitus.

Objective: Determine the prevalence of subclinical bacteriuria (i.e., positive urine culture without LUTS) in dogs with diabetes mellitus.

Animals: 107 dogs with diabetes mellitus.

Methods: Retrospective study evaluating diabetic dogs with a paired urinalysis and urine culture. Relationship between presence of LUTS, pyuria, and bacteriuria and urine culture results were compared using Fisher exact testing.

Results: Fifteen dogs (14%) had a positive urine culture, of which 8 (53%) had pyuria, 15 (100%) had bacteriuria, and 4 (27%) had LUTS. Of the 88 dogs (82%) without LUTS, 11 (13%) had a positive culture. A significant association was found between a positive urine culture and the presence of pyuria (OR infinity; 95% CI 20.34–infinity, P<0.00001) and bacteriuria (OR infinity; 95% CI 164.4–infinity, P<0.00001). No association was found between urine culture results and the presence of LUTS (OR 1.87; 95% CI 0.59–6.85, P=0.46).

Conclusion and Clinical Importance: Subclinical bacteriuria occurred in this cohort of dogs, and our findings reinforce the recommendation that urine cultures should not be routinely performed in diabetic dogs particularly if pyuria and bacteriuria are also absent.

96.1.1.1. ¹Animal Endocrine Clinic; ²Professor Emeritus, Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University; ³Department of Clinical Sciences, College of Veterinary Medicine, Cornell University

Background: Many clinicians believe that hyperthyroidism can cause polyuria and sub‐optimally concentrated urine in cats. Prevalence of dilute vs. concentrated urine concentration in hyperthyroid cats, and if this changes after treatment, are unknown.

Objectives: To determine proportion of hyperthyroid cats with serum creatinine (SCr) <2.0 mg/dL and USG <1.035 prior to treatment. To determine if urine concentration changes after euthyroidism is restored, and whether an initial USG <1.035 predicts post‐treatment azotemia (SCr >2.0 mg/dL).

Animals: 614 hyperthyroid cats.

Methods: Prospective, before‐after study. Cats had serum T₄, T₃ and creatinine concentrations and USG measured before and 6 months after successful treatment with radioiodine. Cats with pretreatment serum creatinine >2.0 mg/dL or treatment failure were excluded.Results: 296/614 (48.2%) hyperthyroid cats had USG <1.035, and 318/614 (51.8%) had USG ≥1.035 before treatment. After treating cats with initial USG <1.035, 257/296 (86.8%) remained <1.035, but 39/296 (13.2%) became concentrated (Figure). After treating cats with USG ≥1.035, most still had USG ≥1.035, but 63/318 (20.4%) became USG <1.035. Fewer cats with pre‐treatment USG ≥1.035 (19/318; 6.0%) developed post‐treatment azotemia than cats with pre‐treatment USG <1.035 (121/296; 40.9%; P<0.001, Chi‐squared). USG <1.035 had a high negative predictive value (94%) for post‐treatment azotemia, but poor positive predictive value (41%).

Conclusions/Clinical Importance: Hyperthyroid cats have equal prevalence of UGS ≥1.035 vs. <1.035, which tends not to change after radioiodine treatment. Of cats with USG <1.035, nearly half developed SCr >2.0 mg/dL after treatment, whereas cats with USG ≥1.035 had a low probability of developing SCr >2.0 mg/L.

97.1.1.1. ¹School of Veterinary Medicine and Animal Science of University, Sao Paulo; ²Grandfood Ind and Com LTDA, Premier Pet

Background: Hemostatic abnormalities in diabetic subjects result in hypercoagulability. Omega‐3 is an interesting supplementation to endocrinopathies, but it can impact hemostatic system.

Objectives: To compare the hemostatic parameters between healthy and diabetic dogs and to investigate impact of omega‐3 supplementation on those parameters.

Animals: Seventeen healthy dogs and sixteen diabetic dogs with good glycemic control.

Methods: Healthy dogs (HE group) ate for three months the same higher‐fiber extruded diet recommended to diabetic dogs. Diabetic dogs received, randomly and by a double‐blind way, that higher‐fiber diet (DM group) and the same extruded diet enriched with 5.0% inclusion fish oil (DMω3), for three months each period. After each period, a blood sample was collected and coagulation panel [prothrombin time (PT), activated partial thromboplastin time (aPTT), plasmatic fibrinogen levels (FB), and platelet count (PC)] analysis was used to assess coagulation. Data of HE vs. DM and DM vs. DMω3 were compared using T test or Wilcoxon test (p<0.05).

Results: There was no age difference between healthy and diabetic dogs (p=0.86). Diabetic dogs had lower PT (p<0.001) and aPTT (p<0.01), and higher FB (p<0.001) and PC (p=0.02) than healthy dogs (Table 1). Supplementation of omega‐3 was approximately 142 mg EPA+DHA/kg body weight. There were no differences (p>0.05) on coagulation panel among dogs receiving or not omega‐3 (Table 1).

Conclusions: Diabetic dogs had hypercoagulopathy profile and fish oil supplementation did not impact PT, aPTT, plasmatic fibrinogen level and platelet counts.

98.1.1.1. ¹Gulfcoast Veterinary Oncology and Internal Medicine; ²Private Consultant; ³Medical Director, Thyroid and Endocrine Center of Florida; ⁴Sarasota Apothecary; ⁵Professor Emeritus, College of Veterinary Medicine, The Ohio State University

Background: Cinacalcet is approved to treat severe hypercalcemia in humans with primary hyperparathyroidism (PHPT) who are unable to undergo parathyroidectomy. In 2018, generic cinacalcet became available and potentially affordable for veterinary patients.

Hypothesis: Cinacalcet will be well‐tolerated and effective in the management of hypercalcemia associated with PHPT in dogs.

Animals: Fourteen dogs diagnosed with symptomatic PHPT and ionized calcium values greater than 1.5 mmol/L as measured by the Zoetis iSTAT Alinity.

Methods: Cinacalcet was initially dosed at 0.5 mg/kg once daily. Dosages were increased weekly until the iCa fell below 1.45 mmol/L, side effects developed, or a dose of 10 mg/kg BID was reached. PTH was measured at baseline, and again when the target iCa was achieved. SPSS v24 was used to analyze the data with significance at p<0.05.

Results: Mean iCa concentration was 1.74±0.19 mmol/L at baseline and 1.39±0.28 mmol/L post‐cinacalcet (p=0.033). PTH levels were 13.68±7.35 pmol/L pretreatment and 9.96±7.31 pmol/L post‐cinacalcet (p=0.194). With doses ranging from 1 mg/kg every 4 days to 9.4 mg/kg BID, 11 (78%) dogs achieved the target iCa <1.45 mmol/L. The average number of days dogs remained within target iCa was 33.6 days (7–84). Side‐effects were observed in 64% of dogs: lethargy and anorexia (9), vomiting (2), and nonhypocalcemic associated episodes of shaking (3).

Conclusion: Cinacalcet was effective in significantly decreasing iCa in dogs with PHPT. However, there was considerable inter‐individual variation in dose response, side‐effects, and the duration of within‐target iCa.

99.1.1.1. ¹Royal Veterinary College; ²Queen Mary, University of London

Background: Cats have one CYP11B enzyme for aldosterone and cortisol synthesis; no validated immunohistochemical marker of aldosterone production exists. Ki67 is a marker of proliferation used to predict malignancy in other tumours.

Hypothesis/Objectives: Feline aldosterone production is associated with CYP11B, KCNJ5 and VSNL1 presence, and CYP17A1 absence. Ki67 index is higher in malignant lesions.

Animals: Two normal cats and 18 with primary hyperaldosteronism.

Methods: Retrospective case series. CYP11B, KCNJ5, VSNL1, CYP17A1 and Ki67 were detected by immunohistochemistry on formalin‐fixed paraffin‐embedded adrenal tissue (normal; n=2, adenoma; n=11, adenocarcinoma; n=7). Tumour protein expression was semi‐quantified by multiplying staining intensity (0–3) by percentage area, giving an overall H‐score (0–300). Percentage of Ki67 positive nuclei was calculated using QuPath. Mann‐Whitney U and Wilcoxon signed‐rank tests were used for independent and paired variables respectively. Results are reported as median [range].

Results: Normal zona glomerulosa was CYP11B, KCNJ5 and VSNL1 positive, and CYP17A1 negative. CYP11B (215, [80–300]) had significantly higher expression than CYP17A1 (72.5, [6–155]) in adenomas (p=0.004) and carcinomas (p=0.016). KCNJ5 and VSNL1 expression varied and was not different between adenomas and carcinomas (KCNJ5 220, [0–268] versus 35, [0–290]; p=0.2, VSNL1 210, [0–300] versus 130, [0–295]; p=0.3). Ki67 percentage did not differ (p=0.13) between adenomas (2.09, [0.02–4.29]) and adenocarcinomas (2.6, [0.13–13.46]).

Conclusions and Clinical Importance: High CYP11B expression and low, but not absent, CYP17A1 expression appears to be the most reliable indicator aldosterone producing adrenal tissue. Ki67 proliferative index is not predictive of malignancy.

100.1.1.1. ¹University of Ghent; ²University of Bern; ³Vetscope Pathologie Dettwiler; ⁴DOCVET Clinica Veterinaria Nervianese; ⁵Tierklinik Hofheim

Background: Aetiology and pathophysiology of canine medullary thyroid carcinoma (cMTC) and functional feline thyroid adenomatous hyperplasia (fTAH) remain poorly understood. 3D organoid culture could constitute an excellent in vitro model.

Hypothesis/Objectives: Establish and characterize patient‐derived organoid cultures of cMTC and fTAH, explore new therapeutic options for cMTC in cMTC organoids and evaluate the effect of TSH on growth and thyroid hormone (TH) production in fTAH organoids.

Animals: One frozen cMTC and fTAH patient‐derived tissue sample.

Methods: Organoids from one cMTC and fTAH were cultured. Histology and immunohistochemistry for thyroglobulin, calcitonin, vimentin and Ki‐67 were performed. While immunohistochemistry for thyroid transcription factor‐1 (TTF‐1) and synaptophysin was also performed on cMTC organoids. CellTiter‐Blue® viability assays were used to evaluate the effect of carboplatin, meloxicam and toceranib phosphate on cMTC organoid growth and to evaluate the effect of different concentrations of TSH on fTAH organoid growth. TH production by fTAH organoids was evaluated after incubation with sodium iodide and different TSH‐concentrations.

Results: Organoids were cultured for 99 (cMTC) and 65 (fTAH) days. Organoids resembled the primary tissue histologically. TTF‐1 and calcitonin immunolabeling in cMTC organoids confirmed thyroid and C‐cell origin. The fTAH organoid cells showed thyroglobulin immunolabeling (5–10%) and no calcitonin immunolabeling. No drug had effect on cMTC organoid growth. TSH had no effect on fTAH organoid growth and no TH production was detected.

Conclusions and Clinical Importance: Organoid cultures provide a promising in vitro model to explore pathophysiology and new treatment modalities for cMTC and fTAH.

101.1.1.1. ¹Veterinary Teaching Hospital, Texas A&M University; ²Associate Professor, Veterinary Teaching Hospital, Texas A&M University; ³Veterinary Medicine & Biomedical Sciences, Texas A&M University; ⁴Clinical Assistant Professor, Veterinary Teaching Hospital, Texas A&M University

Background: Flash glucose monitoring systems (FGMS) are frequently used for convenient interstitial glucose (IG) monitoring in dogs with diabetes mellitus and are typically placed between the scapulae.

Objective: To evaluate the variability between glucose measurements from FGMS placed in 2 locations (between the scapulae and over the hip) in non‐diabetic dogs during rapidly induced hypoglycemia.

Animals: Twenty‐four apparently healthy colony dogs that were subjects in a teaching laboratory.

Methods: Prospective interventional study. FGMS (FreeStyle Libre, Abbott) were placed between the scapulae and over the hip of all dogs. Regular insulin was administered at 0.3 u/kg intravenously and the subsequent hypoglycemia was corrected. Prior to insulin administration and every 10 minutes over a 90‐minute period, IG was recorded from both locations with FGMS. Concurrently at each timepoint, blood glucose was obtained with a portable blood glucose meter (PBGM) from a venous blood sample. Measurement between FGMS locations were compared using Bland Altman plots (average vs. scapulae—hip) and Passing‐Bablok regression.

Results: A total of 209 out of a possible 240 paired FGMS measurements were available for comparison. There was a constant bias of 5.6 mg/dL with 95% limits of agreement of ‐26.3 to 37.5 mg/dL between FGMS locations. There was a correlation between FGMS locations (r=0.731; intercept=‐2.676; slope=0.971).

Conclusions and Clinical Importance: In this model of rapidly induced hypoglycemia, there was wide divergence between IG paired sample readings from FGMS placed between the scapulae and over the hip.

102.1.1.1. Banfield Pet Hospital

Background: The etiology of feline hyperthyroidism has not been completely elucidated. As part of ongoing research, the geographic prevalence of feline hyperthyroidism was investigated.

Hypothesis/Objectives: A non‐uniform geographic distribution of feline hyperthyroidism could generate additional insight into pathophysiology.

Animals: Privately owned cats with naturally occurring hyperthyroidism presented to 1084 Banfield Pet Hospitals from July 12, 2020, to July 17, 2021. All hospitals function as primary care facilities and do not offer specialized services (e.g., radioactive iodine). There was no directed recruitment of affected cats, and this population was anticipated to approximate naturally occurring illness.

Methods: Electronic medical records were retrospectively reviewed to identify cases. Inclusion criteria were a clinical diagnosis of feline hyperthyroidism and/or total thyroxine concentration greater than or equal to 5.5 μg/dl or 60 nmol/L. The percentage of unique patients diagnosed with hyperthyroidism was calculated. Direct standardization was applied to calculate age‐adjusted disease prevalence by state using Banfield's overall feline population as the standard population and six age groups (delineated by years): 0–<8, 8–10, 11–13, 14–16, 17–19 and 20+.

Results: Overall hyperthyroid prevalence was 177 per 10,000 cats seen (95% CI, 174–180 per 10,000). The Pacific Northwest and the Eastern Seaboard had the highest prevalence of feline hyperthyroidism (≥200 cases per 10,000; Image 1).

Conclusions and Clinical Importance: This study presents new information that may be helpful in determining risk factors for the development of feline hyperthyroidism. By delineating specific areas of increased risk, further investigations may be focused on dietary, environmental, genetic or other factors.

103.1.1.1. ¹Zomedica Inc.; ²CRW Consulting

Background: Hyperadrenocorticism, commonly referred to as Cushing's disease, is a common endocrine disorder in dogs. For suspected Cushingoid dogs, a low‐dose dexamethasone test (LDDST) is considered the diagnostic test of choice. In‐clinic testing for Cushing's disease would be more convenient for the practitioner and allow for more timely results. An appropriate testing platform with well‐defined and accurate hormone measurements is needed.

Objective: Assess the diagnostic performance of canine cortisol measurements for Cushing's disease at the point of care.

Animals: Forty‐one canines with clinical and clinicopathologic signs of Cushing's disease from three veterinary clinics and 40 healthy research colony canines.

Methods: All animals underwent a LDDST and serum cortisol levels were measured using the in‐clinic TRUFORMA® diagnostic platform. Animals suspected of Cushing's disease underwent a physical exam and screening workup, including CBC, chemistry, urinalysis, and LDDST performed at a reference lab. Blinded to TRUFORMA results, a board‐certified endocrinologist diagnosed the animals based on the reference lab results in addition to clinical and clinicopathologic signs.

Results: Healthy animals showed reduced cortisol levels 8 hours post‐dexamethasone injection compared to animals suspected, but not diagnosed with Cushing's disease (p=0.0225). Animals suspected and then diagnosed with Cushing's disease showed an increased cortisol level at 8 hours post‐dexamethasone injection (p=<0.0001). Using a diagnostic cutoff of 2.5 μg/dL for the 8‐hour TRUFORMA cortisol concentration, 40 of the 41 suspected animals (97.6%) were diagnosed in agreement with the endocrinologist.

Conclusion: Using an appropriate cutoff, the TRUFORMA point‐of‐care cortisol assay successfully distinguished animals with hyperadrenocorticism.

104.1.1.1. Virginia‐Maryland Regional College of Veterinary Medicine

Background: Diabetes mellitus is a common endocrine disease in cats. Good glycemic control may help achieve remission. The flash glucose monitoring system (FGMS—FreeStyle Libre system; FSL) allows for 14‐day continuous monitoring of interstitial blood glucose.

Hypothesis/Objectives: To evaluate the correlation of blood glucose concentration when measured by the FGMS as compared to the portable blood glucose monitor (PBGM—AlphaTRAK; AT) and the laboratory standard (hexokinase method; HM).

Animals: Twenty‐nine diabetic cats from hospital population. Exclusion criteria: diabetic ketoacidosis, dehydration, dermatological disease, fractious nature.

Methods: Prospective, observational study. Sensors placed on the lateral thorax on cats at initial visit. In‐hospital and at‐home blood glucose curves (BGC) performed using AT and FSL in each cat. Plasma glucose concentration assessed by HM once per in‐hospital curve. Correlations between FSL and AT, and FSL and HM evaluated using Lin's concordance correlation coefficient (CCC). Strength of agreement defined as excellent (>0.99), substantial (0.95–0.99), moderate (0.90–0.95), and poor (<0.90) based on Lin's CCC value. Adverse sensor wear events recorded.

Results: Two‐hundred fifty‐two glucose measurements compared between AT and FSL, and poor agreement was found. Thirty‐seven glucose measurements compared between FSL and HM, and substantial agreement was found. Median duration of sensor operation 9 days (range 1–13 days). Adverse events minor including skin irritation and temporary alopecia.

Conclusions and Clinical Importance: The FGMS is accurate compared to the HM and should be considered to facilitate at home BGC for diabetic cats. Sensors do not operate for the full period in a majority of patients.

106.1.1.1. ¹North Carolina State University, Raleigh, NC, USA; ²Department of Clinical Sciences, College of Veterinary Medicine, North Carolina State University, Raleigh, NC, USA; ³Department of Small Animal Clinical Sciences, University of Tennessee, College of Veterinary Medicine, Knoxville, TN, USA; ⁴Department of Clinical Sciences, College of Veterinary Medicine, North Carolina State University, Raleigh, NC, USA; ⁵Department of Clinical Sciences, College of Veterinary Medicine, North Carolina State University, Raleigh, NC, USA; ⁶Gastrointestinal Laboratory, Department of Small Animal Clinical Sciences, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University, College Station, TX, USA

Background: Gastric hyperacidity is reported to cause gastric ulceration in dogs with chronic kidney disease (CKD), yet there are no studies evaluating gastric pH in dogs with CKD.

Hypothesis/Objectives: Compare gastric pH and serum gastrin concentrations in dogs with CKD to healthy, age‐matched dogs.

Animals: Client‐owned dogs with CKD (International Renal Interest Society [IRIS] Stage I‐IV) and age‐matched, healthy dogs. Inclusion criteria for control dogs were absence of azotemia, USG >1.015, and normal body condition. Exclusion criteria for all dogs included historical or biochemical evidence of hepatobiliary disease, primary gastrointestinal disease, uncontrolled endocrine disease, or use of interfering drugs, including acid suppressants within 7 days of pH monitoring.

Methods: Prospective, case‐control study. Minimum database and serum gastrin concentration were evaluated within 24 hours of pH monitoring, which was performed by radiographic‐assisted placement of a pH capsule in the gastric fundus. Continuous pH recordings were compared over time and between groups using a repeated measures mixed‐model ANOVA.

Results: Thirteen dogs with CKD (mean±SD, age: 8.9±5.0 y) and 11 healthy dogs (age: 8.3±3.3 y) were enrolled. No significant differences were observed between groups for any pH parameters, including mean gastric pH (CKD: 2.37±0.87; healthy: 2.31±0.98; p=0.77). Serum gastrin concentrations for all dogs were within the reference interval.

Conclusions and Clinical Importance: Dogs with CKD do not have gastric hyperacidity compared to age‐matched, healthy dogs. These results do not support the routine use of acid‐suppressing drugs in dogs with CKD.

107.1.1.1. ¹College of Veterinary Medicine, University of Tennessee; ²Department of Small Animal Clinical Sciences, College of Veterinary Medicine, University of Tennessee; ³ASA PStat Accredited Professional Statistician, The Office of Information Technology, Veterinary Medical Center, University of Tennessee; ⁴Clinical Associate Professor of Medicine, Veterinary Medical Center, Department of Small Animal Clinical Sciences, University of Tennessee

Background: Previously evaluated gastroprotectants did not prevent glucocorticoid‐induced gastric bleeding in dogs, but neither probiotics or twice‐daily omeprazole were evaluated.

Hypothesis/Objectives: Compare gastrointestinal bleeding among dogs administered prednisone, prednisone with omeprazole, or prednisone with probiotics.

Animals: Twenty‐four healthy research dogs.

Methods: Double‐blinded, placebo‐controlled randomized trial. Dogs received placebo, prednisone (2 mg/kg q 24 h), prednisone/omeprazole (1 mg/kg q 12 h), or prednisone/probiotics (11.2–22.5 billion CFU/kg q 24 h) for 28 days. Clinical signs and endoscopic gastrointestinal mucosal lesions were determined at baseline, day 14, and day 28 of treatment. Results were compared using split‐plot repeated measures mixed model ANOVAs.

Results: Attitude, vomiting, and food intake did not differ among groups. Fecal score differed over time depending on treatment received (F [6,40]=2.65, P<0.029). Post hoc tests revealed scores increased in the prednisone‐receiving groups between baseline and day 28 (P<0.001, for each). Nineteen of 33 (58%) instances of fecal score ≥5 occurred in the prednisone/omeprazole group. Gastric mucosal lesion scores differed by treatment‐by‐time (F [6,60]=2.86, P=0.016), between treatments (F [3,60]=4.9, P=0.004), and over time (F [2,60]=16.5, P<0.001). Post hoc tests revealed lesion scores increased between baseline and days 14 and 28 for prednisone‐receiving groups (P≤0.03, for all). Ulcers occurred in 6 dogs: prednisone alone, 2; prednisone/omeprazole, 2; and prednisone/probiotics, 3.

Conclusions and Clinical Importance: Consistent with previous studies, prednisone induced gastric bleeding. Bleeding was not mitigated by co‐administration of omeprazole or probiotics.

108.1.1.1. ¹Royal Veterinary College; ²Internal Medicine, University of Bristol; ³Internal Medicine, The Royal (Dick) School of Veterinary Studies

Background: The impact of undernutrition in dogs with inflammatory protein‐losing enteropathy (iPLE) is currently unknown.

Objectives: To determine if an undernutrition screening score (USS) at the time of histopathologic diagnosis is associated with length of hospitalization, treatment response, clinical and biochemical remission, and death in dogs with iPLE.

Animals: Fifty‐seven dogs with iPLE prospectively recruited from three referral hospitals in the U.K.

Methods: Dogs with iPLE were recruited at the time of histopathologic diagnosis. The USS and the canine chronic enteropathy clinical activity index (CCECAI) were recorded at this time, along with historical and clinicopathologic data. The USS was scored out of 18 and consisted of parameters related to appetite, weight loss, and body, muscle and skin condition, with higher scores reflecting worse undernutrition. Follow‐up was obtained for at least 6‐months after histopathologic diagnosis.

Results: Dogs that were hospitalized for >5 days had higher USS (p=0.005) and CCECAI (p<0.001) at diagnosis. Dogs that failed to respond to treatment prescribed at diagnosis had higher USS (p=0.011) and CCECAI (p=0.002). Dogs that failed to achieve clinical remission within 6‐months had higher USS (p=0.037) and CCECAI (p=0.041). Dogs that died or were euthanized due to iPLE within 6‐months of diagnosis had higher CCECAI (P=0.001). The USS and CCECAI were positively correlated (p=0.001, correlation‐coefficient=0.414).

Conclusions and Clinical Importance: A higher USS at diagnosis is associated with negative outcomes in dogs with iPLE. Further studies are required to determine if nutritional intervention to correct undernutrition at diagnosis, helps to improve these outcome parameters.

109.1.1.1. ¹Small Animal Teaching Hospital, Texas A&M University; ²Clinical Assistant Professor, Veterinary Medical Teaching Hospital, Small Animal Clinical Sciences, Texas A&M University; ³Director, Gastrointestinal Laboratory, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University; ⁴Associate Director for Research, Gastrointestinal Laboratory, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University; ⁵Associate Professor, Veterinary Medical Teaching Hospital, Small Animal Clinical Sciences, Texas A&M University

Background: Canine parvovirus causes significant intestinal damage, potentially resulting in malabsorption. Measurement of cobalamin is used as a marker of ileal malabsorption in patients with chronic enteropathy. Prevalence of cobalamin deficiency in dogs with parvoviral enteritis has not been described.

Objective: Determine serum cobalamin and methylmalonic acid (MMA) concentrations in dogs with parvoviral enteritis, healthy dogs, and dogs with non‐parvovirus acute diarrhea (non‐PVAD) of similar age.

Animals: Thirty‐one healthy control dogs, 30 dogs with parvoviral enteritis (positive fecal parvovirus antigen test or PCR), and 24 dogs with non‐PVAD (negative antigen test and PCR). Dogs were <10 months and had not received cobalamin supplementation.

Methods: Prospective, case‐control study. Serum cobalamin (all dogs) and MMA (63 dogs) concentrations were measured within 24 hours of presentation.

Results: Median (min–max) cobalamin concentrations were 784 ng/L (293–1,619) in the healthy group, 447 ng/L (<150–805) in the parvovirus group, and 527 ng/L (160–8,838) in the non‐PVAD group. Serum MMA concentrations were 815 mmol/L (588–1,466), 858 (554–2,870), and 776 (392–831). Serum cobalamin concentrations were significantly lower in the parvovirus group than in the healthy group (p=0.0002), but not the non‐PVAD group (p=0.1517). No significant differences in serum MMA were found (p=0.1623).

Conclusions and Clinical Importance: Dogs with parvoviral enteritis had lower serum cobalamin concentrations than healthy controls, but did not appear to have cobalamin deficiency at the cellular level.

110.1.1.1. ¹College of Veterinary Medicine, The Ohio State University; ²Post Doctoral Scholar, Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University; ³Internal Medicine Resident, Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University; ⁴Associate Professor, Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University; ⁵Veterinary Student, Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University; ⁶Professor, Clinical, Veterinary Clinical Sciences, College of Veterinary Medicine The Ohio State University; ⁷Assistant Professor, Clinical, Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University

Background: Fecal collection in cats is required for research and clinical purposes but fecal identification is limited by feline elimination behaviors in group‐housing and multi‐cat households. Crayon shavings and glitter have been used as fecal markers, but their impact on the fecal microbiome is unknown.

Objective: To determine the impact of fecal identification markers, specifically crayon and glitter, on the feline microbiome.

Animals: This clinical trial was performed in six healthy and sterilized purpose‐bred cats (3 male, 3 female) group‐housed in a dedicated research facility.

Methods: A randomized crossover experiment was performed. Cats received either glitter or crayon shavings mixed in a feline adult maintenance diet for 14 days. A 14‐day washout was performed between crossovers. Feces was collected daily from each cat for 16S rRNA gene sequencing (V4 region). Amplicon analysis was performed in R‐studio (dada2; phyloseq). Pairwise PERMANOVA analysis compared samples across experimental phases and marker order.

Results: Fecal identification markers were readily identified in the feces of all cats. Statistically significant changes to the microbiome were identified with marker order (P<0.05) and across experimental phases (P<0.05) for 5/6 cats. Microbiome changes were cat‐dependent (P<0.001), suggesting idiosyncratic microbiome responses to glitter and crayon.

Conclusions and Clinical Importance: Both glitter and crayon allow for identification of individual feline fecal samples; however, these markers may impact the microbiome and therefore are not recommended in studies with microbiome endpoints. The difference in response to these markers among cats highlights the individuality of the feline microbiome.

111.1.1.1. ¹Kansas State University; ²Assistant Professor, Small Animal Internal Medicine, Clinical Sciences, Kansas State University; ³DVM Candidate 2022, Clinical Sciences, Kansas State University

Background: Glucagon‐like peptide‐2 (GLP‐2) is an enteroendocrine hormone responsible for gastrointestinal (GI) health with nutrient‐dependent secretion.

Hypothesis/Objectives: Maximal GLP‐2 plasma concentration occurs 30 minutes after meal ingestion, and a high‐fat meal provides greater GLP‐2 secretion stimulus than high‐protein or high‐carbohydrate meals.

Animals: Nine healthy research cats.

Methods: This was a randomized, cross‐over study. Cats were fed a standardized high‐fat, high‐carbohydrate, or high‐protein meal after an overnight fast. Cats received all 3 diets in random order with a 24‐hour period between sampling days. Blood samples were collected at baseline, 30, 60, 75, 90, and 120 minutes post‐meal. Plasma GLP‐2 concentration was measured using a commercial feline ELISA. The Friedman test (non‐normal data) or one‐way repeated measures ANOVA (normal data) was used to evaluate maximal GLP‐2 secretion within each diet. GLP‐2 concentrations were compared among diets using a mixed analysis of variance accounting for repeated measures.

Results: Plasma GLP‐2 concentration 30 minutes post high‐fat meal (1.64±0.23 ng/mL) was higher than 90 minutes (1.39±0.31 ng/mL; P=0.03) and 120 minutes (1.44±0.27 ng/mL; P=0.03) post‐meal. There was no difference in post‐prandial GLP‐2 concentration after the high‐protein or high‐carbohydrate meal compared to baseline. There was no difference in post‐prandial GLP‐2 concentrations between the meal types.

Conclusions and Clinical Importance: Results suggest that ingested fat is a small but significant stimulus for GLP‐2 secretion in cats. Further understanding of GLP‐2 secretion may help diagnose and treat feline GI disorders.

112.1.1.1. ¹Michigan State University; ²Assistant Professor, Small Animal Clinical Sciences, Michigan State University; ³Professor, Pathobiology and Diagnostic Investigation, Michigan State University; ⁴Associate Professor, Small Animal Clinical Sciences, Michigan State University

Background: Dogs with chronic enteropathy (CE) or protein‐losing enteropathy (PLE) and low serum 25‐hydroxyvitamin D (25[OH]D) variably develop ionized hypocalcemia and/or secondary hyperparathyroidism (SHPT). Serum magnesium concentrations influence calcium homeostasis and secretion of PTH. However, little is known about the relationship between magnesium and the development of ionized hypocalcemia and SHPT in dogs with chronic gastrointestinal (GI) disease and decreased 25[OH]D.

Objectives: Evaluate relationships between ionized magnesium (iMg) and PTH, ionized calcium (iCa), and 25(OH)D in dogs with CE +/‐ PLE and decreased 25(OH)D.

Animals: 50 dogs with CE +/‐ PLE and decreased serum 25(OH)D.

Methods: Retrospective search of submissions database at Michigan State University Veterinary Diagnostic Laboratory for vitamin D profiles submitted 2017–2020. Submitting practices contacted to obtain consent for use of patient data, confirmation of GI disease, and case information. Cases excluded if supplemented with Ca, Mg, or vitamin D. Spearman correlation performed to evaluate relationships between iMg and PTH, 25(OH)D, and iCa. Ionized Mg concentrations compared between dogs with SHPT and those with normal parathyroid hormone concentrations.

Results: Concentrations of iMg were weakly negatively correlated with PTH concentrations (rho,‐0.31; P=0.032), and weakly positively correlated with serum 25(OH)D (rho,.34, P=0.02) and iCa (rho,.42, P=0.003) concentrations. Dogs with decreased 25(OH)D and SHPT were more likely to have ionized hypomagnesemia compared to dogs with decreased 25(OH)D and normal parathyroid hormone concentrations (P=0.015).

Conclusions and Clinical Importance: Hypomagnesemia may contribute to alterations in iCa and parathyroid hormone in dogs with CE +/‐ PLE and decreased 25(OH)D.

113.1.1.1. ¹Iowa State University; ²President, LifeDiagnostics, Inc.; ³Professor of Veterinary Biochemistry, School of Veterinary Medicine, University of Glasgow; ⁴PhD Student, Biomedical Sciences, Iowa State University; ⁵Dual Degree Student, Biomedical Sciences, Iowa State University; ⁶Post‐Doctoral Scientist, Biomedical Sciences, Iowa State University; ⁷Associate Professor of Pharmacology, Biomedical Sciences, Iowa State University

Non‐invasive biomarkers to predict treatment responses in canine chronic enteropathies (CE) are urgently needed. In a recent exploratory proteomic study, several pancreatic/intestinal enzymes, acute phase proteins and immunoglobulins (Ig) were found to be differentially present in feces from healthy dogs and dogs with CE. In this follow‐up study, we sought to evaluate the discriminatory ability of these biomarkers to predict subgroups of dogs with CE.

Fifty‐eight dogs with CE were included in the study (25 food‐responsive diarrhea (FRD), 12 antibiotic‐responsive diarrhea (ARD), 10 steroid‐responsive diarrhea (SRD), 11 chronic diarrhea of other causes). Spatial Proximity Analyte Reagent Capture Luminescence (SPARCL^TM) assays were developed to measure the fecal concentrations of the following analytes: pancreatitis‐associated‐protein‐1 (PAP1), haptoglobin (HP), S100A8, IgG1, IgG2, IgGFc, alpha‐1‐acid glycoprotein (AGP), ceruloplasmin, C‐reactive protein (CRP), intestinal fatty acid‐binding protein (FABP), liver‐FABP, and gastrotropin. Multivariate logistic regression was used to assess the predictive ability of analytes to differentiate between CE and chronic diarrhea of other causes, and between subgroups of CE.

Fecal levels of gastrotropin were predictive of CE vs. chronic diarrhea due to other causes. In addition, dogs with lower fecal HP and CRP were more likely to be FRD. Furthermore, higher fecal concentrations of PAP‐1 were significantly associated with ARD. For SRD, higher CCECAI and weight loss, and higher fecal concentrations of IgG1, IgG2, CRP and gastrotropin were identified as markers with predictive ability.

In conclusion, this study suggests that novel fecal biomarkers PAP‐1 and gastrotropin could be predictive of ARD and SRD, respectively, in dogs with CE.

114.1.1.1. ¹School of Veterinary Medicine, UC Davis; ²Gastrointestinal Laboratory, Department of Small Animal Clinical Sciences, College of Veterinary Medicine and Biomedical Sciences, Texas A&M; ³Pathology and Laboratory Medicine, UC Davis Health; ⁴Axys Análises; ⁵Postdoctoral Scholar, Veterinary Medicine and Epidemiology, School of Veterinary Medicine, UC Davis; ⁶Assistant Adjunct Professor, Pathology and Laboratory Medicine, UC Davis Health; ⁷Professor and Vice Chair for Strategic Technologies, Pathology and Laboratory Medicine, UC Davis Health; ⁸Assistant Professor, Veterinary Medicine and Epidemiology, School of Veterinary Medicine, UC Davis

Background: Fluorescence imitating brightfield imaging (FIBI) is a novel microscopy method allowing for real‐time, non‐destructive, slide‐free tissue imaging of fresh, formalin‐fixed, or even paraffin‐embedded tissue. The non‐destructive nature of technology permits tissue preservation for further downstream analysis.

Hypothesis/Objectives: To assess the utility of FIBI compared to conventional H&E‐stained histology slides in feline gastrointestinal histopathology.

Animals: Formalin‐fixed paraffin‐embedded (FFPE) full‐thickness small intestinal tissue specimens from 50 cases of feline chronic enteropathy (FCE).

Methods: Observational study. The ability of FIBI to evaluate predetermined morphological features (epithelium, villi, crypts, lacteals, fibrosis, submucosa, muscularis propria) and inflammatory cells was assessed on a 3‐point scale (0=FIBI cannot identify the feature; 1=FIBI can identify the feature; 2=FIBI can identify the feature with more certainty than H&E). H&E and FIBI images were also scored according to World Small Animal Veterinary Association (WSAVA) Gastrointestinal Standardization Group guidelines.

Results: FIBI identified morphological features with similar or in some cases higher confidence compared to H&E images (subscore 0.90). The identification of inflammatory cells was less consistent (subscore 0.48). FIBI and H&E showed an overall poor agreement with regards to the assigned WSAVA scores.

Conclusions and Clinical Importance: While FIBI showed an equal or better ability to identify morphological features in intestinal biopsy specimens, identification of inflammatory cells is currently inferior compared to H&E‐based imaging. Future studies on the utility of FIBI as a diagnostic tool for non‐inflammatory histopathologic lesions are warranted.

115.1.1.1. Gastrointestinal Lab, Veterinary Small Animal Clinical Sciences, Texas A&M University

Background: Amino acids (AA) affect many metabolic processes and serve as an important energy source for enterocytes. Alterations in plasma or serum AA concentrations have previously been reported in dogs and cats with gastrointestinal (GI) disease. However, studies suggest that measurement of fecal AA may be more useful in discriminating human inflammatory bowel disease (IBD) patients from healthy controls.

Objectives: Compare fecal and serum AA concentrations between dogs with chronic enteropathy (CE) and healthy control dogs (HC).

Animals: Feces and serum were collected from HC dogs (n=29 and 23, respectively) and dogs with CE (n=29 and 15, respectively). GI biopsies were also collected (stomach, small intestine, colon; HC=13 and CE=8).

Methods: Retrospective cross‐sectional study. Fecal and serum AA concentrations (measured on an Agilent HPLC and Biochrom 30+ AA analyser, respectively) were compared between groups with Mann‐Whitney tests. Correlation analysis was performed between fecal and serum AA concentrations and histopathological scores. Results were corrected for multiple comparisons using Benjamini‐Hochberg method.

Results: Fecal concentrations of tryptophan were significantly higher (Q=0.023) in dogs with CE than in HC dogs. Serum concentrations of valine were significantly higher (Q=0.026) in dogs with CE than in HC. After correction for multiple comparisons, all correlations with histopathological scores failed to reach statistical significance.

Conclusions: Fecal and serum AA profiles are altered in dogs with CE. Therefore, some AA may have the potential to be used as future therapeutic targets.

116.1.1.1. ¹Swedish University of Agricultural Sciences; ²Assistant Professor, Clinical Sciences, Swedish University of Agricultural Sciences; ³Associate Professor, Clinical Sciences, Swedish University of Agricultural Sciences; ⁴Anicura Albano; ⁵Professor, University of Helsinki; ⁶Professor, Clinical Sciences, Swedish University of Agricultural Sciences

Background: Information about prevalence of canine chronic enteropathy (CE) is limited.

Objectives: Investigate period prevalence, clinical features and diagnostic results of CE in dogs at two Swedish animal hospitals.

Animals: Eight hundred fourteen dogs with CE.

Methods: A retrospective medical record search was performed to identify dogs with CE including dogs with ≥3 visits because of gastrointestinal disease and/or undergone gastroduodenoscopy/colonoscopy during 2013–2018. Dog characteristics, case history, physical examination, diagnostic results, therapeutic protocol and response were recorded.

Results: Eight hundred fourteen dogs met inclusion criteria for CE. Period prevalence of CE was 1.1% of total number of dogs visiting the hospitals, and 4.4% of all dogs presenting with gastrointestinal signs. Breeds with highest risk were Norwegian Lundehund, West Highland White Terrier, and Miniature Poodle. Dogs presented at median age of 3.8 (IQR 1.8–6.8) years. French Bulldogs and Miniature Schnauzers presented younger (below 2.5 years) compared to other breeds (P<0.05). In a subset of dogs, hypoalbuminemia (116/662, 17.5%), hypocobalaminemia (98/647, 15.1%), and increased C‐reactive protein (145/267, 54.3%) were recorded. Treatment outcome was classified in 72.9% of dogs, and characterized as immunosuppressant‐responsive (55.2%), food‐responsive (11.4%), non‐responsive (5.2%) and antibiotic‐responsive (1.1%). Non‐responsive dogs were more likely to present with hypoprotein/albuminemia, anemia, and high CRP (P<0.05).

Conclusions and Clinical Importance: Prevalence of dogs with CE at Swedish hospitals was in agreement with earlier reports. Risk breeds differed slightly from previous reports and earlier age of CE onset was found in two breeds. The largest proportion of dogs was immunosuppressant‐responsive and the smallest antibiotic‐responsive.

117.1.1.1. ¹Louisiana State University; ²Professor, Veterinary Clinical Sciences, Louisiana State University; ³Graduate Research Assistant, Gastrointestinal Laboratory, Department of Small Animal Clinical Sciences, Texas A&M University; ⁴Research Assistant Professor, Gastrointestinal Laboratory, Department of Small Animal Clinical Sciences, Texas A&M University; ⁵Professor and Associate Director for Research, Head of Microbiome Sciences, Gastrointestinal Laboratory, Department of Small Animal Clinical Sciences, Texas A&M University

Background: Intestinal dysbiosis occurs frequently in cats with gastrointestinal diseases or following antibiotic administration. Fecal microbiota transplantation (FMT) may be beneficial in people and dogs with gastrointestinal disease, but data in cats is limited.

Objectives: To investigate the effects of FMT in shortening the recovery of the fecal microbiome in healthy cats after amoxicillin and clavulanate administration.

Animals: Eighteen healthy research cats.

Methods: Prospective, randomized, controlled, IACUC‐approved clinical trial. All cats received a 10‐day course of antibiotics, then were randomized to receive FMT via enema (n=10) or no treatment (n=8). Fecal samples were collected at baseline, immediately after antibiotic administration, and 7, 14, 21 and 56 days later. Abundance of 8 bacterial taxa was determined via qPCR and a fecal feline dysbiosis index (FDI) was calculated. A 2‐way repeated measures ANOVA with post hoc pairwise comparisons was used.

Results: The FDI increased in all cats after antibiotics (baseline ‐2.4±0.1, post‐antibiotics 2.7±0.4, p<0.0005). No adverse reactions were observed during or following FMT. The FDI of control cats was still increased compared to baseline 7, 14 and 21 days after antibiotic discontinuation (0.4±1.9; 0.5±1.9; ‐0.1±2.2 and ‐1.3±2.4 respectively, all p<0.05) while it was not significantly different from baseline in FMT‐treated cats.

Clinical Importance: FMT was a well‐tolerated procedure that has the potential to accelerate reversal of dysbiosis in this feline model of acute antibiotic‐induced dysbiosis. The efficacy of FMT seen in this study may serve as a platform for investigation of its role in cats with various enteropathies.

118.1.1.1. ¹Iowa State University; ²Associate Professor, Iowa State University; ³The Ohio State University

Background: Dysbiosis is associated with mucosal inflammation in dogs with inflammatory bowel disease (IBD). Treatments attempting to reduce mucosal inflammation by normalizing dysbiotic states are a rapidly growing research area. A dietary supplement containing a synbiotic and avian Immunoglobulin Y (IgY) was developed to correct gut dysbiosis, reduce gastrointestinal (GI) signs, and target intestinal inflammation.

Objectives: To investigate the efficacy of a synbiotic/IgY supplement on clinical disease activity, intestinal histopathology, gut microbiome, and serum metabolome in dogs with IBD.

Animals: Twenty dogs diagnosed with IBD were randomized to receive a hydrolyzed diet with (n=11) or without (n=9) synbiotic/IgY supplement for 6 weeks. Dogs failing remission after two weeks of primary treatment also received oral cyclosporine at 5 mg/kg.

Methods: Clinical disease activity, endoscopic lesions, histologic inflammation, and frequency of treatment escalation were assessed in both groups before and after treatment. Biologic samples for microbiologic and metabolomic analyses were archived for later analysis.

Results: There were no differences in baseline clinical parameters between groups (p>0.05). Final data analysis showed decreased clinical disease activity and endoscopic scores in both groups after treatment (p<0.05). Five dogs required treatment escalation with four of five dogs maintaining remission until trial completion. There was no difference in histologic inflammation pre‐ versus post‐treatment for either group (p>0.05).

Conclusions: While this study was likely underpowered, the administration of a synbiotic/IgY supplement with hydrolyzed diet over 6 weeks was well tolerated and associated with reduced clinical disease activity and endoscopic scores, like placebo.

119.1.1.1. ¹Royal Veterinary College; ²Westerdijk Fungal Biodiversity Institute

Background: Fungi play a role in the pathogenesis of human inflammatory bowel disease.

Objective: To explore the presence of culturable fungi in the duodenum of dogs with chronic gastrointestinal signs undergoing routine endoscopic examination.

Animals: Forty‐five client‐owned dogs with chronic gastrointestinal signs from our referral hospital.

Methods: Quantitative microbial culture was performed on duodenal juice; samples were cultured on Sabouraud's dextrose agar (30 & 37°C) and modified Dixon's agar (32°C) for 14‐days. Isolates were identified phenotypically, genotypically (ITS‐sequencing), and by Matrix‐Assisted Laser Desorption Ionization‐time Of Flight. Yeast presence was evaluated by cytologic and histopathologic examination of smears and biopsy specimens.

Results: Forty‐five dogs were recruited with chronic inflammatory enteropathy (CIE; n=38), granulomatous colitis (n=2), gastric adenocarcinoma (n=2), duodenal small cell lymphoma (n=1), and idiopathic gastrointestinal hemorrhage (n=2). Fungi were cultured from 14 dogs: M. pachydermatis from eight (CIE [n=7] [along with Candida albicans n=1]; granulomatous colitis [n=1]), and M. sympodialis from another (gastric adenocarcinoma). Five dogs with CIE yielded other yeasts (C. albicans, C. glabrata, Kazachstania slooffiae, K. telluris, Pichia kudriavzevii [syn. C. krusei]). Dogs with yeast growth had significantly higher serum vitamin B12 (p=0.003), canine chronic enteropathy clinical activity index (p=0.034), and dermatologic signs (p=0.043) at the time of isolation versus those with no growth.

Conclusions and Clinical Importance: To our knowledge, this is the first report of isolation of M. pachydermatis, and M. sympodialis spp. from the canine duodenum. Further studies are needed to determine whether their presence has a pathogenic effect and whether antifungal therapy is warranted.

120.1.1.1. ¹College of Veterinary Medicine, Michigan State University; ²School of Veterinary Medicine, University of Wisconsin; ³College of Veterinary Medicine and Biomedical Sciences, Texas A&M University; ⁴Ocean State Veterinary Specialists; ⁵MiraVista Diagnostics; ⁶College of Veterinary Medicine, Midwestern University

Background: It has been suggested that a panel of IgA‐based serologic assays aids in the diagnosis of chronic enteropathy (CE) in dogs, a syndrome which includes conditions such as food‐responsive enteropathy, immunosuppressant responsive enteropathy, and inflammatory bowel disease (also referred to as chronic inflammatory enteropathy). However, it is unclear whether these markers discriminate between CE and other types of primary intestinal disorders.

Objectives: To evaluate a commercially available diagnostic panel that measures serum levels of IgA directed against OmpC (ACA), canine calprotectin (ACNA), and gliadin‐derived peptides (AGA) in dogs with well‐characterized intestinal diseases.

Animals: Thirty‐two dogs with primary intestinal disease.

Methods: Serum ACA, ACNA, and AGA levels were measured in dogs with CE and dogs with other forms of primary intestinal disease (non‐CE population), including bacterial, fungal, parasitic, and neoplastic enteropathies. Serum IgA levels were compared between populations, and sensitivities and specificities were calculated using laboratory‐provided cut‐points.

Results: Twelve of 15 (80.0%) CE dogs and 14 of 17 (82.4%) non‐CE dogs had abnormally high levels of at least 2 markers; these proportions were not significantly different (P=0.99). A serum ACA level ≥15 EU/mL was 80.0% sensitive for CE, but specificity was only 29.4%.

Conclusions and Clinical Importance: The serologic panel was moderately sensitive, but poorly specific for a CE diagnosis. Common treatments for CE, such as dietary modifications and immunosuppressants, might delay appropriate therapy or even be detrimental for many of the patients with non‐CE diseases that commonly showed high ACA, ACNA, and AGA levels.

121.1.1.1. ¹Colorado State University; ²Professor, Colorado State University; ³Nestle Purina

Background: Diet may induce clinical remission in dogs with chronic inflammatory enteropathy (CIE). Elemental diets (ED) made from individual amino acids are highly digestible with minimal antigenic stimulation and can be effective for remission from Crohn's disease in humans.

Hypothesis/Objectives: To evaluate whether exclusive ED feeding would help CIE dogs achieve clinical remission.

Animals: Client‐owned dogs (n=23) not experiencing adequate relief from their CIE were enrolled in this prospective, uncontrolled study.

Methods: A thorough workup including upper and lower gastrointestinal endoscopy and biopsy was performed. CIBDAI, serum chemistry and cytokine concentrations (MILLIPLEX®) were evaluated before and after 2 (T2) and 8 weeks (T3) of ED feeding. Values were compared using a mixed model with P values adjusted for multiple comparisons.

Results: Eight dogs exited the study before T3; one would not eat the diet, 1 required prednisone for unrelated reasons, and 5 had ongoing signs of CIE. At T2, 16/22 dogs (73%) accepting the diet were experiencing adequate relief with median [range] decrease in CIBDAI from 6 [3–12] to 2 [0–9] (P<0.005). All 16 dogs remaining at T3 were experiencing adequate relief with a median CIBDAI of 2 [0–3]. Serum ALP concentrations [mean±SD] were higher at T3 [106 U/L±57]) than before feeding ([50±34]; P<0.005). Differences in serum albumin, cholesterol, and cytokine concentrations were not detected.

Conclusions and Clinical Importance: The ED was palatable for 95% of dogs and successfully induced clinical remission in 73%. Work is ongoing to determine the mechanisms behind the ED's positive effects.

122.1.1.1. ¹College of Veterinary Medicine ,The Ohio State University; ²Internal Medicine Resident, Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University; ³Veterinary Student, Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University; ⁴Associate Professor, Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University; ⁵Professor, Clinical, Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University; ⁶Assistant Professor, Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University

Background: Microbiome‐based fecal evaluations are promising diagnostic tools for assessing feline health. However, given inherent daily changes in the gut microbiome, the ideal sampling frequency is unknown.

Hypothesis: High‐frequency fecal sampling is more effective than weekly sampling at capturing changes in the feline fecal microbiome in response to a new diet.

Animals: This clinical trial was performed in six healthy and sterilized purpose‐bred cats (3 male, 3 female) group‐housed in a research facility.

Methods: All cats were fed an adult commercial wet food diet, at maintenance needs, to which they had no previous exposure. Feces was collected daily for 36 days. The 16s rRNA microbiome analysis (V4 region) was performed using R‐Studio. Pairwise PERMANOVA evaluated changes in composition. A subset of data (once‐weekly samples) was compared to the high‐frequency (daily) dataset to evaluate daily versus weekly fecal sampling effects.

Results: With high‐frequency sampling, the fecal microbiome composition of each cat differed by day (P<0.01). Despite being co‐housed, daily sampling identified distinct microbiome compositions between cats at all time points (P<0.0001). With once weekly sampling, no compositional differences between cats were detected and there were decreased differences over time (sampling week) (P<0.05) for all cats.

Conclusions and Clinical Importance: Daily sampling or pooled samples may more accurately capture fecal microbiome dynamics compared to weekly sampling. Ongoing analysis will evaluate taxa contributing to daily versus weekly microbiome changes and compare sampling frequencies (e.g., twice weekly) to daily sampling to establish appropriate sampling frequencies for future microbiome studies.

123.1.1.1. ¹Gastrointestinal Laboratory, Department of Small Animal Clinical Sciences, Texas A&M University; ²University of São Paulo; ³Veterinary Nutrology Service, Teaching Veterinary Hospital, School of Veterinary Medicine and Animal Science, University of São Paulo; ⁴Grandfood Industry and Commerce LTDA (Premier Pet)

Background: Exocrine pancreatic insufficiency (EPI) is characterized by insufficient synthesis and secretion of enzymes by the exocrine pancreas. This in turn leads to maldigestion and intestinal dysbiosis, and may or may not lead to disturbances in intestinal integrity. Disturbances of the intestinal integrity can be assessed by evaluation of intestinal permeability (IP) and have been reported in humans with EPI. Iohexol has recently been described as a marker for IP in healthy dogs and dogs with chronic enteropathy.

Objectives: To compare IP between dogs with EPI and healthy control dogs receiving the same diet.

Animals: Twenty dogs were selected for this study: 10 dogs with EPI, currently showing no clinical signs and receiving pancreatic enzyme replacement therapy (PERT) and 10 healthy control dogs. Study and control group were matched according to size and age

Methods: Dogs received the same extruded commercial diet. Food was withheld for 12 hours before oral administration of 2.0 mL/kg of iohexol. Blood samples were collected from the jugular vein two hours after iohexol administration, centrifuged, and frozen

Results: Serum iohexol concentrations were significantly higher (p=0.0147) in dogs with EPI on PERT (median: 60.6 μg/mL range: 21.0–186.3) compared to healthy control dogs (median: 29.0 μg/mL range: 20.7–56.7).

Conclusions: This study suggests that despite receiving PERT, showing no clinical signs, and maintaining normal body condition, dogs with EPI show an increased IP when compared to healthy control dogs. An increase in IP may contribute to the gastrointestinal dysfunctions in dogs with EPI.

124.1.1.1. ¹University of California, Davis; ²Veterinary Specialty Hospital; ³Flagstaff Veterinary Internal Medicine Consulting (FLG VIM‐C); ⁴Gastrointestinal Laboratory, Texas A&M University

Background: Alterations of the fecal metabolome have previously been described in cats with chronic enteropathy (CE). Characterizing the serum metabolome could provide further information on alterations of metabolic pathways and may identify diagnostic or therapeutic targets.

Hypothesis/Objectives: To characterize the serum metabolome of cats with CE before and after treatment.

Animals: Serum samples from 14 healthy cats and 26 cats with CE (13 cats with inflammatory bowel disease (IBD), 13 cats with small cell intestinal lymphoma (SCL)), before and 4 weeks after treatment with corticosteroids and/or chlorambucil.

Methods: Untargeted metabolomic analysis of serum using high performance chromatography–tandem mass spectroscopy. Differences between healthy and CE cats were evaluated using the Wilcoxon rank sum test, adjusted using false discovery rate and expressed as q‐values. Statistical significance was set at q<0.05.

Results: Metabolomics analysis identified 1,082 named metabolites, of which 130 were significantly different between cats with CE and healthy controls at baseline. Random forest analysis revealed a predictive accuracy of 85% for differentiating controls from cats with CE. Metabolic pathways found to be significantly altered included polyunsaturated fatty acid, phospholipid, amino acid, and indole metabolism. Several metabolites were found to be significantly different between cats with IBD versus SCL, including 3,4‐dihydroxybenzoic acid, malic acid, arginine, and several fatty acids. Metabolic profiles of cats with CE after 4 weeks of treatment were similar to healthy control cats.

Conclusion: FCE cats exhibited disruptions in several metabolic pathways, some differential for IBD and SCL. Metabolic profiles partially recovered following treatment.

125.1.1.1. ¹Veterinary Student, College of Veterinary Medicine, Veterinary Clinical Sciences, The Ohio State University; ²Resident and PhD Student, College of Veterinary Medicine, The Ohio State University; ³Internal Medicine Resident, Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University; ⁴Assistant Professor, Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University

Background: Fecal microbiota transplantation (FMT) is the delivery of fecal material from a healthy donor into a diseased recipient to confer a health benefit. FMT is routinely delivered as a slurry via enema or nasogastric tube, but can also be encapsulated for convenient oral administration. The microbial impacts of processing feces for encapsulation are unknown.

Hypothesis/Objectives: We aimed to investigate if standardized fecal processing into a 10% glycerol FMT slurry alters the microbial community structure. We hypothesized that microbial community structure would not vary between processing steps.

Animals: Feces from four screened healthy canine fecal donors at The Ohio State University Companion Animal Fecal Bank were utilized for this study.

Methods: Batches of canine feces underwent a standardized FMT processing protocol. Samples from each donor, across multiple batches, were pulled from specific steps during processing and underwent DNA extraction and Illumina sequencing of the V4 region of the 16S rRNA gene. Amplicon sequence analysis was performed using MOTHUR, DADA2 and R‐Studio.

Results: Analysis revealed significant differences in alpha diversity (observed, Shannon index) and beta diversity (NMDS, Jaccard index) measures between donors. Within an individual donor, no significant differences were observed between processing steps (AMOVA). Sequence analysis using amplicon sequence variants (ASVs) and operational taxonomic units (OTUs) both yielded similar results.

Conclusions and Clinical Importance: Standardized FMT processing for encapsulation does not significantly alter the microbial composition of feces from healthy canine donors, further promoting the practicality of oral FMT administration in canine medicine. Additional studies evaluating microbial viability are required.

126.1.1.1. ¹Clinic of Small Animal Medicine, Centre for Clinical Veterinary Medicine, LMU Munich, Munich, Germany; ²Gastrointestinal Laboratory, Department of Small Animal Clinical Sciences, Texas A&M University, College Station, TX, USA; ³Clinic for Small Animal Internal Medicine, Vetsuisse Faculty, University of Zurich, Zurich, Switzerland

Background: Acute hemorrhagic diarrhea syndrome (AHDS) is characterized by the occurrence of acute bloody diarrhea. Histopathologic examination of intestinal biopsies from dogs with AHDS typically reveals necrotizing enteritis. Therefore, the presence of intestinal barrier dysfunction is suspected, which might be associated with clinical severity.

Hypothesis: Dogs with AHDS have increased intestinal permeability, which is associated with clinical severity.

Animals: Thirty client‐owned dogs with AHDS and 25 healthy control dogs (HC).

Methods: Prospective clinical trial. The AHDS‐index was evaluated in all dogs at the day of presentation. On the same day, dogs were administered 2 ml/kg of iohexol orally. After two hours, serum was obtained to measure iohexol concentration (SIC) by enzyme‐linked immunosorbent assay. The same procedure was performed in the HC group. A Mann‐Whitney test was used to compare SIC between AHDS and HC. The association between the AHDS‐index and SIC was evaluated using Spearman's rank correlation.

Results: SIC were significantly higher (p=0.002) in AHDS (median: 46 μg/ml; min–max: 9–246) compared to HC (30 μg/ml; 11–57). There was a significant positive correlation between AHDS‐index and SIC (r_s=0.4; p=0.020). Furthermore, SIC was significantly higher (p=0.001) in dogs with severe (AHDS‐index ≥9) (median: 80 μg/ml; min–max: 25–246) compared to dogs with mild to moderate disease (AHDS‐index <9) (31 μg/ml; 9–54).

Conclusions and Clinical Importance: Dogs with AHDS have significantly increased intestinal permeability as assessed by iohexol. In addition, a significant correlation between clinical severity and permeability was detected.

127.1.1.1. ¹Department for Small Animals, College of Veterinary Medicine, University of Leipzig, Leipzig, Germany; ²Clinical Instructor Pathology, Institute of Animal Pathology, Vetsuisse Faculty, University of Bern, Switzerland; ³Research Scientist, Institute of Immunology/Molecular Pathogenesis, Center for Biotechnology and Biomedicine, College of Veterinary Medicine, University of Leipzig, Leipzig, Germany; ⁴Clinical Scientist, Institute of Pathology, Faculty of Veterinary Medicine, University of Leipzig, Leipzig, Germany; ⁵Laboratory for Veterinary Pathology, Munich, Germany; ⁶Professor, Veterinary Immunology, Institute of Immunology/Molecular Pathogenesis, Center for Biotechnology and Biomedicine, College of Veterinary Medicine, University of Leipzig, Leipzig, Germany; ⁷Institute of Anatomy, College of Medicine, Leipzig University, Leipzig, Germany; ⁸Professor, Histology and Embryology, Institute of Veterinary Anatomy, College of Veterinary Medicine, University of Leipzig, Leipzig, Germany; ⁹Professor, Small Animal Internal Medicine; Director, Gastrointestinal Laboratory, Department of Small Animal Clinical Sciences, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University; ¹⁰Professor, Small Animal Internal Medicine, Department for Small Animals, College of Veterinary Medicine, University of Leipzig, Leipzig, Germany

Background: Diagnosis of chronic inflammatory enteropathies (CIE) in cats and differentiation from intestinal lymphoma using currently available diagnostics is challenging. The S100/calgranulins appear as useful non‐invasive biomarkers for canine CIE, with increased intestinal expression, but have not been evaluated in cats.

Hypothesis: We hypothesized that the S100/calgranulins are differentially expressed between feline CIE and intestinal lymphoma, and correlate with histological and/or clinical disease severity.

Animals: Retrospective case‐control study including patient data and gastrointestinal tissue from 16 cats with CIE and 8 with intestinal lymphoma.

Methods: Gastrointestinal tissue biopsies were immunohistochemically stained using species‐specific polyclonal α‐S100A8/A9 and α‐S100A12 antibodies. Epithelial and lamina propria S100A8/A9⁺ and S100A12⁺ cells per 10,000 μm² were compared between groups of cats and tested for a correlation with clinicopathologic abnormalities, clinical disease severity (feline chronic enteropathy activity index [FCEAI] score), and histologic lesions.

Results: S100A8/A9⁺ and S100A12⁺ cells were detected in all gastrointestinal segments, without significant differences between CIE and lymphoma. Segmental inflammatory lesions were moderately to strongly correlated with increased S100/calgranulin‐positive cell counts. Increased ileal S100A12⁺ cell counts correlated with hypocobalaminemia (P=0.0229). Hypofolatemia was linked to higher duodenal S100A8/A9⁺ cell counts. FCEAI scores were higher in cats with lymphoma than CIE (P=0.0101) and were linked to higher duodenal S100A12⁺ cell counts (P=0.0385).

Conclusions and Clinical Importance: These findings suggest a role of the S100/calgranulins in the pathogenesis of the spectrum of feline chronic enteropathies and emphasize the diagnostic value of B vitamin determination and clinical disease grading.

128.1.1.1. Cornell University

Background: Dietary modification can induce clinical remission in dogs with chronic enteropathy. Palatability impacts dietary therapy in dogs with protein losing enteropathy (PLE).

Objectives: To compare the ability of isocaloric diets composed of (A) hydrolyzed fish, rice starch and fish oil (HF), (B) HF plus prebiotics, turmeric and high cobalamin (HF+), or (C) highly digestible non‐hydrolyzed mixed protein source diet (Placebo) to resolve clinical signs and maintain serum B12 in dogs with chronic enteropathy. To evaluate palatability and weight gain in dogs with PLE.

Animals: Thirty‐one client owned dogs with chronic enteropathy: 23 non‐PLE, 8 PLE.

Methods: Randomized, blinded, placebo‐controlled clinical trial. Each diet was fed for 2 wks, with responders continuing for 12 wks. Non‐responders crossed over to another diet for 12 wks. Clinical response was determined by CCEAI and stool consistency, with long term follow‐up at 26 wks. Concurrent medications were allowed in PLE.

Results: Nineteen of 23 dogs with non‐PLE chronic enteropathy responded solely to their initial diet, with no difference between diets (P<0.05. Table 1). Four dogs crossed over, with sustained remission in 18/18 at 6 months (Table 1). Serum B12 was maintained by diet in chronic enteropathy. Dogs with PLE fed HF or HF+ found it palatable and gained weight, with 2/8 responding to diet alone. Table 1

Conclusions and Clinical Importance: Changing diet, independent of antigen restriction or supplementation, was associated with long‐term clinical remission in dogs with chronic enteropathy, and a subset of PLE. Further study is required to determine the basis of this clinical response.

130.1.1.1. ¹Iowa State University; ²Veterinary Pathology, Iowa State University; ³Population Medicine and Diagnostic Sciences, Cornell University; ⁴Clinical Sciences, Auburn University

Background: Immune thrombocytopenia (ITP) is a common hemostatic disorder of dogs leading to significant morbidity and mortality, often from side effects of immunosuppression. In ITP, autoantibodies target platelets for macrophage clearance. Human studies demonstrate autoantibodies also trigger platelet desialylation. Desialylation marks platelets for clearance, thus preventing desialylation could lead to new, safer treatment strategies. Studying desialylation in canine ITP requires a platelet desialylation assay.

Objective: Develop a flow cytometric assay to detect exposed platelet β‐galactose residues as a marker of desialylation based on binding of FITC‐labeled Ricinus communis agglutinin (RCA).

Animals: 17 healthy, client‐owned dogs recruited from a veterinary teaching hospital.

Methods: Platelet‐rich plasma (PRP) was isolated from EDTA‐anticoagulated blood. Ex vivo desialylated platelets were generated by neuraminidase C treatment (positive control); β‐lactose incubated PRP served as negative control. Controls and native PRP were labeled with RCA‐FITC and desialylation was quantified flow cytometrically by mean fluorescence intensity/forward scatter (MFI/FSC). Following optimization, analytic precision, performance in serially diluted PRP, and sample stability were assessed.

Results: RCA MFI/FSC remained stable for 48 hr post‐blood collection at 25°C (p=0.06) and 4°C (p=0.24). Precision was acceptable with coefficient of variation=4.8%. RCA MFI/FSC was not influenced by platelet counts ranging from 200,000–10,000 platelets/μL (p=0.14).

Conclusions and Clinical Importance: This flow cytometric assay detects desialylated platelets with good analytic performance through a wide range of platelet counts. This assay will enable pathophysiologic investigations of platelet desialylation in canine ITP and set the stage for novel therapeutics.

131.1.1.1. ¹School of Veterinary Medicine, University of California‐Davis, Davis, CA, USA; ²Technician, Veterinary Medical Teaching Hospital, School of Veterinary Medicine, University of California; ³Associate Professor, Medicine and Epidemiology, School of Veterinary Medicine, University of California

Background: TPE has emerged as adjunctive therapy for severe IMHA in dogs but has not been compared to conventional therapy. We report the efficacy of adjunctive TPE to comparator dogs managed medically.

Hypothesis/Objectives: Effective TPE delivery improved survival in dogs with severe IMHA.

Animals: 34 client‐owned dogs with primary or secondary IMHA (HCT <20%, spherocytosis, reticulocytosis, transfusion dependency, known outcome) treated with adjunctive TPE at UC Davis between 2010–2020 were compared to 72 similarly affected, randomly selected dogs managed medically over the same interval. Both cohorts were censored for euthanasia within 3 days for financial concerns yielding 32 TPE and 68 Control dogs.

Methods: 100 TPE treatment were delivered on a Spectra Optia. Controls were managed with conventional immunosuppression. Kaplan‐Meier (Logrank) survival analysis was compared between all TPE, dogs with ≥3TPE treatments, and Controls at 30 days.

Results: TPE dose was 1.4±0.3 plasma volumes (PV)/session. 4/6 non‐surviving dogs received ≤2 TPE treatments (2.0±0.6 PV); 28 dogs received ≥3TPE treatments delivering 4.2±0.6 PV/dog. 30‐day survival for all TPE and medically managed dogs was 76.5% and 57.5%, respectively (P=0.06). When censored for financial‐directed euthanasia, survival for TPE and Control cohorts was 81.3% vs 61.8%, respectively (P=0.058, Fig). Survival for dog receiving ≥3TPE (>4.0 PV delivered) was 92.9% vs Control (P=0.003). Hazard ratio for Control/≥3TPE was 6.4.

Conclusions: Delivery of TPE (>4 total PV) was associated with improved survival in dogs with IMHA compared to medical management.

132.1.1.1. ¹Ghent University; ²Small Animal Department, Ghent University; ³Faculty of Medicine and Health Sciences, Heart Centre, Ghent University

Background: Activated clotting times (ACT) are measured to titrate anticoagulant dose during intravascular procedures. Point‐of‐care (POC) analyzers could improve test practicality and accuracy compared to manual tests, but a reference interval (RI) for dogs is lacking.

Objectives: To establish a RI for ACT in adult dogs using a POC analyzer, to quantify longitudinal intra‐subject variability and to investigate device reliability and inter‐device agreement.

Animals: Forty‐two healthy dogs, including 28 client‐owned and 14 research dogs.

Methods: Prospective measurements were performed on fresh venous blood samples using the i‐STAT 1 (Abbott) POC analyzer and kaolin cartridges. The RI was determined using the Robust method. Intra‐subject within‐ and between‐day variability were quantified between baseline and 2 hour (n=8) or 48 hour (n=10). Device reliability and inter‐device agreement were studied by duplicate measurements (n=12) on identical devices.

Results: Mean, lower and upper reference limits for ACT were 92.9±9.1 seconds, 74.4 seconds (90% confidence interval (CI): 70.1–78.3 seconds) and 111.2 seconds (90% CI: 107.5–115.2 seconds), respectively. Coefficients of variation of intra‐subject within‐day (8.1%) and between‐day (10.4%) variability were low, yet resulted in a small but significant between‐day difference (P=0.020). Device reliability assessed by intraclass correlation coefficient was excellent (0.99, 95% CI 0.997–1.000) and measurements between devices were not significantly different (P=0.593).

Conclusions and Clinical Importance: The POC analyzer has an excellent reliability and test simplicity, which favors its use over manual ACT assays.

133.1.1.1. Ontario Veterinary College

Background: The previously undescribed Primary Hemostasis Components (PHC) 1 and 2 on the PFA‐200 analyzer are for additional variables that normalize platelet function to a transferable, intuitive, percentage‐based system. Closure curves may also be visually inspected and may predict closure vs non‐closure in animals with flow obstruction.

Objectives/Hypothesis: To investigate the PHC parameters and curve analysis for determination of platelet function in cats.

Animals: 20 prospective healthy cats between 2–10y with normal physical examination, CBC and biochemistry; 34 healthy retrospective blood donor cats, and 14 retrospective cats receiving therapy with clopidogrel.

Methods: Normal ranges were determined for CT, PHC(1), and PHC(2). ROC curves were created to determine utility of these values to assess clopidogrel effect. Visual analysis of the closure curves was performed.

Results: There was no difference in CT for healthy cats and blood donors. Reference interval for CT was 20–90s (COL/ADP) and 36–119s (P2Y). RI for PHC(1) was 51.1s–145.1s (COL/ADP) and 41.5s–165.6s(P2Y). RI for PHC(2) was 94.2s–105.9s (COL/ADP) and 87.2s–105.2s(P2Y).

AUCROCs for clopidogrel effect were excellent with no difference between tests. Optimal cut‐offs for clopidogrel response were >76s (COL/ADP CT), >211s (P2Y CT); <74.1s (COL/ADP PHC(1)), <72.1s (P2Y PHC(1)); and <96.6s (COL/ADP PHC(2)),<76.0s (P2Y PHC(2)).

Six classes of closure curve were identified, of which two were associated with non‐closure.

Conclusions and Clinical Importance: PHC(1) and PHC(2) perform comparably to CT and provide an expression of platelet function that may be compared between cartridges, and laboratories. Closure curves may be used to predict non‐closure.

134.1.1.1. Ontario Veterinary College

Background: The Platelet Function Analyzer‐200 (PFA‐200) allows determination of clopidogrel effect in cats, however, samples must be analyzed within four hours of collection making routine monitoring infeasible. Samples in dogs and pilot data in cats have demonstrated feasible analysis after shipping and storage.

Objectives/Hypothesis: To validate the shipment of feline blood samples for analysis on the PFA‐200 24h after collection.

Animals: 20 healthy cats between 2–10y with normal physical examination, CBC and biochemistry, and four cats receiving clopidogrel therapy.

Methods: Blood was collected by jugular venipuncture. A CBC was analyzed at point‐of‐care to confirm adequate platelet counts. Samples were divided and analyzed in duplicate at point of‐care with the COL/ADP and P2Y cartridges; and packaged in insulated containers with ice packs, transported to a distant location and then back the following day, and analyzed 24h after initial collection. Results of point‐of‐care and shipped samples were compared.

Results: When used to categorize cats as clopidogrel responsive vs resistant, agreement was 0.85 (COL/ADP) and 0.95 (P2Y). Bland‐Altman analysis showed minimal outliers and no bias. Median closure time was significantly different between point‐of‐care (COL/ADP 51.5s, P2Y 50.5s) and shipped samples (COL/ADP 78.75s, P2Y 65.5s) (p<0.001). The occurrence of flow obstructions was not significantly different between groups (P=0.525).

Conclusions and Clinical Importance: Feline blood may be shipped and analyzed up to 24h after collection with the PFA‐200 with comparable results to point‐of‐care testing. This allows practical, remote determination of clopidogrel effect in cats.

135.1.1.1. Ontario Veterinary College

Background: Determining platelet function in cats may be useful for monitoring clopidogrel effect. Plateletworks (PW) exposes citrated blood to a platelet agonist, and determines “percent aggregation” based on platelet counts obtained with a hematology analyzer. PW was developed for impedance analyzers and has been validated for the Siemens ADVIA 2120 optical analyzer, but not for point‐of‐care optical analyzers.

Objectives/Hypothesis: To validate Plateletworks for the IDEXX ProCyte Dx analyzer and visual slide review.

Animals: 20 healthy cats between 2–10y and 5 cats receiving clopidogrel therapy.

Methods: Jugular blood samples were aliquoted into ADP agonist and EDTA tubes, divided, and analyzed simultaneously on ADVIA 2100 and ProCyte Dx analyzers at 10min, 30min, 45min, 1h. Slides were prepared and degree of clumping scored by a blinded observer. Percent aggregation was the primary outcome.

Results: Association of aggregation was 0.8166 between analyzers. Median values were significantly different (p=0.0004), but Bland‐Altman analysis showed minimal outliers and no bias. Maximal coefficient of variation was lower for ProCyte Dx than ADVIA 2120. When classifying as non‐, low‐ or high‐aggregators, Cohen's Kappa was 0.714. Correlation was substantial between both analyzers and slide review. There was no significant change in aggregation within the first hour, however there was a trend to increase at 30 min, followed by a slight decline. Cats on clopidogrel were categorized as responsive on both analyzers.

Conclusions and Clinical Importance: Plateletworks may be used in clinics with ProCyte Dx analyzers or visual slide review. The timespan for analysis is longer than previously reported.

136.1.1.1. ¹Hokkaido University; ²Rakuno Gakuen University

Background: To our knowledge, there are no reports summarizing the therapeutic effects of splenectomy in dogs with non‐regenerative immune‐mediated anemia (NRIMA).

Hypothesis/Objectives: Evaluation of the therapeutic efficacy of splenectomy as an alternative to immunosuppressive therapy.

Animals: Twenty client‐owned dogs with NRIMA who had splenectomies performed at Hokkaido University Veterinary Teaching Hospital between 2015 and 2021.

Methods: Retrospective case series.

Results: The median age of the included dogs was 11.5 years (range 3–15 years). The reasons for splenectomy were:

• 5

  No response to immunosuppressive therapy (n=13)

• 6

  Temporary response to immunosuppressive therapy but subsequent loss of response (n=5)

• 7

  Suspected complications of immunosuppressive therapy (n=1)

• 8

  First‐line treatment (n=1)

Nineteen dogs were treated with immunosuppressive therapy prior to splenectomy, and the median duration of treatment was 113 days (range 50–295 days). Two dogs died within 7 days of surgery, and the remaining 18 dogs showed an increase in reticulocyte count at a median of 16 days (range 1–49 days) after splenectomy, with a median maximum reticulocyte count of 137,900/μl (range 71,700–368,200/μl). During the study period, seven dogs had complete remission of anemia, eight dogs had persistent mild to moderate anemia but could be maintained without blood transfusion, and three dogs had worsening non‐regenerative anemia that required therapeutic intervention.

Conclusions and Clinical Importance: Fifteen out of 20 dogs (75%) improved after splenectomy and decreased their dependency on transfusions, suggesting that splenectomy is an effective treatment for NRIMA.

137.1.1.1. ¹Ontario Veterinary College; ²University College Dublin

Background: Measurement of rivaroxaban‐specific anti‐Xa activity is recommended for monitoring rivaroxaban efficacy in veterinary medicine, but this assay is performed only at specialized reference laboratories. Accurate detection of rivaroxaban effect using other hemostatic tests would make therapeutic monitoring more timely and accessible.

Objective: To assess the results of rivaroxaban‐specific anti‐Xa activity with prothrombin time (PT), partial thromboplastin time (PTT), fibrinogen concentration, tissue‐factor and kaolin‐activated thromboelastography (TEG) and thrombin generation (TG) activity in hypercoagulable dogs.

Animals: 9 client‐owned dogs diagnosed with hypercoagulability and/or thromboembolic disease and prescribed rivaroxaban, recruited from the OVC‐HSC population from August 2020 to November 2021.

Methods: Prospective clinical trial. Blood samples were collected by direct jugular venipuncture prior to treatment, then 1‐week and 1 to 3‐months after initiation of therapy. PT, PTT, fibrinogen, anti‐Xa activity, TEG, and TG were performed at each visit (3 hours after rivaroxaban dosing).

Results: The average rivaroxaban dose was 1.4 mg/kg/day. There was a strong correlation between the anti‐Xa assay and PT (r=0.806, p=0.0002). Anti‐Xa was strongly positively correlated with TG variables ttpeak (r=0.819, P=0.0001) and lag time (r=0.746, P=0.0009), and negatively correlated with TG variables ETP (r=‐0.568, P=0.022) and peak (r=‐0.706, P=0.002). There was a poor correlation between anti‐Xa assay and PTT, fibrinogen, and all TEG variables.

Conclusion and Clinical Importance: Prothrombin time is strongly correlated with rivaroxaban‐specific anti‐Xa activity and may be a convenient method to monitor dogs receiving rivaroxaban therapy.

138.1.1.1. ¹College of Veterinary Medicine, Purdue University; ²Associate Professor, Small Animal Internal Medicine, Veterinary Clinical Services, College of Veterinary Medicine, Purdue University; ³Assistant Professor of Veterinary Clinical Pathology, Comparative Pathobiology, College of Veterinary Medicine, Purdue University; ⁴Director, Clinical Trials and Department of Epidemiology, Veterinary Administration, College of Veterinary Medicine, Purdue University

Background: Storage of canine packed red blood cells (pRBC) leads to time‐dependent lesions which may increase risk of transfusion reactions. Oxidative stress is a contributor of storage lesions, but supplementation of antioxidants in canine pRBC has not been investigated.

Objective: Describe storage and oxidative lesions in canine pRBC during routine storage when supplemented with saline (control, Group 1), N‐acetylcysteine (NAC) and ascorbic acid (AA; Group 2), and AA and a‐tocopherol (VE; Group 3)

Hypothesis: Storage and oxidative lesions in canine pRBC [measured by glutathione (GSH), thiobarbituric acid reactive substances, and flow cytometry for intraerythrocytic reactive oxygen species (ROS)] will be decreased with antioxidant supplementation.

Methods: Nine leukoreduced units of canine pRBC were aseptically separated into three aliquots (Groups 1, 2, 3). Antioxidants were supplemented on Day 1 after baseline samples collected. Additional samples were collected on days 7, 28, and 42. Units were collected in 3 batches, with assays performed at the end of each storage period. Type 3 tests of fixed effects compared the impact of group and time on each measurement.

Results: All groups showed storage lesions and GSH depletion by day 42 compared to baseline, regardless of antioxidant supplementation. Intraerythrocytic ROS accumulation was lower in Group 3 (AA & VE) compared to other groups at all time points after baseline (p<0.0001).

Conclusions and Clinical Importance: Supplementation of canine pRBC with AA and VE reduced oxidative stress but not storage lesions. Future studies should evaluate the clinical use and incidence of transfusion reaction with AA/VE‐supplemented pRBC.

139.1.1.1. ¹Michigan State University; ²Rotating Intern, College of Veterinary Medicine, Michigan State University; ³Veterinary Student, College of Veterinary Medicine, Michigan State University; ⁴Assistant Professor, College of Veterinary Medicine, Michigan State University; ⁵Professor, College of Veterinary Medicine, Mississippi State University; ⁶Associate Professor, College of Veterinary Medicine, Mississippi State University

Background: Evidence supporting different therapeutic protocols for non‐associative immune‐mediated hemolytic anemia (na‐IMHA) is weak.

Hypothesis/Objectives: Investigate the impact of drug selection on outcome in na‐IMHA.

Animals: 242 client‐owned dogs.

Methods: Multi‐institutional retrospective study (2015–2020). Immunosuppressive efficacy was determined by time to PCV stabilization, duration of hospitalization, and disease relapse and mortality rates. Anti‐thrombotic efficacy was determined by presence of clinically suspected or confirmed thrombosis. Data was analyzed using mixed model linear and logistic regression.

Results: Use of corticosteroids alone versus a multi‐agent protocol had no significant effect on time to PCV stabilization (p=0.55), duration of hospitalization (p=0.13), or mortality (p=0.58): mortality with 1 immunosuppressive drug was 17.6%, and with 2 drugs was 28.3%. A higher rate of relapse (p=0.04, OR: 3.97) during follow‐up (median 31 days) was detected in dogs receiving a corticosteroid alone (11.3%) compared to multiple agents (3.1%). When comparing specific drug protocols, there was no significant effect on time to PCV stabilization (p=0.31), relapse (p=0.44), or mortality (p=0.08). Duration of hospitalization was significantly longer, by 1.8 days, for the corticosteroid with mycophenolate group (p=0.02). Use of clopidogrel versus a multi‐agent protocol had no effect on development of confirmed (p=0.41) or suspected (p=0.37) thromboses.

Conclusions and Clinical Importance: Corticosteroids alone may be suitable for the initial management of na‐IMHA, although an increased risk of relapse may be noted. Use of multiple anti‐thrombotic agents does not reduce risk of thrombosis.

140.1.1.1. ¹Oregon State University; ²Assistant Professor, Carlson College of Veterinary Medicine, Oregon State University

Background: Dogs with congenital portosystemic shunts commonly have hematological abnormalities. Unlike dogs with extrahepatic portosystemic shunt (EHPSS), dogs with intrahepatic portosystemic shunt (IHPSS) are predisposed to gastrointestinal ulceration and bleeding.

Objective: To compare hematological parameters between dogs with IHPSS and EHPSS.

Animals: Thirty‐two dogs with EHPSS and 23 dogs with IHPSS.

Methods: Retrospective study of client‐owned dogs with a confirmed congenital portosystemic shunt. Erythrogram parameters and history of blood transfusion prior to shunt closure were recorded. An unpaired Student t‐test was used to compare values between dogs with IHPSS and EHPSS.

Results: Twelve IHPSS dogs (52%) and 3 EHPSS dogs (9%) had a hematocrit below the reference interval (<37%). Two IHPSS dogs and no EHPSS dogs had a history of a blood transfusion. Compared to EHPSS dogs, IHPSS dogs had a significantly lower hemoglobin (P=0.0004; mean difference +/‐ SEM: 2.4 +/‐ 0.6 g/dL), hematocrit (P=0.004; 5.1 +/‐ 1.7%), mean cell volume (P<0.0001; 8.3 +/‐ 2.0 fL), mean cell hemoglobin concentration (P=0.001; 1.9 +/‐ 0.5 g/dL), and plasma protein (P=0.04; 0.4 +/‐ 0.01 g/dL). Reticulocyte count was measured in 16 IHPSS dogs and 16 EHPSS dogs, and no significant difference was found between groups (P=0.7).

Conclusions: IHPSS dogs had lower hematological values than EHPSS dogs. In a few IHPSS dogs, a blood transfusion was required for treatment of anemia. These differences may be partly attributable to blood loss secondary to gastrointestinal bleeding in dogs with IHPSS.

141.1.1.1. ¹North Carolina State University; ²Assistant Professor in Emergency and Critical Care, Department of Clinical Sciences, North Carolina State University; ³Research Specialist, Veterinary Hemostasis and Neuro‐Oncology, Department of Clinical Sciences, North Carolina State University; ⁴Assistant Professor in Emergency and Critical Care, Surgical and Radiological Sciences, North Carolina State University; ⁵Clinical Assistant Professor in Emergency and Critical Care, Department of Clinical Sciences, North Carolina State University

Background: The platelet inhibitory effect of clopidogrel can vary between individuals. A proprietary modified thromboelastography protocol (TEG Platelet Mapping [TEG‐PM]) can be used for therapeutic monitoring but is not widely available. Thrombin generation (TG) might offer a viable monitoring alternative.

Hypothesis/Objectives: Evaluate TG for antiplatelet monitoring in dogs administered clopidogrel.

Animals: 6 healthy mix‐breed dogs.

Methods: 2 mg/kg clopidogrel was administered orally once daily for 7 days. Blood was collected for TEG‐PM and TG on day 0 and day 7. Platelet inhibitory effect was assessed by percentage platelet inhibition (% inhibition) on TEG‐PM and with the TG variables lag time, peak, and endogenous thrombin potential. Comparisons between day 0 and day 7 results were made using a paired student's t‐test. Data are presented as mean +/‐ SD.

Results: Significant changes in % inhibition were noted on TEG‐PM [day 0: 57.7 +/‐ 27.3%, day 7: 98.6 +/‐ 4.0%, P=0.02)], but without significant changes in TG variables: lag time (day 0: 1.8 +/‐ 0.2 min, day 7: 1.8 +/‐ 0.2 minutes, P=0.42); peak (day 0: 76.4 +/‐ 6.9 nM, day 7: 72.7 +/‐ 10.4 nM, P=0.49); and endogenous thrombin potential (day 0: 399.2 +/‐ 26.9, day 7: 392.2 +/‐ 32.4, P=0.49).

Conclusions and Clinical Importance: No significant changes in thrombin generation variables were detected in healthy dogs administered clopidogrel, despite concomitant evidence of platelet inhibition with TEG Platelet Mapping. Thrombin generation does not appear to be a useful antiplatelet therapeutic monitoring tool in dogs administered clopidogrel.

142.1.1.1. ¹Hokkaido University; ²Specially Appointed Associate Professor, Translational Research Unit, Veterinary Teaching Hospital, Graduate School of Veterinary Medicine, Hokkaido University; ³Assistant Professor, Veterinary Teaching Hospital, Graduate School of Veterinary Medicine, Hokkaido University; ⁴Assistant Professor, Laboratory of Veterinary Surgery, Graduate School of Veterinary Medicine, Hokkaido University; ⁵Associate Professor, Laboratory of Advanced Veterinary Medicine, Graduate School of Veterinary Medicine, Hokkaido University; ⁶Assistant Professor, Laboratory of Veterinary Internal Medicine, Department of Clinical Sciences, Faculty of Veterinary Medicine, Hokkaido University; ⁷Associate Professor, Laboratory of Veterinary Internal Medicine, Department of Clinical Sciences, Faculty of Veterinary Medicine, Hokkaido University; ⁸Professor, Laboratory of Veterinary Internal Medicine, Department of Clinical Sciences, Faculty of Veterinary Medicine, Hokkaido University

Background: Some non‐regenerative immune‐mediated anemia (NRIMA) patients resist immunosuppressive therapy. Our preliminary study indicated that splenectomy improved non‐regenerative anemia for these cases.

Hypothesis/Objectives: The spleen releases humoral factors that suppress hematopoiesis leading to non‐regenerative anemia.

Animals: Fifteen dogs with NRIMA and three healthy control dogs.

Methods: Protein expression and gene expression were analyzed. Four serum samples pre‐ and post‐splenectomy were used in liquid chromatography with tandem mass spectrometry. Protein identification and quantification were performed by Scaffold DIA and Prosit (database UP000002254). Differentially expressed proteins were defined using paired t‐tests with p‐value <0.05 and fold change >1.5 or <0.7. RNA was extracted and subjected to RNA sequencing (Illumina NextSeq 500). Sequencing reads (1M single‐Read/sample) were aligned to the canine reference genome assembly CanFam3.1. Gene expression was measured, and differentially expressed genes (DEG) identified using DESeq2 with adjusted p‐values <0.05.

Results: Of 505 proteins detected, 59 (including haptoglobin and ficolin 1) were identified as differentially expressed between pre‐ and post‐splenectomy dogs. Pathway analysis for these revealed the enrichment of 16 pathways including complement activation. A total of 1389 DEGs were identified between NRIMA and healthy dogs; 681 were downregulated including EPB41, and 708 were upregulated including S100A12, S100A8, and S100A9 in NRIMA.

Conclusions and Clinical Importance: We obtained lists of proteins and genes that might be involved in suppressing normal hematopoiesis before splenectomy, including those that regulate complement activation. These findings could help understand the pathological mechanisms of NRIMA.

144.1.1.1. ¹Veterinary Hospital, North Carolina State University; ²NCSU Microbiome Methods in Health and Disease Laboratory; ³UNC Microbiome Core Facility

Background: Bacterial cholangitis is a common disease of cats yet existence of a bile microbiome in cats is unknown. This knowledge gap may limit our understanding of the pathogenesis of hepatobiliary diseases in cats.

Objectives: To establish 1) if healthy cats harbor a core microbiota in gallbladder bile and 2) any relationship between results of conventional bile culture, cytology, and 16S rRNA gene amplicon sequencing (16S sequencing).

Animals: 43 cats with suspected hepatobiliary disease and 18 control cats from a local animal control facility.

Methods: Case‐control study. Bile was collected via percutaneous ultrasound‐guided cholecystocentesis from cats with suspected hepatobiliary disease and post‐mortem from control cats having concurrent histopathologic examination to exclude underlying disease. Bile cytology, aerobic/anaerobic culture, and 16S sequencing were performed.

Results: Both bile culture and 16S sequencing results were attained for 40 cats with suspected hepatobiliary disease and 18 controls. A positive culture was obtained from 10 cats with suspected hepatobiliary disease and 0 controls. 16S sequencing did not identify existence of a core bile microbiome. Cats with positive culture results for E. coli (n=7) had a concurrent bloom of additional pathogenic bacteria identified by 16S sequencing. Cats that were culture‐positive for non‐E. coli species (n=3) had neither the cultured bacteria nor other bacteria identified by sequencing.

Conclusions and Clinical Importance: Cat bile does not harbor a microbiome but may contain bacterial pathogens not identified by culture. These pathogens may contribute to pathogenesis of hepatobiliary disease, particularly in cats with bile E. coli infection.

145.1.1.1. ¹Southern Counties Veterinary Specialists; ²RCVS and European Diplomate in Small Animal Medicine, Internal Medicine, Southern Counties Veterinary Specialists; ³The Rodhams

Background: Serum bile acid concentrations represent an early predictor of short‐term survival in critically ill people; however, their use as biomarkers in septic dogs has not been previously established.

Objective: Evaluate if fasting serum bile acid concentrations differ between septic and non‐septic dogs and identify any relationship with survival and hospitalization period in septic dogs.

Animals: Twenty‐six client‐owned dogs diagnosed with sepsis. Dogs presenting with a non‐hepatobiliary medical condition (non‐septic group; n=21) and dogs admitted for an elective orthopedic procedure, considered otherwise healthy (orthopedic group; n=29), were included as control groups.

Methods: Retrospective observational study of medical records of a single referral center from 2008 to 2020. Fasting serum bile acids, biochemical and hematological parameters were compared between groups, using the Kruskal‐Wallis with post hoc Dunn's test or one‐way ANOVA with post hoc Tukey's test. For septic dogs, the relationship between fasting serum bile acids and hospitalization period was assessed using Spearman rank correlation, and survival to discharge using the Mann‐Whitney U test.

Results: Fasting serum bile acids were significantly higher in the septic (median 17.5 μmol/L, interquartile range [IQR] 7.5–26.9 μmol/L) compared to the non‐septic group (median 6.9 μmol/L, IQR 4.7–9 μmol/L; P=0.013). However, no difference was identified between the septic and the orthopedic group (median 12 μmol/L, IQR 9–14 μmol/L; P=0.471). Fasting serum bile acid concentrations were not associated with hospitalization period (Spearman's rho =‐0.129; P=0.529) nor with survival (P=0.918).

Conclusions and Clinical Importance: Fasting serum bile acid concentrations were not considered prognostic indicators.

146.1.1.1. ¹Gastrointestinal Laboratory, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University; ²Director of Bioinformatics, MicroGen DX; ³Microbiology Technician I, Vet Med‐Teaching Hospital, Clinical Microbiology Laboratory, VMTH, Texas A&M University; ⁴Technical Laboratory Coordinator, Vet Med‐Teaching Hospital, Clinical Microbiology Laboratory, VMTH, Texas A&M University; ⁵Professor and Associate Director for Research/Head of Microbiome Science, Department of Small Animal Clinical Science, Gastrointestinal Laboratory, Texas A&M University; ⁶Professor in Veterinary Pathobiology/Director, Clinical Microbiology and Immunology, VTPB, Department of Veterinary Pathobiology, Texas A&M University; ⁷Associate Professor, Small Animal Internal Medicine Associate Director for Clinical Services, Department of Small Animal Clinical Science, Gastrointestinal Laboratory, Texas A&M University; ⁸University Regents Professor/University Distinguished Professor (with Tenure), Small Animal Internal Medicine/Dr. Mark Morris Chair in Small Animal Gastroenterology and Nutrition/Director, Gastrointestinal Laboratory (GI Lab), Department of Small Animal Clinical Science, Gastrointestinal Laboratory, Texas A&M University

Background: Culture has traditionally been used to detect and identify bacteria. Results can take several days and some species of bacteria are fastidious. Molecular methods could allow results to be obtained more quickly and might allow detection of fastidious species.

Objective: To compare bacteriological culture with 16S rRNA gene next‐generation sequencing (NGS) and qPCR for identifying bacteria in liver samples from dogs.

Animals: Liver biopsies obtained from 18 dogs presented for diagnostic workup of chronic liver disease.

Methods: Prospective cross‐sectional study. Liver biopsies were macerated and plated on standard media for aerobic and anaerobic bacteriological culture. The same biopsy sample was transported overnight in an anaerobic container on ice packs for NGS and qPCR. For 10 dogs, negative controls were obtained by swabbing the macerator and culture broth (together, or separately) prior to sample processing.

Results: Seventeen samples had a negative bacteriological culture. One culture positive sample (Corynebacterium tuberculostearicum) was negative by molecular methods. Four of 18 samples (22%) were positive by molecular methods only; reported bacteria varied widely. Concordant negative results between bacteriological culture and molecular methods were obtained in 13/18 (72%) samples. In 5/10 (50%) negative controls, NGS detected E. coli DNA (10⁵–10⁷ copies/gram).

Conclusions and Clinical Importance: Concordant negative results from bacteriological culture and molecular methods were obtained in a majority of liver samples from dogs. However, the clinical relevance of bacterial DNA in liver samples from dogs remains uncertain as half of the negative controls were reported to contain bacterial DNA.

147.1.1.1. ¹Cummings School of Veterinary Medicine, Tufts University; ²Veterinary Internist, MedVet Medical and Cancer for Pets; ³Associate Professor, Cummings School of Veterinary Medicine, Tufts University; ⁴Professor, Clinical Sciences, Cummings School of Veterinary Medicine, Tufts University

Background: Dogs with hepatocutaneous syndrome (HCS) have plasma hypoaminoacidemia. The specificity of this finding among dogs with other chronic hepatopathies (CH) is unknown.

Hypothesis/Objectives: To compare plasma amino acid panels (PAA) from a retrospective cohort of dogs with HCS with PAA in dogs with non‐HCS CH.

Animals: Forty client‐owned dogs with CH.

Methods: Dogs undergoing hepatic biopsy for chronic serum liver enzyme elevations were recruited. Twelve dogs each with vacuolar hepatopathy or chronic inflammatory disease and 8 dogs with congenital vascular/biliary disease were enrolled. Eight cases of HCS with PAA were retrospectively enrolled. PAA analysis was done at the UC Davis Amino Acid Laboratory. PAA concentrations in the 4 groups were compared using one‐way ANOVA and the Kruskal‐Wallis test, followed by the Tukey Kramer HSD and Steel‐Dwass test for post hoc analysis, respectively. To find a unique PAA pattern in dogs with HCS, Lasso regression analysis with AIC forward selection was performed.

Results: There was no difference in breed, age or sex among the 4 groups. Compared to non‐HCS liver diseases, dogs with HCS had significantly lower concentrations of aspartic acid, threonine, serine, asparagine, glutamine, glycine, alanine, citrulline, methionine, lysine, histidine, arginine, and proline (P<0.0019 after Bonferroni correction). The Lasso regression analysis revealed that low glutamine, glycine, citrulline, arginine and proline concentrations were characteristic of dogs with HCS.

Conclusions and Clinical Importance: Plasma hypoaminoacidemia does not occur in dogs with non‐HCS CH. Determination of PAA is warranted as a noninvasive biomarker for diagnosis of HCS in dogs.

148.1.1.1. ¹GI Lab, Texas A&M University; ²Director of Gastrointestinal Laboratory, Small Animal Internal Medicine, GI Lab, Texas A&M University; ³Associate Director for Research, Small Animal Internal Medicine, GI Lab, Texas A&M University; ⁴Associate Director for Clinical Services, Small Animal Internal Medicine, GI Lab, Texas A&M University

Background: Isoprostanes are stable end products of lipid peroxidation that have been used as markers of oxidative stress in humans and dogs. It has previously been reported that a cohort of dogs with liver disease (inflammatory, metabolic, vascular, or neoplastic) had increased urine isoprostane concentrations compared to healthy control (HC) dogs. However, urine isoprostane concentrations have not been individually reported for different types of canine liver disease.

Hypothesis/Objectives: To assess oxidative stress in dogs with different types of liver disease.

Animals: Urine was collected from 20 HC dogs and from dogs with liver disease, including 25 with chronic hepatitis (CH), 7 with steroid hepatopathy (SH), and 8 with congenital portosystemic shunt (CPSS).

Methods: Prospective, observational study. Urinary 15‐F_2t‐isoprostane concentrations were measured by gas chromatography/negative ion chemical ionization mass spectrometry and normalized to urine creatinine concentrations. Concentrations were compared between groups using a Kruskal‐Wallis test, followed by Dunn's multiple comparisons tests. Significance was set at P<0.05.

Results: The median [range] urine 15‐F_2t‐isoprostane to creatinine ratios (ng/mg UCr) were 3.6 [2.2–12.4], 5.7 [2.4–11.3], 4.8 [2.4–8.6], and 12.5 [2.9–22.9] for HC dogs and dogs with CH, SH, or CPSS, respectively. CPSS dogs had significantly higher urine isoprostane concentrations compared to HC dogs (P=0.004).

Conclusions and Clinical Importance: These findings suggest that dogs with CPSS experience increased oxidative stress compared to HC dogs. If confirmed in a larger group of dogs, this may have therapeutic implications.

149.1.1.1. Veterinary Gastrointestinal Laboratory, Texas A&M University

Background: MicroRNAs 1275, 222, and 21 are involved in copper homeostasis and their serum expression has been reported to be altered in human patients with Wilson's disease.

Objective: To quantify microRNAs 1275, 222, and 21 in serum from dogs with copper‐associated chronic hepatitis (CuCH), dogs with idiopathic chronic hepatitis (iCH), and healthy control dogs.

Animals: Stored serum samples from 8 dogs with histologically confirmed CuCH, 8 dogs with histologically confirmed iCH, and 8 healthy control dogs.

Methods: Retrospective cross‐sectional study. Target MicroRNAs were quantified by real‐time, reverse transcription polymerase chain reaction (RT‐qPCR). MicroRNA 26b was used as housekeeping gene. Relative quantification was calculated using the ΔΔCt method. Kruskal‐Wallis tests were used to compare differences in copy numbers between groups. Statistical significance was set at P<0.05.

Results: Dogs with CuCH showed a decreased expression of MicroRNA 1275 compared to healthy controls (CuCH [median; range]: 0.15; 0.0–0.40; healthy controls: 1.7; 0.10–10.3) but this difference did not reach statistical significance (P=0.06; Figure 1). Other post hoc comparisons between groups were not significant (P=0.13). No significant differences between groups were observed for MicroRNA 222 (P=0.10) or 21 (P=0.15), respectively.

Conclusions and Clinical Importance: These results, from a small number of dogs, do not support the diagnostic utility of microRNA 222 or 21 for the diagnosis of CuCH in dogs. However, larger cohort studies evaluating the utility of MicroRNA 1275 are warranted to determine the diagnostic utility of MicroRNA 1275.

150.1.1.1. ¹University of California, Davis; ²Colorado State University; ³Professor, Microbiology, Immunology, Pathology, Colorado State University; ⁴Professor, Colorado State University

Background: Copper associated hepatopathy (CAH) has become an important and prevalent malady since the 1990s, coincidental with changes in copper content in commercial dog foods. Understanding the demographic and histopathologic features related to hepatic copper (Cu) concentrations may aid in identifying and better understanding dogs with CAH.

Hypothesis/Objectives: The primary aim was to describe associations between hepatic Cu concentrations and demographic and histopathologic features.

Animals: Dogs that underwent liver histopathology and Cu quantification at a veterinary diagnostic laboratory between July 2010 and February 2020.

Methods: Data was retrospectively collected from medical records. A Gaussian multiple regression model on the log scale was used to evaluate associations between hepatic Cu and demographic features and key histologic features of necroinflammation.

Results: Of 4,559 cases meeting criteria, 49% had hepatic Cu >400 and 19% had Cu >1000 μg/g dw (dry weight) (reference range 120–400 μg/g dw). Mean hepatic Cu was 704 μg/g dw, range 4.5–31,500. Severity of inflammation (mild, moderate, and severe) and necrosis/apoptosis were associated with increased hepatic Cu (p<0.01). Age was negatively associated (p<0.02), but specific breeds (Doberman pinscher, Labrador, and West Highland white terrier) and the diagnosis of chronic hepatitis was associated with increased hepatic Cu (p<0.001).

Conclusion and Clinical Importance: Elevated hepatic Cu is prevalent amongst dogs. Severity of inflammation and necrosis/apoptosis are predictive of higher hepatic Cu. Dogs with necroinflammation of the liver have significantly higher hepatic Cu compared to dogs without.

151.1.1.1. ¹College of Veterinary Medicine, The Ohio State University;²Internal Medicine Resident, Veterinary Clinical Sciences, The Ohio State University; ³Assistant Professor, Veterinary Clinical Sciences, The Ohio State University

Background: Ursodiol is a naturally occurring bile acid prescribed to cats for a wide range of gallbladder and liver conditions. Ursodiol is known for its ability to expand the bile acid pool by increasing bile flow and displacing hepatotoxic bile acids. Additionally, ursodiol has anti‐inflammatory and anti‐fibrotic properties. Therefore, ursodiol is used in a variety of feline hepatic diseases, including acute/chronic hepatopathies, cholangiohepatitis, and for gallbladder abnormalities. Additionally, veterinarians prescribe a wide range of dosages and duration of ursodiol administration. Prescribing trends and clinical characteristics of feline patients receiving ursodiol is unknown.

Objective: This retrospective study aimed to determine the clinical use, dosage, and duration of therapy of ursodiol administration in feline patients at a teaching hospital over 13 years (2008–2021).

Animals: Since 2008, 88 feline patients were prescribed ursodiol at the Ohio State University Veterinary Medical Center and were included in this retrospective study.

Methods: Metadata from these patients, including signalment, history, diagnostics (including laboratory data and diagnostic imaging), ursodiol dosage/duration, and patient outcomes were recorded.

Results: Ursodiol was most prescribed as a hepatoprotectant (29/88, 33%), for gallbladder sludge (26/88; 30%), or for cholestasis (17/88; 19%) in feline patients. The most common prescribed dose was 10.89 mg/kg/day (range 5.2 to 33.9 mg/kg/day). An accurate duration of ursodiol administration was unable to be determined.

Conclusions and Clinical Importance: Overall, these findings document ursodiol prescribing trends and clinical characteristics of feline patients presenting to a teaching hospital. These results can help guide novel applications of ursodiol administration in feline patients.

152.1.1.1. ¹Tufts VETS; ²Staff Internist, Tufts VETS; ³Head of Pathology, Pathology, School of Veterinary Medicine, University of Surrey; ⁴Cummings School of Veterinary Medicine at Tufts University

Background: Granulomatous hepatitis (GH) is a form of chronic hepatitis in dogs on which limited information has been published.

Hypothesis/Objective: The objective was to describe the presentation, imaging findings, survival time, and response to immunosuppressants in dogs with GH.

Animals: Twenty‐nine client‐owned dogs with confirmation of GH on hepatic biopsy.

Methods: This was a retrospective observational study. Pathology records from an academic medical center were searched. Inclusion criteria included a histopathologic diagnosis of GH based on WSAVA guidelines, absence of an identified etiology and a medical record which included clinical presentation, clinical pathologic and ultrasound findings, treatment, and adequate follow‐up to determine survival times.

Results: The median age was 7 yr (range, 0.66 to 12 yrs). Nineteen breeds were represented. Decreased appetite (19/29), lethargy (16/29), and fever (13/29) were seen. All dogs had increases in serum transaminases, while 21/29 and 12/24 had hyperbilirubinemia and neutrophilia, respectively. Hepatomegaly (12/26) and hypoechoic nodules (7/26) were common ultrasound findings. Histopathologic necroinflammatory scores were moderate to severe in 90% of dogs, while fibrosis scores were commonly graded as mild (75%). Overall survival was 635 days (range, 1–6442 days). Kaplan Maier analysis showed that 19 dogs treated with immunosuppressants survived longer (1074 days, range 18 to 2482) than the 8 dogs not treated with them, (480 days, range 1 to 1526). (p=0.04). Effect size was 0.80 (95% confidence interval, 0.36–1.23)

Conclusion and Clinical Importance: GH in dogs is associated with high histopathologic grade, fever, neutrophilia, and an apparent improved survival with immunosuppressants.

154.1.1.1. ¹College of Veterinary Medicine, University of Leipzig; ²Veterinary Specialist in Porcine Health Management, VetaRegio; ³Vetsuisse Faculty, University of Zurich; ⁴Small Animal Clinic Aarau West; ⁵Gastrointestinal Laboratory, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University

Background: Serum C‐reactive protein (CRP) is a biomarker of systemic inflammation in dogs and is used in clinical practice to aid in diagnosing and monitoring primary inflammatory conditions (e.g., immune‐mediated polyarthritis). Treatment of these diseases often includes anti‐inflammatory or immunosuppressive doses of corticosteroids. However, potential effects of corticosteroid administration on serum CRP concentrations in dogs have not been studied.

Objective: To evaluate serial serum CRP concentrations in healthy dogs receiving high‐dose corticosteroid treatment of >4 weeks.

Animals: Samples from a prospective cohort study including 12 healthy Beagle dogs were used.

Methods: Serum samples obtained before initiation (baseline) and on days 1, 5, and 28 of receiving either 8 mg/kg hydrocortisone PO q12h (6 dogs) or placebo (6 control dogs) were used. Serum CRP concentrations were measured in all samples (stored at ‐80°C) using the TriDelta canine CRP‐ELISA (total of 2 assay runs using the same assay reagents). A MANOVA model with Box‐Cox transformed (l=‐0.93) data followed by individual t‐tests served for statistical analysis, with P<0.05 indicating significance.

Results: Serum CRP concentrations slightly decreased over the 4 weeks in both groups of dogs, without a significant effect of hydrocortisone administration (P=0.7605; MANOVA). Two baseline samples and one sample obtained on day 1 of hydrocortisone treatment yielded serum CRP concentrations above the upper reference limit (>7.6 mg/L); all other measurements were within the reference interval.

Conclusions and Clinical Importance: These findings suggest a lack of a confounding effect of hydrocortisone administration on serum CRP concentrations in dogs.

155.1.1.1. College of Veterinary Medicine and Biomedical Sciences, Colorado State University

Background: Lokivetmab (Cytopoint®) is an anti‐IL‐31 monoclonal antibody used for management of canine allergic pruritis. IL‐31 is a cytokine shown to activate the neuronal signaling pathway associated with itch in dogs. Multiple publications confer the efficacy and safety of lokivetmab for allergic disease when given at recommended dosages. However, there have been anecdotal concerns that lokivetmab could trigger an immune‐mediated response leading to immune‐mediated hemolytic anemia (IMHA) or immune thrombocytopenia (ITP). No study to date has evaluated for the development of antibodies against erythrocytes or platelets when using this drug.

Hypothesis/Objectives: Determine if anti‐erythrocyte antibodies (AEA) or anti‐platelet antibodies (APA) develop following a standard dose therapy of lokivetmab in healthy dogs. We hypothesized that the development of AEA or APA following a single‐dose of lokivetmab would be rare and unlikely of clinical significance.

Animals: 8 healthy adult purpose‐bred research beagles.

Methods: Prospective descriptive study. Whole blood samples were taken at baseline (day 0) prior to lokivetmab administration and on days 14 and 28 following lokivetmab administration. A CBC and AEA and APA direct flow cytometry assays were performed at all time points.

Results: No dogs exhibited anemia or thrombocytopenia and all were negative for AEA and APA at all measured time points.

Conclusions and Clinical Importance: There was no evidence of APA or AEA formation in healthy dogs administered lokivetmab at standard dosages. Future work includes a clinical trial for the evaluation of anemia, thrombocytopenia, AEA, and APA in client‐owned allergic dogs receiving lokivetmab.

156.1.1.1. ¹Clinic of Small Animal Medicine, LMU Munich; ²Viro Vet Diagnostic, Biomedical Research Centre, Giessen

Background: To minimize the risk of feline injection site‐associated sarcomas, cats should be vaccinated with non‐adjuvanted feline leukemia virus (FeLV) and rabies vector vaccines, which are currently not licensed for simultaneous use and have to be administered on separate visits.

Hypothesis: The aim of the study was to investigate the immune response of cats after vaccination against rabies and FeLV at simultaneous versus separate visits using canarypox‐vectored vaccines.

Animals: 106 healthy, FeLV antigen‐negative, client‐owned kittens (8–12 weeks) were included.

Methods: In this prospective randomized study, all cats received primary vaccinations against rabies (day 0) and FeLV (day 28, 56). After one year, the study group (n=52) received booster vaccinations against rabies and FeLV simultaneously (day 364). The control group (n=54) received booster vaccinations against rabies (day 364) and FeLV (day 392) separately (4 weeks apart). Anti‐rabies antibodies were determined by fluorescent neutralization assay (days 28, 364, 392) and significant differences in anti‐rabies antibodies between both groups were determined by Mann‐Whitney U test.

Results: Four weeks after the one‐year booster vaccination against rabies, all cats had anti‐rabies antibodies (≥0.5 IU/ml), and there was no significant difference (p=0.141) in the median titer between the study group (14.30 IU/mL) and the control group (21.39 IU/mL).

Conclusions and Clinical Importance: According to the World Health Organization and the World Organization for Animal Health, anti‐rabies antibodies ≥0.5 IU/ml indicate protection against rabies. Simultaneous administration of FeLV and rabies vector vaccines does not interfere with the development of anti‐rabies antibodies.

157.1.1.1. ¹BluePearl Maitland; ²Kaleidoscope Statistics, LLC; ³Internal Medicine, BluePearl Maitland

Background: IMHA (immune mediated hemolytic anemia) is defined as an immune mediated destruction of red blood cells. Relapses are known to occur but risk factors are poorly defined.

Hypothesis: We hypothesize that patients with more severe IMHA on presentation (lower PCV at diagnosis, more transfusions given) would have higher relapse rates.

Animals: IMHA was defined as a PCV less than 30% at diagnosis and if one of the following were identified: spherocytes, positive Coombs’ test, positive slide agglutination. Records between January 2005 to December 2019 from one specialty hospital found 223 dogs who met these criteria.

Methods: This was a retrospective case‐control study with a significance threshold of 0.05.

Results: There were 21 relapses. Overall relapse probabilities (95% CI) at 3 and 12 months were 5% (2–11) and 12% (7–20). Dogs who received two or more transfusions had a higher probability of relapse than dogs who received one transfusion or none (p=0.044). Dogs who had PCV <10% had a 27% (95% CI 10–63) probability of relapsing within 1 year compared to dogs who had PCV >10% who had a 9% (95% CI 5–18) probability of relapsing within 1 year (p=0.010 log‐rank).

Conclusions: Patients with extremely low PCVs at diagnosis or need multiple transfusions are significantly more likely to relapse. These patients would likely have a poorer prognosis.

158.1.1.1. ¹Colorado State University; ²Center for Companion Animal Studies, Colorado State University; ³Department of Clinical Sciences, College of Veterinary Medicine and Biomedical Sciences, Colorado State University; ⁴Department of Microbiology, Immunology, and Pathology, College of Veterinary Medicine and Biomedical Sciences, Colorado State University

Background: Anti‐erythrocyte antibodies play a major role in primary and secondary immune‐mediated hemolytic anemia (IMHA). A direct flow cytometry assay to detect antibodies on the RBC is quantitative (% IgG binding) and an alternative to direct Coombs testing.

Hypothesis/Objectives: To report the % IgG binding among groups of anemic dogs and to compare to direct Coombs test results when available.

Animals: Final diagnoses for 116 client‐owned, anemic dogs were classified as idiopathic IMHA, neoplasia, blood‐borne pathogen infection, and other etiologies.

Methods: Whole blood in EDTA was stored at 4°C and analyzed for % IgG binding and direct Coombs test (positive >1:4) within 24 hours of collection. The % IgG binding among disease categories was not normally distributed and so medians were compared by Mann‐Whitney U test.

Results: % IgG binding was detected in some dogs with IMHA (30/46 [65.2%]; median 10.63%, range 0.24–99.8), neoplasia (3/34 [8.8%]; 3.16% 0.14–16.85), blood‐borne pathogens (1/1 [100%]; 5.18%), and with other etiologies (5/35 [14.3%]; 0.83%, 0.16–95.1). Results from flow and Coombs test were discordant for 13 of 42 dogs (30.9%) and glucocorticoids had been administered to 11/13 (84.6%) dogs with discordant results before testing.

Conclusions and Clinical Importance: Dogs with idiopathic IMHA have significantly higher % IgG binding than dogs with other causes of anemia. Explanations for the discordant results for the 2 assays are currently being explored.

159.1.1.1. ¹Royal Veterinary College; ²University of Pennsylvania; ³School of Veterinary Medicine, University of Pennsylvania; ⁴Auburn University; ⁵Department of Clinical Sciences, Auburn University; ⁶Cummings School of Veterinary Medicine, Tufts University; ⁷Cornell University; ⁸School of Veterinary Medicine, Louisiana State University

Background: Little is known about the impact of the intestinal microbiome on immune‐mediated hemolytic anemia (IMHA) and immune thrombocytopenia (ITP) in dogs.

Hypothesis/Objectives: We hypothesize that IMHA and ITP in dogs are associated with intestinal dysbiosis. We evaluated the fecal microbiome of diseased dogs at presentation and after initiation of treatment.

Animals: 25 affected and 13 healthy control dogs were included in the study. Samples were collected from affected dogs at presentation, and 2 and 8 weeks after initiation of treatment.

Methods: 16S rRNA gene sequencing was used to profile the fecal microbiome of affected dogs and healthy controls. Microbiome diversity and composition were analyzed in the context of disease status.

Results: The fecal microbiota of diseased dogs were significantly more diverse than those of healthy dogs at presentation (Shannon and Simpson indices, p≤0.03). The genus Mitsuokella was enriched in diseased dogs (q<0.001), whereas the genus Treponema was enriched in healthy dogs. The genus Treponema was also an enterotype of all study dogs (odds ratio [OR]=0.19, 95% confidence interval [CI]=0.04–0.68). The Pirellulaceae p−1088−a5 (OR=2.59, 95% CI=1.14−7.27) and Lachnospiraceae NK4A136 (OR=3.13, 95% CI=1.02−11.77) groups were associated with disease. Beta diversity analysis demonstrated a shift in microbiome composition toward that of healthy dogs with treatment.

Conclusions and Clinical Importance: IMHA and ITP in dogs may be associated with intestinal dysbiosis. Specific bacterial genera may serve as disease biomarkers. Additional studies are warranted to evaluate whether microbiome composition or diversity play a causal role in disease initiation or progression.

160.1.1.1. ¹Seoul National University; ²Director, Internal Medicine, Haemaru Referral Animal Hospital; ³Internal Medicine, Seoul National University; ⁴Associated Professor, Internal Medicine, Seoul National University; ⁵Professor, Internal Medicine, Seoul

Background: Romiplostim is a thrombopoietin receptor agonist and is used for the treatment of immune‐mediated thrombocytopenia (IMT) in humans and there is limited literature in dogs.

Hypothesis/Objectives: The aim of this study is to investigate the effects and adverse events of romiplostim in dogs and cats with primary thrombocytopenia or secondary thrombocytopenia.

Animals: Twenty‐eight dogs and 2 cats were included.

Methods: A retrospective study was conducted by extracting data from electrical medical records from 2020 to 2021. We evaluated the effects of treatment and adverse events of romiplostim for primary IMT and secondary thrombocytopenia caused by several underlying conditions including lymphoma, babesiosis, pancytopenia, feline leukemia virus infection, post‐vaccination, liver failure, and systemic lupus erythematous.

Results: The platelet count increased to the reference range in 15 dogs, and 10 of them survived. Response rate was primary IMT 58.3% (7/12), pancytopenia of unknown cause 100% (3/3), chemotherapy‐induced thrombocytopenia 66.6% (2/3), hepatic failure 100% (1/1), post‐vaccination 100% (1/1), babesiosis 100% (1/1), primary pancytopenia 0% (0/4), paraneoplastic thrombocytopenia 0% (0/3). The recovery period of platelet count after administration of romiplostim was 7 (range, 2–10) days. Romiplostim was well‐tolerated.

Conclusions and Clinical Importance: Romiplostim was well tolerated and effective in approximately one‐half of primary IMT and secondary thrombocytopenia in dogs.

161.1.1.1. College of Veterinary Medicine and Biomedical Sciences, Colorado State University;

Background: Telmisartan is licensed in the USA to control hypertension in cats. In other species, telmisartan also has immunomodulatory properties and is an anti‐oxidant.

Objective/Hypothesis: To determine whether telmisartan administered to cats has antioxidant or immune modulation effects.

Animals: Eight healthy, young adult, purpose‐bred, mixed sex cats.

Methods: Cats (n=8) were administered telmisartan at 1.5 mg/kg PO twice daily for 14 days followed by 1.5 mg/kg PO once daily for 14 days. Superoxide dismutase (SOD) and plasma glutathione peroxidase (GPx) activities were determined in plasma using commercial kits and compared between day 0 and 14. The mRNA of select cytokine genes detected in blood by RT‐PCR from the telmisartan cats on day 14 and 28 and compared to 6 control cats. Data were assessed for normalcy and then compared by Student's T test or Mann Whitney U test.

Results: Plasma SOD concentrations between day 0 and day 14 were similar. Plasma GPx activity was increased on day 14 (mean=27, std=21.9) compared to day 0 (mean=23.5, std=26.3) but not significant (P=0.78). While no differences in IL1‐beta, IL6, IFN‐alpha, and IFN‐beta were detected, TNF‐alpha levels were greater in the day 14 (P=0.04) and 28 (P=0.01) samples and IFN‐gamma levels were significantly greater in the day 28 samples (P=0.02).

Conclusions and Clinical Importance: Telmisartan has immunomodulatory effects in healthy cats, which suggests potential alternate uses of this drug.

163.1.1.1. ¹University of California‐Davis, Davis, CA, USA; ²Assistant Professor, Medicine and Epidemiology, University of California‐Davis, Davis, CA, USA

Background: Amphotericin‐B (AmB) is an essential medication for the treatment of systemic mycoses. Nephrotoxicity is a known complication of AmB treatment but the incidence and risk factors for acute kidney injury (AKI) following its administration are not known in dogs.

Objective: Determine the incidence of and risk factors for AKI in dogs receiving AmB.

Animals: 51‐client owned dogs receiving AmB for the treatment of systemic infections.

Methods: Retrospective study. Signalment, potential risk factors (dehydration, renal disease, recent non‐steroidal anti‐inflammatory drug administration, recent anesthesia, ICU hospitalization, neurologic deficits, inpatient status), AKI development (creatinine >0.3 mg/dL from baseline), drug formulation (deoxycholate (AmB‐D) or lipid complex (ABLC)), dose, and treatment duration were recorded. Incidence of AKI was determined, and odds ratios were calculated for potential risk factors.

Results: Incidence of AKI was 5/12 (41%) for dogs receiving AmB‐D and 14/39 (36%) for dogs receiving ABLC. Of the 19 dogs that developed AKI, 16 (84%) continued treatment after a break in the planned dosing protocol. Median cumulative dose administered prior to first AKI was 6.9 mg/kg for AmB‐D and 22.5 mg/kg for ABLC. ICU hospitalization (odds ratio (OR) 0.21, 95% confidence interval (CI) 0.58–0.87) and inpatient status (OR 0.25, and 95% CI 0.07–0.86) were associated with a decreased odds of AKI.

Conclusions and Clinical Importance: Incidence of AKI with AmB is common, but does not always preclude continued treatment. The incidence of AKI is similar between AmB‐D and ABLC, but dogs receiving ABLC tolerated a higher total dose before AKI developed.

164.1.1.1. ¹Cummings School of Veterinary Medicine; ²Massachusetts Veterinary Referral Hospital, Ethos Veterinary Health;³Emerson Hospital; ⁴School of Medicine, Tufts Medical Center, Tufts University

Background: Antimicrobial stewardship programs (ASPs) are increasingly prevalent in companion animal hospitals, but despite the critical role of veterinary technicians in patient care, there is often a lack of ASP focus on support staff.

Hypothesis/Objectives: To establish baseline attitudes and practices of veterinary technicians in a tertiary care center around antimicrobial use, prior to the institution of a hospital ASP.

Animals: None.

Methods: This was a prospective, survey‐based study. Data were collected through means of an online questionnaire, consisting of a combination of closed and open‐ended questions, sent via a targeted email list. Descriptive statistics were used for data analysis.

Results: A total of 39 completed responses were received. Seventy‐nine percent of participants felt that antimicrobial stewardship was important, and 54% wanted to be more involved in the ASP. Most participants were comfortable suggesting non‐antimicrobial treatment alternatives with doctors or students (92%), but fewer were comfortable discussing the need for antimicrobial time‐outs (54%). Technicians listed a lack of familiarity with appropriate treatment choices, and a perception that some doctors may not be open to input from support staff among their barriers to conversations with doctors. Technician suggestions on how to include support staff in the ASP included education on antimicrobial use, and a focus on non‐prescribing practices such as infection control.

Conclusions and Clinical Importance: Our research demonstrates an interest in and a desire to learn more about antimicrobial stewardship in this technician population. Targeted engagement of veterinary support staff may increase success within hospital ASPs.

165.1.1.1. ¹College of Veterinary Medicine, North Carolina State University; ²Associate Professor and Director of Diagnostic Laboratories, Population Health and Pathobiology, College of Veterinary Medicine, North Carolina State University

Background: Conflicting data exist regarding the effect of urine concentration on the growth of canine uropathogens in clinical or subclinical urinary tract infection (UTI). Assessing bacterial growth of uropathogens in varying urine concentrations may improve our understanding of factors influencing pathogenesis or treatment of UTI.

Hypothesis: Different uropathogens will exhibit unique growth patterns that are dependent on urine concentration and specific in vitro urine diluent.

Animals: Stored, frozen, culture negative canine urine was used, which was previously collected from healthy patients.

Methods: The urine was diluted from a starting specific gravity (USG) of 1.035 with phosphate buffered saline or sterile water to final USGs of 1.005, 1.015, or 1.025. Each was inoculated with 1:100 0.5 McFarland concentration of 1 of 4 clinical uropathogens (Escherichia coli, Enterococcus faecalis, Staphylococcus pseudintermedius, and Klebsiella pneumoniae) and incubated. The bacterial concentration (CFU/mL) was determined at hour 0 and 8. All counts were log transformed and the differences from pre‐ to post‐incubation were represented graphically for each organism.

Results: Some organisms exhibit stunted growth in very dilute (E. coli, S. pseudintermedius) and well concentrated urine (E. coli, K. pneumoniae), while others exhibit modest growth independent of urine concentration (E. faecalis) and diluent. There was not a consistent diluent effect on supporting uropathogen growth in vitro.

Conclusions: Different bacterial genera previously associated with UTI exhibit unique growth patterns that are dependent on urine concentration and diluent, which may suggest that uropathogens have individually specific environmental preferences for optimal growth.

166.1.1.1. ¹LexaGene; ²Director, Applications and Assay Development, LexaGene; ³Senior Staff Scientist, LexaGene; ⁴Vice President of Applications, Bioinformatics, LexaGene; ⁵CEO and Founder, LexaGene

Bacterial culture and antimicrobial susceptibility testing is considered the gold standard to confirm the presence of an infection and antimicrobial resistance, but typically requires 3–5 days to obtain results. Genotypic detection of specific pathogens and their associated antimicrobial resistance genes can be used to predict antimicrobial resistance in bacterial pathogens and can deliver accurate results in hours rather than days.

This study was carried out as a proof‐of‐principle evaluation of MiQLab System to identify bacterial pathogens and their resistance genes. The MiQLab System equipped with MiQLab Bacterial and AMR test V2 can detect 10 common bacterial pathogens infecting cats and dogs and resistance genes for common first‐line veterinary antimicrobials such as beta‐lactams, sulphonamide‐trimethoprim combinations, lincosamides and tetracyclines.

The study evaluated the ability of the MiQLab to detect bacterial pathogens and resistance genes in highly characterised bacterial isolates with available genetic information. The overall percent agreement for bacterial identification between the MiQLab and bacterial culture was 100%. Similarly, the overall percent agreements between antimicrobial resistance gene detection on the MiQLab and genetic sequence data was 96.0%.

These results indicate that MiQLab's qPCR‐based genotypic detection technology can be used to identify bacterial pathogens and their associated antimicrobial resistance genes accurately with a rapid turn‐around time.

167.1.1.1. ¹MiraVista Diagnostics; ²Oklahoma State University; ³Senior Scientist, MiraVista Diagnostics; ⁴Assistant Teaching Professor, Veterinary Medicine and Surgery, University of Missouri; ⁵Veterinary Specialists of North Texas; ⁶Medical Director, MiraVista Diagnostics

Background: Anti‐Histoplasma antibody detection by enzyme immunoassay (EIA) or immunodiffusion (ID) is used to support the diagnosis of histoplasmosis in cats and dogs. There is a paucity of published diagnostic performance data.

Objectives: The primary objective was to describe the performance of IgG EIA and ID for anti‐Histoplasma antibody detection in cats and dogs. A secondary aim was to describe the clinical utility of antibody detection when combined with antigen detection.

Animals: 51 animals with pathology proven histoplasmosis and 160 animals without histoplasmosis.

Methods: This is a retrospective cohort study. Results of Histoplasma antigen EIA at the time of diagnosis was included in extracted data. Stored serum samples were tested for anti‐Histoplasma antibodies by EIA and ID. Diagnostic sensitivity was calculated.

Results: Diagnostic sensitivity of IgG EIA was 24/30 (80.0%) in cats and 17/21 (81.0%) in dogs. Diagnostic sensitivity of ID was 0/30 (0%) in cats and 3/21 (14.3%) in dogs. Diagnostic sensitivity of Histoplasma antigen in urine was 28/30 (93.3%) in cats and 19/21 (90.5%) in dogs. When interpreted in parallel, combined diagnostic sensitivity of antigen EIA and IgG EIA was 51/51 (100%) in both species. In animals not diagnosed with histoplasmosis, anti‐Histoplasma IgG antibodies were detected in 1/21 (4.8%) of cats and 10/138 (7.2%) of dogs.

Conclusions and Clinical Importance: Immunodiffusion is not recommended for anti‐Histoplasma antibody detection in cats and dogs. Anti‐Histoplasma antibody detection by EIA should be considered with suspected histoplasmosis but no detectable Histoplasma antigen in urine.

168.1.1.1. ¹Antech Diagnostics; ²Manager Molecular Diagnostics, Antech Diagnostics; ³Molecular Diagnostics Technician, Antech Diagnostics; ⁴VP Medical Science and Innovation, Antech Diagnostics

Background: Three single nucleotide polymorphisms (SNP) in the highly conserved beta‐tubulin gene confer Benzimidazole resistance in many nematode parasites of veterinary and human importance. The F167Y SNP is associated with Ancylostoma caninum benzimidazole resistance in Greyhounds and other dog breeds.

Objective: A F167Y allele‐specific real‐time PCR (qPCR) was adopted and validated to screen hookworm positive samples.

Animals: Ninety‐two fecal samples that tested positive for Ancylostoma by routine ova and parasite (O&P) and confirmed by Ancylostoma qPCR were included and screened for the presence of the 167Y resistance marker.

Methods: The Ancylostoma O&P positive fecal samples were verified by Ancylostoma specific qPCR. The 167Y resistance allele‐specific qPCR was used to determine the frequency of Benzimidazole resistance genetic marker.

Results: The 167Y allele resistance marker was detected in 25 of the 87 (28.7%) Ancylostoma O&P and qPCR positive samples. Resistant allele positive samples were confirmed by beta‐tubulin gene sequencing using flanking primers. No breed information was available from 10 samples; two samples were from a mixed breed and German Shepherd Dog, respectively. One sample was represented by each of the following breeds: Boxer, Golden doodle, Yorkshire terrier, Labrador retriever, Basenji, Siberian husky, American Staffordshire terrier, Beagle, Australia cattle dog, poodle and dachshund, respectively.

Conclusions and Clinical Importance: Anthelmintic resistance in hookworms is an emerging concern in veterinary medicine. Routine identification of the benzimidazole resistance genetic marker for Ancylostoma positive samples will aid in monitoring the prevalence of hookworm resistance and provide veterinarians with information that guides appropriate treatment decisions.

169.1.1.1. ¹Antech Diagnostics; ²Research Associate, Translational Medicine Institute, Colorado State University; ³Research Coordinator, Center for Companion Animal Studies, Colorado State University; ⁴Molecular Diagnostics Technician, Innovation, Antech Diagnostics; ⁵Assistant Professor, Pathobiology, Auburn University; ⁶Pathobiology, Auburn University; ⁷VP Medical Science and Innovation, Innovation, Antech Diagnostics; ⁸University Distinguished Professor, Pathobiology, Auburn University; ⁹Kenneth W. Smith Professor in Small Animal Clinical Veterinary Medicine, Small Animal Internal Medicine, Colorado State University

Background: Ova and parasite (O&P) by centrifugal flotation is a common method for the detection of hookworm eggs in fecal samples. Molecular diagnostics using real‐time quantitative PCR (qPCR) is a highly sensitive method to detect nucleic acids of infectious agents, including parasites.

Objective: This study compared O&P with qPCR for the detection of Ancylostoma caninum in experimentally and Ancylostoma spp. in naturally infected dogs and cats.

Animals: Five beagles treated with an anthelmintic drug at day minus 10 were experimentally infected on day 0 with 300 Ancylostoma caninum infective larvae. Fecal samples were collected at regular intervals.

Field Samples: A total of 790 fecal samples were included in the comparison of O&P with qPCR to detect Ancylostoma spp.

Methods: Fecal samples from experimentally infected beagles were analyzed within 48 hours using O&P zinc sulfate centrifugation and flotation and qPCR for Ancylostoma. The O&P Ancylostoma positive field samples were tested for Ancylostoma spp. by qPCR.

Results: In the experimentally infected beagles, all five dogs tested qPCR positive by day 17 whereas all five dogs tested positive by O&P by day 28. In the field samples, overall agreement was 97.5% between O&P and qPCR.

Conclusions and Clinical Importance: The high correlation of O&P Ancylostoma positive samples with qPCR confirms the high sensitivity of this molecular diagnostic assay. In experimentally infected dogs, qPCR indicates earlier and more consistent detection of Ancylostoma caninum eggs in feces than O&P.

170.1.1.1. ¹Purdue University; ²Director, Veterinary Science, Banfield Pet Hospital; ³Banfield Pet Hospital

Background: Immediate‐type hypersensitivity reactions and other acute adverse events have been documented in dogs following vaccination. Risk factors have been investigated >15 years ago, but changes in vaccines or shifting genetic profiles may have altered risk factors.

Objectives: Determine incidence rates and potential risk factors for adverse events (AEs) recorded within 3 days of vaccine administration in dogs.

Animals: 4,954,565 dogs vaccinated at 21,103,149 office visits at Banfield Pet Hospitals.

Methods: Electronic medical records from 2016–2020 were used in a retrospective cohort study identifying all dogs receiving vaccinations and documenting potential vaccine‐associated AEs within 3 days following vaccination. Incidence rates for AEs with 95% CIs were calculated, and associations for potential risk factors were assessed by logistic regression.

Results: The overall documented AE rate was 0.001845 (0.1845%, or 18.45 AE/10,000 dogs), or <1/500. The 5 breeds (n>10,000) with highest AE rates were Dachshund, Boston terrier, miniature Pinscher, French Bulldog, and Havanese (68.62, 63.28, 53.73, 49.28, and 35.82 AE per 10,000, respectively). Increasing the number of vaccines administered increased AE risk in dogs ≤20 kg (OR 1.23; 95%CI: 1.22–1.24) but not in dogs >20 kg (OR 1.01; 95% CI: 0.99–1.02). AE rates for DAPP, leptospirosis, rabies, or Borrelia vaccines administered alone were similar (19.22–21.31 AE/10,000).

Conclusions and Clinical Importance: Small dogs receiving multiple vaccines per office visit remain at increased risk of AEs following vaccination, but AE rate by breed suggests genetics may be the greater risk factor for an acute AE.

171.1.1.1. ¹Universidade Federal do Acre; ²Anesthesiology Resident, Department of Veterinary Medicine, UFRGS; ³Veterinary; ⁴Center for Biological and Natural Sciences, UFAC/Rio Branco; ⁵Adjunct Professor, Department of Veterinary Medicine, UFRGS; ⁶Adjunct Professor, Department of Veterinary Medicine, Federal University of Paraná; ⁷Adjunct Professor, Center for Biological and Natural Sciences, UFAC/Rio Branco

Background: Canine hemoparasitosis is a set of diseases caused by intracellular microorganisms, such as bacteria and protozoa, transmitted mainly by ticks. Its diagnosis is usually made based on the observation of suggestive hematological changes such as anemia, thrombocytopenia, and leucopenia, however, tests with greater sensitivity are recommended, such as serology and PCR. Despite the wide distribution, there are few reports on the prevalence in the North, more specifically in the Western Amazon.

Hypothesis/Objectives: The present study aimed to assess the prevalence of Ehrlichia spp., Babesia spp., and Anaplasma spp. in thrombocytopenic dogs or with inclusions suggestive of hemoparasites in the municipality of Rio Branco, Acre.