Abstract OBJECTIVE The objective of this study was to develop a robust machine-learning approach for efficient detection and grading of sesamoiditis in horses using radiographs, specifically in data-limited conditions. SAMPLE A dataset of 255 dorsolateral-palmaromedial oblique (DLPMO) and dorsomedial-palmarolateral oblique (DMPLO) equine radiographs were retrospectively acquired from Hagyard Equine Medical Institute. These images were anonymized and classified into 3 categories of sesamoiditis severity (normal, mild, and moderate). METHODS This study was conducted from February 1, 2023, to August 31, 2023. Two RetinaNet models were used in a cascaded manner, with a self-attention module incorporated into the second RetinaNet's classification subnetwork. The first RetinaNet localized the sesamoid bone in the radiographs, while the second RetinaNet graded the severity of sesamoiditis based on the localized region. Model performance was evaluated using the confusion matrix and average precision (AP). RESULTS The proposed model demonstrated a promising classification performance with 92.7% accuracy, surpassing the base RetinaNet model. It achieved a mean average precision (mAP) of 81.8%, indicating superior object detection ability. Notably, performance metrics for each severity category showed significant improvement. CLINICAL RELEVANCE The proposed deep learning-based method can accurately localize the position of sesamoid bones and grade the severity of sesamoiditis on equine radiographs, providing corresponding confidence scores. This approach has the potential to be deployed in a clinical environment, improving the diagnostic interpretation of metacarpophalangeal (fetlock) joint radiographs in horses. Furthermore, by expanding the training dataset, the model may learn to assist in the diagnosis of pathologies in other skeletal regions of the horse.

Abstract OBJECTIVE To determine the dose effect of peri-neural liposomal bupivacaine (LB) in an induced forelimb lameness model. ANIMALS 12 clinically normal adult horses. METHODS A randomized cross-over design was performed with 1 limb receiving saline and the other LB: low dose (6), high dose (6). Lameness was induced in 1 forelimb using a frog-pressure model. In the lame limb, peri-neural injection of the palmar nerves at the proximal sesamoid bones was performed using saline, low dose LB (0.25 mg/kg) (LDLB), or high dose LB (0.5mg/kg) (HDLB) in random order with a 1-week washout period between treatments. Distal limb swelling, mechanical nociceptive thresholds (MNT), and objective lameness data were collected before and up to 72 hours after peri-neural anesthesia. Data analysis was performed with mixed model ANOVA, equality of medians test, and Kaplan Meier survival analysis. RESULTS Compared with baseline, horses treated with LDLB and HDLB had improvements in MNT and lameness (P < .001). In the LDLB group, the median duration of analgesia was 4.5 hours (range = 3–6 hours) and the median return to lameness was 7 hours (range = 4–24 hours). In the HDLB group, the median duration of analgesia was 12 hours (range = 4–48 hours) and the median return to lameness was 9 hours (range = 3–48 hours). Mild to moderate swelling was identified in 11/12 (92%) LB limbs. CLINICAL RELEVANCE Both LDLB and HDLB resulted in loss of skin sensation and improvement of lameness. There was high variability among horses in duration of action for both doses.

Abstract OBJECTIVE To test the influence of increasing injectate volumes on the regional effects of xylazine and morphine epidural analgesia, with the hypothesis that increasing volume produces more cranial spread of analgesia as determined by thermal threshold (TT) testing. ANIMALS 6 university-owned research/teaching horses (2 mares, 4 geldings) deemed healthy on physical examination and basic lameness evaluation, aged 6–19 years and weighing 420–560 kg, were used in this prospective, randomized, blinded, cross-over experimental study. METHODS After routine placement of a caudal epidural catheter, all animals were subsequently instrumented with a TT testing system at the withers (Location A), the cranial (Location B), and caudal (Location C) abdominal area, over the tuber coxae (Location D), and the hind limb dorsal pasterns (Location E). All horses underwent five testing cycles with 0.2 mg/kg morphine and 0.2 mg/kg xylazine diluted to 20, 35, 50, 75, and 100 mL. TT testing was performed at 2, 4, 6, 8, and 10 hours by blinded investigators. RESULTS With increased epidural volume, significantly greater cranial spread of analgesic effect was noted. All epidural volumes caused significant changes in TT testing at location E but only the largest volume resulted in a significant TT testing change at location A. CLINICAL RELEVANCE Volume influences the regional effects of caudal epidural analgesia in horses but might affect analgesic reliability.

Abstract OBJECTIVE To determine the effects of using either a linear or curvilinear oscillating, battery-powered saw blade on the extent of bone resection, bone fissure or fragmentation, soft tissue trauma, and surgical time for femoral head and neck excision (FHNE) in feline cadavers. ANIMALS 18 feline cadavers. METHODS Paired feline cadaveric femora were randomly assigned to either a 10 mm linear or 12 mm curvilinear blade for FHNE by 2 surgical residents. CT of each femur pre- and postoperatively were used to create 3D reconstructions of each femur. The residual remaining or excessively resected bone volume at the ostectomy site was compared to an “ideal” ostectomy line made by a board-certified surgeon on preoperative CTs. RESULTS There were no significant differences in residual or excessive bone resection by a saw blade (P = .84), between surgeons (P = .65), or in surgery time (P = .39). When compared to the “ideal” ostectomy, the linear saw blade removed slightly less bone compared to the curvilinear blade, but was not statistically significant (P = .82). No fissures or fractures were noted; however, the curvilinear blade removed the entire lesser trochanter in 1 cadaver and the linear blade partially removed the greater trochanter in 1 femur and 2 lesser trochanters in 2 femora. CLINICAL RELEVANCE The use of a curvilinear blade may be a viable option for performing FHNE in cats. In vivo studies are warranted to determine its efficacy in clinical cases where FHNE is performed and the incidence of complications postoperatively.

Abstract OBJECTIVE To evaluate skin perfusion in cats receiving dexmedetomidine compared to a placebo. ANIMALS 9 healthy adult research cats. METHODS A randomized, blinded, placebo-controlled study design was used. Two sites, the dorsal metatarsus (site: limb) and lateral flank (site: flank), were evaluated with laser speckle contrast imaging (LSCI) at baseline and following administration of dexmedetomidine (1, 3, or 5 mcg/kg, IV) or a placebo (0.9% saline, IV). Mean speckle contrast (MSC), a surrogate for perfusion, was obtained from LSCI and compared between treatments. Heart rate, sedation score, and body temperature were recorded. Skin perfusion to the flank and limb, reported as MSC, was assessed via LSCI at baseline and at 5, 10, and 15 minutes posttreatment. RESULTS There was a significant decrease in heart rate (P < .001) in cats receiving 1, 3, and 5 mcg/kg dexmedetomidine compared to placebo. There was a significant increase in median sedation score at all time points postsedation compared to baseline (P < .018). Changes in MSC for the metatarsus were not significantly different between treatments at any time point (P = .12). For the flank, MSC was significantly higher for cats treated with dexmedetomidine compared to baseline (P ≤ .01). Skin perfusion to the flank decreased as early as 5 minutes posttreatment with dexmedetomidine and persisted for at least 15 minutes, regardless of dexmedetomidine dose. CLINICAL RELEVANCE Dexmedetomidine decreased skin perfusion in cats, even at low doses. Veterinarians may elect for an alternative sedative medication when decreased skin perfusion is a concern.

Abstract OBJECTIVE Right dorsal colitis causes chronic colic associated with long-term treatment with nonsteroidal antiinflammatory drugs (NSAIDs). This study was designed to determine if NSAIDs could inhibit anion transporters that protect against intestinal mucosal injury in other species. ANIMALS 20 healthy horses. METHODS The effects of indomethacin (INDO) and firocoxib (FIR), on short-circuit current (Isc) in mucosa from the right dorsal colon (RDC) and right ventral colon (RVC) were measured in Ussing chambers by standard electrophysiological techniques. Immunohistochemical methods were used to detect apoptosis (caspase-3) with these NSAIDs and phenylbutazone (PBZ) and to locate the NKCC1 transporter. RESULTS The Isc in RDC and RVC incubated with INDO or FIR was increased almost 3-fold (P < .0001) by prostaglandin E2 (PGE2) through a system inhibited by loop diuretics (P < .0001). Although these findings and anion replacement studies were consistent with anion secretion, the RDC also displayed an Isc response suggestive of a unique transporter apparently absent in RVC or NSAID-free solutions. In RDC, FIR, INDO, and PBZ induced apoptosis in the lower half of crypts. However, significant differences in apoptotic index were recorded in the RDC between NSAID-treated and control tissues (no NSAID). CLINICAL RELEVANCE The effects of NSAIDs on Isc were consistent with reduced anion secretion, which could represent the pharmacological equivalent of the transport failure responsible for Cystic Fibrosis (CF) in other species. Failure of anion secretion could interfere with buffering acid from intraluminal fermentation, which could suggest a treatment target for right dorsal colitis.

Abstract OBJECTIVE To determine the effects of cardiac cachexia on the metabolomic profile in dogs with myxomatous mitral valve disease (MMVD). ANIMALS 3 groups of dogs with MMVD enrolled between November 30, 2018, and April 7, 2022: (1) Dogs with congestive heart failure (CHF) and cachexia (CHF-cachexia group; n = 10); (2) dogs with CHF that had no cachexia (CHF-no cachexia group; n = 10); and (3) dogs with asymptomatic disease (American College of Veterinary Internal Medicine [ACVIM] Stage B2) with no cachexia (B2 group; n = 10). METHODS Metabolomic profiles were analyzed from serum samples using ultra-high-performance liquid chromatography-tandem mass spectroscopy. Dogs in the 3 groups were compared, with statistical significance defined as P < .05 with a low false discovery rate (q < .10) and nominal statistical significance defined as P < .05 but q > .10. RESULTS Numerous metabolites were significantly (n = 201) or nominally significantly (n = 345) different between groups. For example, when comparing the CHF-cachexia vs CHF-no cachexia groups, lipids were the predominant metabolite differences, including many medium- and long-chain dicarboxylates and dicarboxylate acylcarnitines. For comparisons of the CHF-cachexia vs B2 groups and the CHF-no cachexia vs B2 groups, amino acids, nucleotides, and cofactors/vitamins were the predominant metabolite differences. CLINICAL RELEVANCE Some significant metabolite differences were identified between dogs with and without cardiac cachexia.

Abstract OBJECTIVE To compare small intestinal inflammation with gastric inflammation in horses with and without equine gastric glandular disease (EGGD), we evaluated endoscopic, macroscopic, and microscopic findings of the glandular stomach and microscopic findings of the small intestine. ANIMALS 36 horses. METHODS Horses underwent endoscopy and were scored for EGGD. After euthanasia, stomachs were collected and macroscopically evaluated. Normal pyloric mucosa, glandular lesions, and small intestinal (duodenum, mid-jejunum, and ileum) samples were collected and processed for microscopic examination. Cellular infiltrate was scored. Immunohistochemistry (CD3, CD20, and Iba-1) was performed on the ventral pylorus and small intestine of horses with mild to moderate lymphoplasmacytic infiltrate. A Spearman’s correlation coefficient was used to evaluate the relationship of EGGD grade with gastric glandular inflammation, and the relationships of cellular infiltrate type and severity among glandular stomach, duodenum, jejunum, and ileum. RESULTS Gastrointestinal inflammation was common, with gastric inflammatory infiltrate identified in 92%, duodenal inflammatory infiltrate in 83%, jejunal inflammatory infiltrate in 92%, and ileal inflammatory infiltrate in 92% of horses. Endoscopic evidence of gastric disease (hyperemia or EGGD grade ≥ 2/4) was not associated with the presence or severity of duodenal, jejunal, or ileal inflammation. Gastric lymphoplasmacytic inflammation grade ≥ 2 was associated with duodenal lymphoplasmacytic inflammation grade ≥ 2. This was a convenience sample of horses presenting for euthanasia. Medical history (including deworming history) was unknown. CLINICAL RELEVANCE Gastric lymphoplasmacytic inflammation is associated with duodenal lymphoplasmacytic inflammation but not more distal small intestinal inflammation. Intestinal inflammation is not associated with endoscopic findings (hyperemia or EGGD grade ≥ 2/4).