Protozoal Hepatitis Associated with Immunosuppressive Therapy in a Dog

Abstract

A4-year-old, female spayed, Standard Poodle was examined for evaluation of presumed immune-mediated hemolytic anemia (IMHA) that had been refractory to therapy. Before referral, weakness and lethargy had been noted and a CBC performed by the referring veterinarian had shown a regenerative anemia with autoagglutination and spherocytosis. The hemoglobin concentration had been markedly reduced at 69 g/L (reference range, 120–180 g/L) and the reticulocyte count had been 35.3%. A PCV measurement taken by the referring veterinarian the same day that the CBC had been submitted was 24% (reference range, 37–55%). Serum biochemistry results revealed a slightly increased creatine kinase activity (611 U/L; reference range, 0–506 U/L), an increased aspartate aminotransferase activity (86 U/L; reference range, 15–55 U/L), an increased amylase activity (1,283 U/L; reference range, 250–1,060 U/L), and an increased bilirubin concentration (1.6 mg/dL; reference range, 0–0.24 mg/dL). There was no history of recent vaccination or drug administration before the onset of anemia. Treatment was with prednisonea (2 mg/kg PO q12h). However, after 1 week of treatment there was no significant improvement and the dog was referred. On physical examination, the dog was weak and had pale mucous membranes. Abdominal palpation and ultrasonography detected hepatosplenomegaly. Thoracic radiography did not reveal any abnormalities. A urine sample obtained by cystocentesis did not yield any growth on bacterial culture. A CBC revealed the hemoglobin concentration was 76 g/L (reference range, 120–180 g/L) and PCV 24% (reference range, 37–55%). A diagnosis of idiopathic IMHA was made and additional immunosuppressive therapy was added to the treatment regime (azathioprineb 1 mg/kg PO q24h and cyclosporinc at 10 mg/kg PO q24h). In an attempt to reduce the risk of thromboembolic disease, aspirind was also administered at 0.5 mg/kg PO q24h. The dog responded well to treatment and medications were gradually withdrawn. The cyclosporin was completely withdrawn after 9 weeks of treatment. At that stage, treatment was prednisone (0.7 mg/kg PO q24h) and azathioprine (1 mg/kg PO q48h). The aspirin dose was unchanged. All medications were discontinued after 1 year at which time the CBC was within normal limits. Approximately 2 months after discontinuing treatment, the dog once again had lethargy and pale mucous membranes. A CBC showed a hemoglobin concentration of 71 g/L (reference range, 120–180 g/L) and a PCV was 24% (reference range 37–55%). Autoagglutination and spherocytosis were present. Serum biochemistry was within normal limits apart from an increase in bilirubin concentration (1.4 mg/dL; reference range, 0–0.24 mg/dL). A diagnosis of a relapse of the idiopathic IMHA was made and immunosuppressive therapy was reinstituted with prednisone (1.7 mg/kg PO q12h), azathioprine (0.9 mg/kg PO q12h), aspirin (0.5 mg/kg PO q24h), and cyclosporin (8.5 mg/kg PO q24h). After 10 days of therapy, there had been a clinical improvement. The dose of prednisone was reduced to 1.1 mg/kg PO q12 h and the azathioprine dose was reduced to 0.9 mg/kg PO q48h. This treatment regime was continued for a further 7 days before the dog developed a poor appetite and lethargy. A CBC showed a hemoglobin concentration of 107 g/L (reference range, 120–180 g/L) and a PCV was 30% (reference range 37–55%). Serum biochemistry results revealed markedly increased activity of creatine kinase (4,714 U/L; reference range, 0–506 U/L), aspartate aminotransferase (2,542 U/L; reference range, 15–55 U/L), alanine aminotransferase (6,156 U/L; reference range, 15–110 U/L) alkaline phosphatase (>2,500 U/L; reference range, 10–160), and concentration of bilirubin (1.4 mg/dL; reference range, 0–0.24 mg/dL). Abdominal ultrasound examination revealed hepatomegaly with a diffusely hyperechoic hepatic parenchyma. Distal acoustic attenuation was noted. A urine sample obtained by cystocentesis yielded a heavy pure growth of Klebsiella pneumonia (>100,000 organisms/mL). Thoracic radiographs did not reveal any important abnormalities. IV fluid therapy was commenced together with antibiotic treatment (metronidazolee 7.5 mg/kg IV q12h, enrofloxacinf 10 mg/kg IV q24h, amoxicillin–clavulanic acidg 22 mg/kg IV q8h, and clindamycinh 10 mg/kg PO q12h). Activated clotting time was prolonged at 308 seconds (reference range, 90–110 seconds). A plasma transfusion was administered but approximately 36 hours after presentation, ventricular fibrillation and respiratory arrest occurred. Cardiopulmonary resuscitation was unsuccessful and the dog died. Ultrasound guided “tru-cut” biopsies were taken from the hepatic parenchyma within 10 minutes of death. On microscopic examination, the multiple “tru-cut” biopsies of liver were composed of hepatocytes, bile ducts, and Kupffer cells. There were locally extensive coalescing foci of coagulative hepatocytic necrosis that was predominantly centrilobular and midzonal, with infiltrates of neutrophils and hemorrhage. Multifocal hepatocytes contained numerous (20–100) basophilic, 2–3 μm diameter, globular to ovoid bodies resembling protozoal tachyzoites of Neospora caninum or Toxoplasma gondii (Fig 1). Many Kupffer cells contained intracytoplasmic hemosiderin, and some hepatocytes contained intracytoplasmic bile. Immunohistochemistry for T. gondii was negative, using a mouse monoclonal antibody to P30 antigen.i H&E histopathology of “tru-cut” biopsies of liver. Multifocal hepatocytes contain numerous (20–100) cytoplasmic basophilic, oval, 2–3 μm diameter, globular to ovoid bodies (Neospora caninum tachyzoites). Scale bar=10 μm. DNA was extracted from tissue sections from liver and subjected to PCR using the JB–SF nested PCR assay.1 A PCR product was obtained confirming the presence of N. caninum DNA in the liver. This case report represents an apparently uncommon manifestation of neosporosis in the dog. The most common forms of canine neosporosis are myositis-polyradiculoneuritis and encephalomyelitis.2,3 Given the markedly increased activity of creatine kinase and aspartate aminotransferase, it is quite possible that this dog could have been affected by a concurrent protozoal myositis. However, no samples of muscle tissue were obtained to investigate this possibility. Similarly, no biopsies of the central nervous system and spinal cord were taken. However, there was no clinical support for encephalomyelitis or polyradiculoneuritis. There have been individual reports of Neospora-associated dermatitis, myocarditis, pneumonia, and septic peritonitis in dogs.4–6 As a full necropsy was not carried out on this dog, the possibility of a more disseminated infection could not be ruled out. Neospora-associated hepatitis has previously been reported in dogs.7,8 There has been one previous report of a dog with cutaneous neosporosis that occurred in association with immunosuppressive therapy.9 This dog was being treated for pemphigus foliaceus with prednisone and azathioprine. Canine hepatic sarcocystosis has been reported10 and could be a differential diagnosis in this dog. However, to the authors' knowledge, neither Sarcocystis neurona or Sarcocystis canis has been reported in New Zealand and the dog had not travelled outside of the country during its lifetime. Cyclosporin is being used with increasing frequency to treat a variety of immune-mediated disease processes in the dog. Anecdotally, the drug appears to be efficacious but there are no published reports of its efficacy in the treatment of canine IMHA. In the author's institution, this drug has been used successfully in several cases of refractory IMHA. Cyclosporin is a potent suppressor of cell-mediated immunity and as such, has the potential to inhibit the body's immune defence against infection by Neospora. Dogs are the definitive host of N. caninum11 with a wide range of intermediate hosts including cattle, sheep, and deer.12 Dogs are infected by eating tissue that contains encysted organisms such as raw beef13 and bovine fetal membranes.14 The source and timing of infection in this dog was not known. The dog lived in an urban environment and so the likelihood of exposure to bovine placental material would have been low. There was no known ingestion of raw meat associated with this dog's hepatitis. However, it was not known whether ingestion of raw meat occurred earlier in the dog's life and Neospora serology was not previously performed. The fact that the dog appeared to tolerate the initial cyclosporin treatment raised the possibility that infection may have occurred subsequent to this initial course. The dose of cyclosporin that should be used for the treatment of immune-mediated disease in the dog is not known with certainty. A dose of 10 mg/kg PO q24h has been anecdotally suggested and this dose is routinely used at the author's hospital without obvious known side effects apart from those in this dog. It is generally recommended that drug doses are adjusted based on serum concentrations although there are no evidence-based recommendations nor broad consensus about the manner and timing of cyclosporin level testing. Serum cyclosporin concentrations were not measured in this case as the drug was planned to be used for a relatively short time course with a reducing dose regime. The optimal duration of cyclosporin treatment for treatment of immune-mediated disease is not known. It is possible that the fatal protozoal hepatitis in this dog could have been associated with toxic levels of cyclosporin. However, the dosage used for the second course of cyclosporin was lower than the first course because the dog's weight had increased and the same oral dose was administered. The duration of the treatment course was also relatively short with a much longer course being previously well tolerated without any of the recognized adverse effects of cyclosporin treatment such as vomiting, anorexia, diarrhea, gingival hyperplasia observed. It is also possible that the fatal protozoal hepatitis in this dog may have been associated with the other immunosuppressive drugs used or indeed the combination of these drugs with cyclosporin. This case report raises an important concern regarding the use of potent suppressors of cell-mediated immunity in the dog, especially as the use of cyclosporin appears to be becoming anecdotally more common in the treatment of immune-mediated disease. The possibility of inducing serious complications such as the fatal protozoal hepatitis reported in this case should be borne in mind. The authors suggest that consideration should be given to performing protozoal serology on dogs before embarking on cyclosporin treatment or indeed other immunosuppressive therapy. In addition, careful attention should be taken to avoid possible sources of protozoal infection such as raw meat ingestion during treatment. This case report also highlights the need for further research into the areas of seroprevalence of protozoal disease and therapeutic monitoring of novel immunosuppressive agents such as cyclosporin. aAPO-Prednisone, Apotex NZ, Aukland, New Zealand bImuran, GlaxoSmithKline, Aukland, New Zealand cAtopica, Novartis, Aukland, New Zealand dCartia, GlaxoSmithKline eMetronidazole-Claris, Claris Lifesciences Ltd, Ahmedabad, India fBaytril, Bayer, Aukland, New Zealand gAugmentin, GlaxoSmithKline hAntirobe, Pfizer, Aukland, New Zealand iNovocastra Ltd, Newcastle on Tyne, UK The authors thank Joel Barrett (UTS) for running the PCR.