Portal de Periódicos da CAPES

Concurrent Hepatic Copper Toxicosis and Fanconi's Syndrome in a Dog

2008; Wiley; Volume: 22; Issue: 1 Linguagem: Inglês

10.1111/j.1939-1676.2007.0040.x

1939-1676

Tracy Hill, Edward B. Breitschwerdt, Thomas E. Cecere, Shelly L. Vaden,

Drug-Induced Hepatotoxicity and Protection

A3-year-old male castrated West Highland White Terrier was presented to the Veterinary Teaching Hospital at North Carolina State University with a 1-week history of intermittent vomiting, polyuria and polydypsia, progressive anorexia, and lethargy. No treatment was administered by the primary veterinarian before referral. The dog was adopted as a puppy and had lived exclusively in North Carolina. The dog received no other medications and had no previous illnesses, except for a transient decrease in appetite 1 year earlier. On physical examination, the dog weighed 8.2 kg, had a body condition score of 6/9, but was approximately 5% dehydrated. Hematologic abnormalities included neutrophilia (14,293 cells/μL, reference range 2,529–12,876 cells/μL) with a left shift (642 band neutrophils/μL) and immature granulocytes (161/μL). Biochemical abnormalities included increased ALT activity (456 U/L; reference range, 16–73 U/L), hyperbilirubinemia (0.3 mg/dL; reference range, 0–0.2 mg/dL), hypokalemia (3.8 mEq/L; reference range, 3.9–5.2 mEq/L), hyperchloremia (121 mEq/L; reference range, 104–117 mEq/L), and increased lipase activity (715 U/L; reference range, 22–216 U/L). Urinalysis findings included a specific gravity of 1.022, pH of 8, 2+ proteinuria, 2+ ketones, 2+ blood, 4+ glucose, 0–2 coarse granular casts/hpf, 0–5 white cells/hpf, and 5–10 red cells/hpf. The urine protein : creatinine ratio was 1.7. A urine sample submitted for bacterial culture yielded no growth. There was mild metabolic acidemia (blood pH 7.26; reference range, 7.36–7.47), bicarbonate 16.2 mEq/L (reference range, 19.8–26.2 mEq/L), and PCO2 36 mmHg (reference range, 30–40 mmHg), on a venous blood sample. Abdominal ultrasonographic findings included periportal lymph node enlargement, microhepatica, and mildly hyperechoic kidneys bilaterally. Fasting and postprandial serum bile acid concentrations were within reference range. The dog was treated with an IV infusion of lactated Ringer's solution containing 30 mEq/L KCl for dehydration, hypokalemia, and metabolic acidosis. Within 48 hours of hospitalization, sodium bicarbonate at 3 mEq/kg/d was administered as a continuous rate infusion to correct the persistent metabolic acidosis. In addition, the dog was treated with a metoclopramide constant rate infusion of 1 mg/kg/d and famotidine 0.5 mg/kg IV q12h. During the 1st 48 hours of hospitalization, the dog remained anorexic and frequently vomited bile; dolasetron 0.6 mg/kg IV q24h and sucralfate 500 mg PO q8h were administered. The dog was supplemented nutritionally by nasoesophageal feeding a liquid diet (Perativea), the continuous rate of which was adjusted to minimize vomiting. Serial assessment of urine dipstick tests and serum glucose measurements indicated that the dog was persistently ketonuric and glucosuric, with normal serum glucose concentrations. Despite a metabolic acidosis, urine pH ranged from 7.0 to 8.0. The findings of proteinuria, glucosuria with normoglycemia, and hyperchloremic metabolic acidosis with alkaline urine supported a diagnosis of Fanconi's syndrome. Results of a urine metabolic profile, performed at the University of Pennsylvania, also were consistent with Fanconi's syndrome with severe generalized amino aciduria and marked glucosuria. By the 4th day of hospitalization, the dog became febrile (103.6°F). An abdominal ultrasound examination indicated mild thickening of the gallbladder wall. Antibiotic therapy was initiated with ampicillin and sulbactam (Unasyn,b 30 mg/kg IV q8h); the fever resolved within 36 hours. Because of the dog's refractory vomiting and unknown underlying hepatic pathology, an exploratory laparotomy was performed on the 5th day of hospitalization. Biopsy specimens were taken of the liver, stomach, duodenum, jejunum, and left kidney. The gallbladder was aspirated and 1 aliquot of bile was submitted for cytologic analysis; another aliquot was submitted for aerobic microbial culture. Gastrostomy and jejunostomy tubes were placed. Recovery from surgery was uneventful. Perativea liquid diet was administered continuously through the jejunostomy tube throughout the remainder of hospitalization without incident. The volume administered was gradually increased to basal energy requirements. Ketonuria resolved within 24 hours after feeding basal energy requirements. Dexamethasone (0.08 mg/kg IV q24h), s-adenosylmethionine (22 mg/kg PO q24h), and ursodeoxycholic acid (15 mg/kg PO q24h) were administered on day 5 postoperatively. Within 24 hours, steady improvement was observed; the vomiting ceased, the dog was more active and alert, and began eating. IV bicarbonate supplementation was necessary to maintain a near normal serum pH. The bile fluid was cytologically unremarkable and bacterial growth was not observed. Lymphofollicular gastritis, which was attributed to refractory vomiting, and mild lymphocytic enteritis were observed histologically on full-thickness stomach and intestinal biopsy specimens. Centrilobular pyogranulomatous hepatitis, characterized by infiltrates of neutrophils, macrophages, and fewer lymphocytes and plasma cells associated with single cell hepatocyte necrosis, was present in liver wedge biopsy specimens. Abundant copper accumulation was present within centrilobular hepatocytes and macrophages (Fig 1). Copper was quantified at 1186 ppm dry weight. Histopathologic abnormalities in the kidney included diffuse mild to moderate tubular atrophy, multifocal tubular epithelial vacuolation, and tubular regeneration. Additionally, there was evidence of abnormal copper accumulation within vacuolated renal tubular epithelium (Fig 2). Liver, dog. (A) Centrilobular zone with individual hepatocellular necrosis (arrow) and infiltrates of macrophages, neutrophils, lymphocytes, and plasma cells. H&E stain. Scale bar = 50 μm. (B) Same section as A with numerous well-demarcated, red–brown copper-containing granules present in the cytoplasm of macrophages and hepatocytes. Rhodanine stain. Scale bar = 50 μm. Kidney, dog. (A) Tubules lined by plump vacuolated epithelial cells (arrow) or mildly atrophied epithelium (arrowhead). H&E stain. Scale bar = 20 μm. (B) Renal tubular epithelial cells containing multiple cytoplasmic well-demarcated, red–brown copper-containing granules. Rhodanine stain. Scale bar = 20 μm. The dog gradually was transitioned to oral medications administered through the gastrotomy tube, including 971 mg of bicarbonate q12h, s-adenosylmethionine, ursodeoxycholic acid, prednisone 0.6 mg/kg q12h that was tapered over the next 4 weeks, famotidine 0.6 mg/kg q12h for 2 weeks, amoxicillin–clavulanic acid 12 mg/kg q12h for 2 weeks, ciprofloxacin 8 mg/kg q12h for 2 weeks, and Marinc 1 medium dog tablet daily for 2 months. d-Penicillamine was given at a dosage of 10 mg/kg PO q12h for 3 months, with no reported adverse effects. After 2 weeks of treatment, the dog was no longer proteinuric, glucosuric, or ketonuric. Venous pH was 7.36 and bicarbonate was 26.6 mEq/L. ALT activity had decreased from 456 to 81 U/L. Ursodeoxycholic acid, s-adenosylmethionine, Marin, bicarbonate, and d-penicillamine therapy were continued. Monthly venous blood gas analysis disclosed normal pH while receiving oral bicarbonate supplementation. At the 3-month recheck examination, the bicarbonate dosage was reduced by half and d-penicillamine was discontinued. Repeat biopsies of the liver and kidney were declined by the owner. Zinc acetate was prescribed (10 mg/kg orally q12h); s-adenosylmethionine, ursodeoxycholic acid, and Marin were continued. Three weeks after decreasing the oral bicarbonate supplementation, venous pH was 7.4 and bicarbonate was 27.9 mEq/L and bicarbonate supplementation was discontinued, but zinc acetate supplementation was continued indefinitely. Thirteen months after initial diagnosis and treatment, the dog was doing well with no clinical signs reported. Copper storage diseases (CSDs) have been described in several breeds, including the Bedlington Terrier, West Highland White Terrier, Skye Terrier, Doberman Pinscher, Labrador Retriever, and Dalmatian, as well as other species, including humans, rats, and sheep. With CSD, copper accumulates in the liver, leading to hepatitis and eventually cirrhosis of the hepatic parenchyma.1 As yet, the genetic basis of CSD has been elucidated only in the Bedlington Terrier, where it is related to a defect in the MURR-1 gene.2 CSD in the Bedlington Terrier results in copper accumulation in the liver as well as renal cortical tissue in some patients.3 We describe a West Highland White Terrier with CSDand concurrent Fanconi's syndrome, which resolved after copper chelation therapy. A genetic mutation that causes abnormal copper accumulation has yet to be identified in West Highland White Terrier. Unlike the Bedlington Terrier, there does not seem to be a correlation between age and hepatic copper concentration in the West Highland White Terrier. In addition, the total hepatic copper concentration is lower in the majority of affected West Highland White Terriers; most West Highland White Terriers have copper concentrations <1,500 ppm dry weight.4, 5 Copper concentrations up to 500 ppm dry weight may occur in normal dogs.6 Thornburg describes the histopathologic lesions of copper toxicosis in West Highland White Terriers as being characterized by multifocal centrilobular hepatitis and cirrhosis. In this dog, the findings were consistent with previous reports, including multifocal centrilobular hepatitis and a copper concentration of 1,186 ppm. It is unclear whether copper toxicosis was the direct cause of Fanconi's syndrome in this dog, because several medical therapies, including antibiotics, steroids, and penicillamine, may have contributed to resolution of Fanconi's syndrome regardless of treating copper toxicosis. The presence of copper in the renal tubules and resolution of Fanconi's syndrome after copper chelation therapy suggest that copper toxicosis may be a cause of Fanconi's syndrome in dogs. In other reports, dogs with copper toxicosis had abnormalities indicative of proximal tubular dysfunction.7, 8 In a study of 10 Dalmatians with copper toxicosis, 3 dogs had proteinuria without pyuria, 2 had glucosuria with normoglycemia, and 1 had renal tubular necrosis with granular casts.7 In a separate report, a 1.5-year-old Dalmatian had copper toxicosis with a positive metabolic screen for Fanconi's syndrome.8 This dog was euthanized because the dog continued to decline clinically. In the dog reported here, glucosuria, metabolic acidosis, amino aciduria, and copper accumulation in the renal tubules were documented with resolution of Fanconi's syndrome after copper chelation therapy. Wilson's disease is a CSD in humans in which copper accumulates to toxic concentrations in the liver and secondarily in the central nervous system and kidneys because of a mutation in the ATP7B gene, which normally allows for copper excretion into the bile and for production of ceruloplasmin. As a result, patients with Wilson's disease have copper accumulation in the liver and in other tissues, including the brain, kidney, red blood cells, and eye. Neurologic signs develop including speech disorders and dysphagia, abnormal and uncoordinated gait, and tremors.9 Renal complications, including urolithiasis and Fanconi's syndrome, have been reported in human patients with Wilson's disease.10, 11 As is reported in association with Wilson's disease in human patients, this dog may represent a subset of patients with copper storage disease that have concurrent renal tubular dysfunction in association with copper accumulation in the proximal tubular epithelium. Several types of proximal renal tubular dysfunction have been described in association with Wilson's disease in humans: failure of renal acidification, amino aciduria, glucosuria, and phosphaturia.11-15 Resolution of proximal tubular dysfunction also may occur after treatment for copper toxicosis with penicillamine.11, 13-15 A renal biopsy in 1 patient, with prior documentation of renal tubular copper accumulation, demonstrated normal proximal tubular ultrastructure 2 years after initiation of penicillamine therapy. Penicillamine therapy was discontinued because of adverse effects, including glomerulonephritis and systemic lupus erythematosus. Eighteen months later, the patient again showed signs of Wilson's disease, and a renal biopsy was repeated. Electron-dense bodies, consistent with copper-bound metalloprotein, were evident in the subapical cytoplasm of the tubular cells, although copper quantification was not performed to confirm increased copper concentrations in the kidney.11 The effect of copper loading has been examined in the rat kidney as a model for copper toxicosis in humans and other species. Rats supplemented with excessive copper, either by injection or by dietary supplementation, develop copper staining in the liver and in the proximal convoluted tubular epithelium.16-18 Haywood demonstrated an increase in copper content of the kidney in rats fed excess copper as well as copper staining granules confined to the proximal tubule. There also were degenerative changes of the tubular cells as well as copper-staining debris in the tubular lumen, suggesting active exocytosis of copper-bound metallothionein.17 A later report of copper loading in the rat kidney described increased copper in renal tubular lysosomes. With time and increased copper concentrations, there was progressive nuclear degeneration in proximal tubular cells and disruption of the mitochondrial membrane.18 These findings are consistent with the observation that copper acts as a prooxidant, disrupting cell membranes and damaging DNA. Eventually, the rats extruded copper-stained lysosomes and copper-laden pinocytotic vesicles into the tubular lumen. After this time point, the copper concentration in the kidney began to decrease and the tubular epithelium recovered to nearly normal. The localization of copper to the renal tubular epithelium, later exocytosis of copper-bound organelles, and recovery of the tubular epithelium suggest a mechanism by which the rat seems able to cope with increased dietary copper intake. Copper also may alter the function of Na-K-ATPase in the proximal tubular epithelium. In vitro, for both rat kidney tissue homogenate and rat synaptic plasma membrane, copper has an inhibitory effect on the function of this important enzyme in a concentration-dependent manner.19, 20 As copper stores accumulate in the canine liver, the kidney may attempt exocytosis of excess copper as occurs in the copper-loaded rat. Exocytosis of copper-bound organelles in the canine and human kidney may be less effective than in the rat, because tubular debris is not described in any published reports of copper-stained kidneys from rats. Because copper accumulates in the kidney as well as the liver, it then may have several effects that ultimately lead to proximal tubular dysfunction and Fanconi's syndrome: necrosis and apoptosis of epithelial cells because copper acts as a pro-oxidant, disrupting mitochondrial membranes and DNA, inducing inflammation that may affect epithelial function, and inhibiting the function of Na-K-ATPase in a concentration-dependent manner that would alter transport mechanisms in the proximal tubule. These effects all may lead to decreased reabsorption of glucose, amino acids, phosphate, and bicarbonate from the tubular lumen. These may be the mechanisms by which copper toxicosis in this West Highland White Terrier resulted in proximal tubular dysfunction characterized as Fanconi's syndrome. Few controlled studies are available in human and canine medicine that describe renal pathology with copper toxicosis or the effect of copper chelation therapy. Additional study is needed to definitively confirm a link between Fanconi's syndrome and copper storage disease in dogs as suggested by this dog, and to elucidate the nature of that association. In dogs with suspected copper storage disease and evidence of tubular dysfunction that are undergoing liver biopsy, it may be warranted to perform a renal biopsy to allow for histopathologic examination of the renal parenchyma, as well as renal copper quantification, which was not performed here. Sequential urine metabolic screening or urinalyses with protein quantification may be a useful diagnostic and therapeutic monitoring tool in those patients with evidence of tubular dysfunction. Additionally, liver and kidney biopsies after 3 months of d-penicillamine therapy may have been informative to document the histological response to treatment. The authors gratefully acknowledge the assistance of the veterinarians involved in the treatment and referral of the dog. Our appreciation is extended to Dr Herman Jeffers, Dr Karyn Harrell, and Dr Sally Bissett. aPerative Specialized Nutrition, Abbot Laboratories, Ross Products Division, Columbus, OH bUnasyn, Pfizer Roerig, Pfizer Inc, New York cMarin, Nutramax Laboratories Inc, Edgewood, MD