Infectious canine hepatitis associated with prednisone treatment

Abstract

An 11-week-old, female Alaskan husky dog housed outdoors in the Yukon, Canada, was diagnosed with infectious canine hepatitis. The predisposing factors in this puppy for such a rare disease included inappropriate vaccination program, potential contact with endemic wildlife, and immunosuppression due to prednisone treatment.

Case description

An 11-week-old, female Alaskan husky dog was presented at the Prairie Diagnostic Services (PDS) at the Western College of Veterinary Medicine (WCVM) for postmortem examination. Kept outdoors in the Yukon, this puppy and her 4 littermates had been vaccinated by the owner against canine distemper virus, canine adenovirus (CAV) type 2, canine parvovirus, and parainfluenza virus at 4 and 6 wk of age. At 7 wk of age, she and 2 littermates developed facial pyoderma and responded to antibiotic treatment (amoxicillin and clavulanate, dosage unknown) administered by the owner. At 9 wk of age the puppy developed facial pustules again and responded well to treatment with prednisone [0.9 mg/kg body weight (BW) PO, q12h] and cephalexin (20 mg/kg BW, PO, q12h) for suspected juvenile cellulitis. A week later, the puppy was presented with anorexia, depression, and facial edema, with no pyrexia. Biochemistry and complete blood (cell) count (CBC) obtained on in-house analyzers showed poorly regenerative anemia, thrombocytopenia, and markedly increased alkaline phosphatase (ALP) activity. Since there were concerns about canine erhlichiosis due to a history of traveling with dogs housed in the same kennel, doxycycline treatment was initiated. Specific testing for ehrlichiosis was not performed due to financial constraints. Vitamin K was also given. Prednisone and cephalexin treatments were discontinued. A week later, the puppy deteriorated and became increasingly depressed. She developed petechiation and hemorrhagic diarrhea. Prednisone therapy was initiated (same dosage as before). The puppy was found dead in the kennel within a day. Blood samples obtained shortly before death showed a mild non-regenerative anemia [hematocrit: 0.298 L/L, reference interval (RI): 0.365 to 0.573 L/L] and no reticulocytes detected on manual cell count (automated reticulocyte count unavailable), which was consistent with acute blood loss and chronic inflammation. The mean corpuscular hemoglobin concentration (MCHC) was mildly decreased (310 g/L, RI: 335 to 357 g/L), which could suggest iron deficiency, but this was considered less likely because mean corpuscular volume (MCV) was well within the reference interval. Small numbers of keratocytes were noted and were suggestive of fibrin strand injury. Marked thrombocytopenia (50.3 × 10⁹/L, RI: 200 to 900 × 10⁹/L) was present and was attributed to platelet clumping, as observed on the blood smear, but increased consumption due to disseminated intravascular coagulation could not be ruled out. There was a moderate left shift (segmented neutrophil count: 5.1 × 10⁹/L, RI: 3.0 to 10 × 10⁹/L; band cell count: 0.469 × 10⁹/L, RI: 0.0 to 0.1 × 10⁹/L), which was consistent with inflammation, and moderate lymphopenia (0.402 × 10⁹/L, RI: 1.2 to 5.0 × 10⁹/L), which was attributed to stress and/or exogenous glucocorticoids. Biochemistry showed moderate hyperbilirubinemia (10 μmol/L, RI: 1.0 to 4.0 μmol/L; fractionated bilirubin measurements were not available), markedly increased ALP (2623 U/L, RI: 9 to 90 U/L), mildly increased gamma-glutamyltransferase (GGT) (11 U/L, RI: 0 to 8 U/L), and markedly increased sorbitol dehydrogenase (SDH; 141.7 U/L, RI: 0.0 to 4.0 U/L) activities, which were consistent with hepatocellular injury and cholestasis. Hemolysis could have contributed to increased bilirubin measurement as well. Increased ALP activity may be due to the increased bone turnover (i.e., age-related), increased steroid-induced isoforms, and cholestasis. Activity levels of alanine aminotransferase (ALT) and glutamate dehydrogenase (GLDH) could not be evaluated due to interference by marked hemolysis. All reference intervals were generated from adult dogs.

Necropsy showed generalized icterus, marked hemorrhage in the thymus, cervical subcutaneous tissues, stomach, renal cortex, and brainstem, and petechiation at all levels of the intestines. The liver was diffusely enlarged and friable, and had a zonal pattern with alternating red and pale tan areas (Figure 1). Histopathology showed that the lymph nodes, thymus, brain, and small intestines were affected by multifocal hemorrhage. The bone marrow was severely hypocellular with decreased cellularity in all cell types. The liver had marked coagulation necrosis with neutrophilic infiltration. Many hepatocytes and Kupffer cells contained marginated chromatin and intranuclear inclusion bodies that were round to ovoid, magenta, and 4–8 μm in diameter (Figure 2). Based on these findings, the diagnosis of infectious canine hepatitis was made. Electron microscopy revealed viral particles resembling canine adenovirus in the nucleus of hepatocytes (Figure 3). Hepatocytes also showed strong nuclear staining on immunohistochemistry using antibodies against CAV-1 (Figure 4).

Figure 1.

Cross section of the liver shows multifocal zonal necrosis. Bar = 1 cm.

Figure 2.

Hematoxylin & eosin-stained liver sections show hepatic necrosis, intranuclear inclusion bodies with margination of chromatin (arrows), and cholestasis (arrowheads). Bar = 50 μm.

Figure 3.

Electron microscopy shows adenoviral particles within the nucleus.

Figure 4.

Hepatocytes show positive nuclear staining for CAV-1. Immunohistochemical staining for CAV-1. Bar = 50 μm.

Discussion

In dogs, there are 2 types of adenoviruses: canine adenovirus type 1 (CAV-1) and canine adenovirus type 2 (CAV-2). The CAV-1 may cause a severe generalized disease called infectious canine hepatitis (ICH), whereas CAV-2 typically causes a mild respiratory disease (1). Infectious canine hepatitis usually affects dogs less than 1 y of age (2) and is transmitted nasal-orally via direct contact, fomites, and ectoparasites (3). Viral replication initially occurs in the tonsils, followed by dissemination of viruses to other tissues and saliva, urine, and feces (1). CAV-1 is known to have tropism for vascular endothelium and hepatocytes (1). Virions assemble in the nucleus and form intranuclear inclusion bodies that may occupy the entire nucleus. Viruses are released by cellular lysis. Most infected dogs are asymptomatic or have mild tonsillitis (4). Animals that show clinical signs are presented with fever, lymphadenopathy, abdominal pain, icterus, anterior uveitis, or petechiation that may indicate disseminated intravascular coagulation. Death from ICH is sporadic (4). In the case presented here, the puppy showed signs associated with inflammation, liver disease, and dysregulation in coagulation, but lacked the classic ocular signs.

Diagnosis of ICH is often made on histopathology, as ante-mortem diagnosis is difficult. Other tests available include complement fixation, hemagglutination inhibition, enzyme-linked immunosorbent assay (ELISA), viral isolation, immunohistochemistry, and polymerase chain reaction (PCR) (1).

Neonates are protected against adenoviral infection by maternal antibody, the level of which generally begins to decline by 5 to 7 wk of age (3). Inactivated and modified-live vaccines (MLV) are both commercially available. Inactivated vaccine does not produce disease in dogs, but has to be given frequently to maintain protective titers (3). Life-long immunity is provided by MLV. Vaccination with CAV-1 MLV is associated with ocular and renal disease, whereas vaccination with CAV-2 MLV is not, but may sometimes result in mild respiratory signs (3). Since CAV-2 MLV is safer than CAV-1 MLV and cross-protects against CAV-1, it is the most commonly used vaccine.

Since the advent of the aforementioned vaccines, IHC has become a rare disease in dogs in Canada. The demise of the puppy in this case was likely due to a combination of factors. Firstly, appropriate handling of the vaccines given to the puppy cannot be verified because the vaccines were administered by the owner. Secondly, the puppy was vaccinated at 4 and 6 wk of age, a time during which interference by maternal antibodies was likely to have occurred. Thirdly, the primary series of vaccination was never completed. The recommended vaccination schedule for puppies less than 16 wk of age consists of 3 doses given at 6 to 8, 9 to 11, and 12 to 14 wk of age (5). Fourthly, prednisone treatment for puppy cellulitis might have resulted in immunosuppression, rendering the puppy more susceptible to infection than her littermates. It was previously shown that German shepherd dogs with experimentally impaired cellular immunity were more likely to die from adenoviral infection compared with control dogs (6). Fifthly, it is known that CAV-1 is endemic in the wildlife in Alaska and Yukon. Among 1122 wolves that were tested, more than 84% had antibodies against CAV-1 (7). Being housed outdoors, the puppy could have acquired the infection via contact with infected wolves or contaminated fomites. Since the entire litter was exposed to all of these predisposing factors, except for prednisone therapy, it is reasonable to assume that immunosuppression associated with prednisone therapy was a crucial contributing factor to development of fatal ICH in this puppy.

In summary, we report a case of ICH in the Yukon. Infectious canine hepatitis should be considered in young dogs with acute hepatic disease housed in rural areas of Alaska and Yukon. To our knowledge, this is the first clinical case report of ICH associated with prednisone treatment at immunosuppressive doses in the presence of the aforementioned predisposing factors. CVJ