Acute Liver Failure Because of Wilson Disease With Overlapping Autoimmune Hepatitis Features

Abstract

Acute liver failure (ALF) is a severe and sudden onset of hepatocyte dysfunction, which could progress to multiorgan failure and death (1). Wilson disease (WD), an autosomal recessive disease of copper transport, is owing to mutation in ATP7B(2). Establishing the diagnosis of WD in the setting of ALF is critical because it allows treatment for prompt removal of serum copper and stabilization of the patient to facilitate lifesaving orthotopic liver transplantation (OLT). Autoimmune hepatitis (AIH) is a progressive inflammatory liver disorder characterized serologically by high levels of transaminases and immunoglobulin G, and presence of autoantibodies, and histologically by interface hepatitis in the absence of known etiology (3). Unlike patients with ALF owing to WD, patients presenting with ALF owing to AIH may respond better to medical treatment with immunosuppressive agents, therefore circumventing the need for liver transplantation. Herein we report a case of ALF caused by WD showing strong autoimmune features. CASE A 15-year-old girl was transferred to our institution for further evaluation of coagulopathy, jaundice, and biochemical evidence of hepatocellular injury. The patient had been in good health until 2 months before presentation when she developed jaundice, ascites, and bilateral leg edema. Family history was negative for metabolic and inherited liver diseases. Physical examination revealed a conscious and oriented adolescent, jaundiced with mild ascites, palpable liver and spleen. Laboratory investigation revealed abnormal liver function tests, hemolytic anemia, and thrombocytopenia and coagulopathy (Table 1). Serum ammonia was 61 μg/dL (reference range [NR] = 19–60). Coombs test and markers for hepatitis B, A, and C viruses (HBV, HAV, HCV) were negative. Ceruloplasmin was low, 13.7 mg/dL (NR = 25–62); serum copper was 65 μg/dL (NR = 50–150); urinary copper 1600 μg/24 h (NR ≤ 50); anti-nuclear antibodies (ANA) were positive with a titre of 1:320, anti-smooth muscle antibodies (SMA) were also positive 1:160 (1:40 dilution as a cutoff of positivity), whereas anti-mitochondrial antibody (AMA) and anti–liver–kidney microsome antibody (LKM) were negative. Immunoglobulin G was 2970 mg/dL (NR = 751–1560), complement C3 and C4 were 9.24 mg/dL (NR = 79–152) and 2.34 mg/dL (NR 16–38), respectively. Abdominal ultrasound demonstrated hepatomegaly with diffuse echogenicity of the liver, moderate splenomegaly, and moderate ascites. Slit-lamp examination revealed the presence of Kayser–Fleischer rings. A core biopsy of liver was obtained and demonstrated diffuse parenchymal remodeling with a severe lobular disarray, nodular regeneration with hepatocyte ballooning, neocholangiolar proliferation, diffuse lymphocyte infiltrates with formation of nodular lymphocyte aggregates with T-cell phenotype, and plasma cell infiltrate. The copper stain was negative. Liver copper content was 388 μg Cu/g dry tissue (NR < 70). DNA analysis of ATP7B showed the compound heterozygous state for the already described c.2532delA and c.3061-1 G->A mutations.TABLE 1: Laboratory parametersThe presence of conclusive data in support of WD and of features suggestive for AIH prompted us to initiate a therapeutic trial with prednisone 60 mg/day and penicillamine initial dose 150 mg/day that was gradually increased to 750 mg/day. The patient also required multiple blood transfusions and vitamin K administration; diuretics, antibiotic prophylaxis, and plasmapheresis were also initiated. Nevertheless, her condition deteriorated with signs of sleepiness and further alteration of laboratory tests (Table 1). Serum ammonia was 82 μg/dL (NR = 19–60). Magnetic resonance imaging demonstrated T1-weighted hyperintensities in the globus pallidus bilaterally, hypothalamic region, cerebral peduncle, and adenohypophisis, suggesting the presence of hepatocerebral degeneration. According to the Wilson index for predicting mortality (4) we calculated the risk, obtaining a score of 11, showing high risk of mortality (Table 2). The patient was transferred to a transplant center where she underwent successful OLT.TABLE 2: Patient's evaluation according the Wilson index to predict mortality without transplantation∗DISCUSSION In our patient, the presence of low ceruloplasmin, increased urinary copper, and Kayser–Fleischer ring positivity oriented the diagnosis toward WD, and was confirmed with liver copper content quantification and genetic study. In addition high titer of both ANA and SMA antibodies along with hypergammaglobulinemia, and low C3 and C4 levels pointed to an autoimmune aspect to her liver disease. This was not fully supported by histology because interface hepatitis was absent, although diffuse lymphocyte infiltrates with formation of nodular lymphocyte aggregates with T-cell phenotype, and plasma cell infiltrates were present. This report shows that ALF as a result of WD can have strong autoimmune features. Patients presenting with ALF and positive autoantibodies require thorough workup to exclude WD. ANA and SMA are suggestive but not specific for the diagnosis of AIH because in children they can be detected also in other diseases including viral infections, celiac disease, and WD (5). In particular, patients with severe liver injury of any etiology can show a decrease of ceruloplasmin level associated with normal or elevated serum copper levels and elevated urine copper as was observed in our case (2). The presence of strong and suggestive autoimmune features in these cases could lead to omission of meticulous screening for WD. It is well known that the clinical and histopathology features of WD overlap with those of AIH, and therefore, in patients with AIH, WD should always be ruled out by liver copper measurement and genetic testing. WD with superimposed features of AIH has been reported in 4 previously described cases with ALF (6–8). The presence of conclusive data in support of WD and of features suggestive of AIH prompted us to initiate a therapeutic trial with both corticosteroid and copper chelation therapy; however, the patient failed to respond. This may be because of the fact that AIH was not the cause of liver failure, although the disease was probably too advanced to respond to any treatment. This report suggests that diagnostic investigations of ALF must include all main causes in older children, such as viruses, toxins, metabolic diseases, and AIH even when an initial cause has been determined, as this may have important implications in treatment and outcome. Patients presenting with ALF and positive autoantibodies require a thorough workup to exclude WD.