Sophie Vandenabeele

Equine multisystemic eosinophilic epitheliotropic disease: A case report and review of literature

CASE HISTORY: A 2-year-old Standardbred gelding presented with a history of fever over 1 week, anorexia and skin lesions on all four legs. The lesions were associated with severe pruritus and oedema, and there was no response to therapy. CLINICAL FINDINGS: The horse was in poor body condition, was lethargic and severely pruritic. Skin lesions consisted of diffuse alopecia and crusting of the distal extremities. Initially it was slightly febrile, but subsequently its temperature increased up to 40°C. Ten days after admission it developed profuse watery diarrhoea and the skin lesions progressed. Skin biopsies revealed superficial and deep perivascular dermatitis with lymphoplasmacytic and eosinophilic predominance. Based on the poor prognosis the horse was subject to euthanasia. PATHOLOGICAL FINDINGS: The most notable lesions included ulcerative gastritis, typhlitis and colitis with prominent oedema of the intestines, marked subcutaneous oedema and severe thickening of the large bile ducts. Histopathology showed marked eosinophilic and lymphoplasmacytic infiltration of various tissues including the skin, gastrointestinal tract, mesenteric lymph nodes, large bile ducts, pancreatic duct and kidney. Immunohistochemistry revealed a clear predominance of CD3-positive cells in the lymphocytic infiltrations. DIAGNOSIS: Based on the clinical findings and histopathology a diagnosis of multisystemic eosinophilic epitheliotropic disease (MEED) was made. CLINICAL RELEVANCE: Multisystemic eosinophilic epitheliotropic disease is rare in horses, and usually chronic. In the current case the horse showed an apparently acute onset with high fever and rapid clinical deterioration. A diagnosis of MEED should be considered in horses presenting with weight loss and skin lesions with or without fever. A final diagnosis is based on histological results of biopsy specimens from affected organs.

Idiopathic sterile inflammation of the epidural fat and epaxial muscles causing paraplegia in a mixed-breed dog

CASE DESCRIPTION A 4-year-old sexually intact male mixed-breed dog was evaluated because of clinical signs of acute-onset pelvic limb ataxia, rapidly progressing to paraplegia with severe spinal hyperesthesia. CLINICAL FINDINGS General physical examination revealed pyrexia, tachycardia, and tachypnea. Neurologic examination demonstrated severe spinal hyperesthesia and paraplegia with decreased nociception. Magnetic resonance imaging revealed extradural spinal cord compression at T13-L1 and hyperintense lesions on T1- and T2-weighted images in the epaxial musculature and epidural space. TREATMENT AND OUTCOME Decompressive surgery, consisting of a continuous dorsal laminectomy, with copious lavage of the vertebral canal was performed. Cultures of blood, urine, and surgical site samples were negative. Histologic examination results for samples obtained during surgery demonstrated suppurative myositis and steatitis. These findings confirmed a diagnosis of sterile idiopathic inflammation of the epidural fat and epaxial muscles with spinal cord compression. The dogs neurologic status started to improve 1 week after surgery. After surgery, the dog received supportive care including antimicrobials and NSAIDs. The dog was ambulatory 1 month after surgery and was fully ambulatory despite signs of mild bilateral pelvic limb ataxia 3 years after surgery. CLINICAL RELEVANCE Although idiopathic sterile inflammation of adipose tissue, referred to as panniculitis, more commonly affects subcutaneous tissue, its presence in the vertebral canal is rare. Specific MRI findings described in this report may help in reaching a presumptive diagnosis of this neurologic disorder. A definitive diagnosis and successful long-term outcome in affected patients can be achieved by decompressive surgery and histologic examination of surgical biopsy samples.

Presumed phenobarbital-induced Stevens–Johnson syndrome in a 4-year-old female Great Dane*

A 4-year-old female Great Dane weighing 57 kg was evaluated for a 6-month history of cluster seizures. The referring veterinarian performed no further investigations, and no treatment was initiated. General physical examination at the time of presentation was unremarkable. On neurological examination, the dog was showing generalized tonic clonic seizures with autonomic involvement (urination, defecation, salivation) only 5 10 minutes apart with regaining of consciousness in between seizures. Complete blood count and serum biochemistry profile were within normal limits. Diazepam (Valium, Roche, Basel, Switzerland) was administered intravenously (IV) (0.25 mg/kg bodyweight (BW)) successfully ending seizure activity after two boluses. The owner declined magnetic resonance imaging of the brain and cerebrospinal fluid examination. Due to the aggressive character of the dog, further hospitalization was impossible and the dog was discharged on phenobarbital treatment (Phenoleptil, Kela Veterinaria, Hoogstraten, Belgium) (5 mg/kg BW/12 h per os (PO)). Since initiation of treatment, no more seizures were witnessed at home. The dog was presented again after 6 weeks for an acute onset of skin lesions, which appeared to be rapidly progressive. At presentation, the dog was lethargic and was walking very slowly and cautiously, appearing markedly uncomfortable. General physical examination revealed a mild tachycardia and hyperemic mucous membranes. Multiple, symmetrical, painful, vesiculobullous lesions with an erythematous peripheral rim were present at the level of the groin, ventrum, axillae and sternum (Figures 1 and 2). These lesions had a positive Nikolskiy sign (the artificial extension of a blister or ulcer induced by digital pressure to adjacent mucous membrane or skin; Gross et al. 2005). A linear erosive lesion was present at the lateral canthus of the left eye, and there were multiple, ill-defined, painful ulcerative lesions on the footpads of the four paws (Figures 3 and 4, respectively). A single, well-delineated vesicle surrounded by an erythematous rim was seen at the left upper labial mucosa. Macroscopically, the main differential diagnoses were a drug reaction such as erythema multiforme (EM), Stevens Johnson syndrome (SJS) or epidermolysis bullosa acquisita (EBA). However, EBA does not exhibit a positive Nikolskiy sign (Grando et al. 2003). The history, clinical distribution of the lesions, aspect of the flat lesions and the positive Nikolskiy sign leaded to a presumptive diagnosis of SJS. To establish this diagnosis, histopathological samples consisting of punch biopsies were obtained from three lesions on the ventrum. According to a previously established method for estimating the probability of an adverse drug reaction (ADR; Naranjo et al. 1981), a probable ADR to phenobarbital was diagnosed (Table 1) and phenobarbital treatment was therefore abruptly stopped. The dog was loaded orally with potassium bromide (Libromide, Dechra Veterinary Products, Bladel, the Netherlands) using a 5-day loading schedule according to Dewey (2008), consisting of 125 mg potassium bromide/kg BW/day, given in 6 doses for 5 consecutive days. Treatment was initiated with gabapentin (Neurontin, Pfizer, city, Belgium) (10 mg/kg BW/8 h PO) for additional seizure management. Prednisolone therapy (Prednisolone, Kela Laboratoria, Hoogstraten, Belgium) (0.3 mg/kg BW/day PO) was initiated for the skin lesions. After termination of the potassium bromide loading schedule, the dog was continued on a maintenance dose (20 mg/kg BW/12 h). The owner was advised to treat the skin lesions once daily topically with povidone-iodine (Iso-Betadine, Meda Pharma, city, Belgium), and the footpath lesions with chlorhexidine (Hibiscrub, M€ onlycke, city, Belgium) bathing. All three skin samples demonstrated identical morphologic changes on histopathology. There was cytotoxic interface dermatitis with the presence of multiple apoptotic keratinocytes in all the layers of the

Atypical canine recurrent alopecia

different dogs. We would like to report the development of an opensource software platform to analyse the punctiform topographical distribution of canine nodular cutaneous lesions directly connected with a database that contains clinical, anamnestic, histopathological, therapeutic and other useful information. A specific relational Web database for collection of the characteristics and location of each lesion was created using open-source software, including the operating system (Linux Ubuntu 10.10) MySQL AB version 5.1. A relational database management system was used to create the database, and a user-friendly graphical interface was obtained from the Apache HTTP Server version 2.2 and Symfony version 1.2.4 (Web application framework written in PHP). A Web GIS platform was integrated into the Web database to allow the registration of topographical data. Nodular skin lesions (n = 387) from 350 dogs were retrieved from the Histopathology Diagnostic Service of the Department of Food Science, University of Udine (Italy). A complete dermatological description was obtained using a multiple-choice format questionnaire to standardize data collection. The accurate identification of the topographical location of each lesion was performed by the clinicians by indicating the lesion on a paper map drawn of canine mesomorphic breeds. The cutaneous location of the lesion was inserted into the Web GIS system with a GeoTIFF image corresponding to the same model. The distribution pattern was analysed by two spatial statistical methods, i.e. average nearest neighbour and kernel density function. The results of the nearest neighbour analysis showed a clustering pattern of sites of onset of lesions examined by query. In particular, the clustering was evident for the following categories of nodular lesion: neoplastic (corrected Z-value, 12.04; P-value <0.001); benign neoplastic (corrected Z-value, 4.81; P-value <0.001); malignant neoplastic (corrected Z-value, 4.77; P-value <0.001); and inflammatory lesions (corrected Z-value, 3.1; P-value <0.01; Figure 1). Through the kernel density estimation, it was possible to prepare topographical maps with remarkable clarity. The maps were easy to interpret, and they showed that the highest relative density of all nodular lesions (n = 387) was concentrated on the front part of the animal and in the scrotal region. The distribution of benign neoplasms (n = 138) showed maximal values of relative density on the dorsal portion of the skull. Malignant neoplastic nodular lesions (n = 139) showed maximal values of relative density in the topographical area of the chest, the scrotum and the posterior surface of the thigh. Inflammatory nodular lesions (n = 32) showed a strong clustering on the dorsal surfaces of the metacarpal–phalangeal region and rostral portion of the dorsal nasal region. The GIS method used in this study has shown interesting findings. In particular, the use of a GIS system in mapping lesions could be considered as a new approach for collection of clinical data from skin lesions. The creation of density maps of the relative frequency of nodular skin lesions could be useful for helping veterinary clinicians to have a general idea about the possible differential diagnoses when presented with such lesions. This technique could be a valuable aid for training, for didactic purposes and for researchers. Furthermore, it allows statistical analysis of nodular skin diseases, including, for example, factors such as their anatomical distribution, breed association and disease incidence.