A 21-year-old woman presented with a 5-day history of loss of appetite, malaise, nausea, and vomiting. On examination, she appeared acutely ill, with jaundice and right-sided hypochondrial tenderness. The stigmata of chronic liver disease were absent, and the results of her neurologic examination were typical. She had previously been well except for bouts of somnolence with occasional emesis during the previous year. There was no history of recent travel, alcohol or illicit drug use, or liver disease in her family. Viral hepatitis was considered the most likely diagnosis, but the patient's clinical deterioration necessitated hospital admission a day later. The results of serology tests for acute infection with hepatitis A virus, hepatitis B virus, or Epstein–Barr virus were negative, and the complete blood count revealed a macrocytic anemia. Liver function tests revealed the following: total bilirubin, 136 μmol/L [upper reference limit (URL),2 26 μmol/L]; direct bilirubin, 62 μmol/L (URL, 7 μmol/L); albumin, 31 g/L [reference interval (RI), 35–50 g/L]; γ-glutamyltransferase, 165 U/L (URL, 44 U/L); aspartate aminotransferase (AST), 166 U/L (URL, 35 U/L); alanine aminotransferase (ALT), 24 U/L (URL, 35 U/L). Serum alkaline phosphatase (ALP) was noteworthy for being undetectable (<5 U/L; RI, 51–117 U/L) on 3 consecutive days. Methodologic interference in the ALP assay by anticoagulants, such as EDTA and fluoride, was excluded on the basis of sodium, potassium, and calcium concentrations that were within reference intervals. Evidence of hemolysis was provided by the results for serum lactate dehydrogenase (786 U/L; RI, 120–230 U/L) and haptoglobin (<60 mg/L; RI, 300–2000 mg/L). The hemoglobin concentration decreased from 83 g/L to 35 g/L (RI, 124–167 g/L), requiring transfusion of 9 units of packed red blood cells. The patient also received 18 units of fresh frozen plasma during 4 plasma-exchange procedures. Although citrate chelation after transfusion of blood products may cause low ALP activity, undetectable ALP activity had been documented before the transfusions. The results for the direct Coombs test were negative, and the international normalized ratio was 2.0 (RI, 0.9–1.2). An initial serum urea concentration of 5.3 mmol/L (RI, 2.1–7.1 mmol/L) and a creatinine value of 78 μmol/L (RI, 53–97 μmol/L) indicated that the patient's renal function was well preserved (Table 1).
QUESTIONS TO CONSIDER
What are the causes and characteristics of fulminant hepatic failure?
Which causes of fulminant hepatic failure are associated with hemolysis?
What are causes of low ALP activity, and how are they related to fulminant hepatic failure?
Fulminant hepatic failure (FHF) is a sudden deterioration of liver function in an otherwise healthy individual that is associated with jaundice, coagulopathy, any degree of encephalopathy, and multiple organ dysfunction (including renal insufficiency in 50% of cases) (1). Causes may be classified as: (a) viral (hepatitis A, B, and/or D; Epstein–Barr virus; and others); (b) drug- and toxin-induced (notably acetaminophen); (c) vascular (“shock liver,” Budd–Chiari syndrome, and veno-occlusive disease); (d) metabolic [acute fatty liver of pregnancy, HELLP (hemolysis, increased liver enzymes, low platelets) syndrome, Reye syndrome, Wilson disease (WD), and other inherited metabolic disorders]; and (e) diverse [sepsis, autoimmune hepatitis (AIH), and hepatic infiltrations] (1). Ingestion of hepatotoxins was not apparent, and the results of an autoantibody screen and serologic tests for common hepatitis viruses were negative. A computed tomography evaluation demonstrated hepatomegaly but no infiltrative or vascular lesions.
Given the prominence of hemolysis in this case, a consideration of FHF etiologies associated with hemolysis is indicated. Hemolysis in FHF may be immune mediated and/or non–immune mediated. Immune-mediated mechanisms are implicated in sepsis, fulminant viral hepatitis, and autoimmune hemolysis associated with AIH (2). Non–immune mediated mechanisms may be categorized as (a) microangiopathic (HELLP syndrome, disseminated intravascular coagulation in sepsis, disseminated malignancy), (b) oxidative (fulminant WD, viral hepatitis in glucose-6-phosphate dehydrogenase deficiency), (c) extravascular destruction (spur cell anemia, hypersplenism), and (d) hemolysin production (Clostridium perfringens sepsis) (2–4). Many of these conditions could be excluded in our patient on the basis of her clinical features and platelet and white cell counts that were within reference intervals.
Methodologic interference in the ALP assay by anticoagulants has been discussed. Other causes of low ALP (EC 126.96.36.199) activity are hereditary hypophosphatasemia, pernicious anemia, hypothyroidism, hypomagnesemia, and zinc deficiency (5). Pernicious anemia and hypothyroidism can be associated with FHF due to AIH in the context of polyglandular autoimmune syndrome, but FHF is unlikely in the other causes. Low serum ALP activity is frequently observed in FHF associated with WD (WD-FHF), which has been called pathognomonic of fulminant WD because it is uncommon in chronic presentations of WD (5, 6). Other findings in FHF that are suggestive of WD are higher bilirubin and lower hemoglobin concentrations, and lower aminotransferase activities (4, 7). Incorporating these features into indices, such as an ALP/total bilirubin ratio <4 and an AST/ALT ratio >2.2, has yielded sensitivities and specificities of 86%–100% for WD-FHF (7). Applying these ratios to our patient's initial results would have led to a WD diagnosis (ALP/total bilirubin ratio indeterminate and an AST/ALT ratio of 6.9). AST/ALT ratios in WD-FHF are high, owing to AST release from necrotic hepatocytes (mitochondrial isoenzyme) and lysing erythrocytes.
The most likely diagnosis in this case was WD-FHF, which was subsequently confirmed by finding a 50-fold increase in urinary copper excretion (29.8 μmol/day; URL, 0.6 μmol/day). The concentration of total copper in the serum was also increased (43.6 μmol/L; RI, 12.6–24.3 μmol/L), but because ceruloplasmin was not measured, the concentration of free copper could not be estimated. A histologic analysis of a liver biopsy sample demonstrated cirrhosis and copper deposition in hepatocytes. WD is an autosomal recessive disorder caused by homozygous or compound heterozygous mutations in the ATP7B gene (ATPase, Cu++ transporting, beta polypeptide) on chromosome 13q14.3. The gene encodes a copper-transporting adenosine triphosphatase required for biliary copper excretion and incorporation of copper into ceruloplasmin. Copper accumulation in tissues manifests as hepatic, neurologic, hemolytic, or psychiatric disease, generally between the first and fourth decades (8). The diagnosis is based on finding decreased serum ceruloplasmin (<200 mg/L), urinary copper excretion >1.57 μmol/day (>100 μg/day), and a hepatic copper content >3.9 μmol/g dry weight (dw) or >250 μg/g dw (URL, 0.8 μmol/g dw or 50 μg/g dw) (8). Heterozygotes are asymptomatic but may exhibit hypoceruloplasminemia, increased urinary copper excretion after penicillamine treatment, and a hepatic copper content of 1.6–3.9 μmol/g dw (100–250 μg/g dw) (9). The heterozygote status should be confirmed to prevent unwarranted lifelong chelation therapy (9).
Hepatic involvement manifests as self-limited acute hepatitis, fatty liver, “autoimmune-like” chronic hepatitis, and cirrhosis (9). Rarely (as seen in this case), the initial presentation is with FHF and Coombs-negative hemolysis (10). Without prompt intervention to remove copper from the circulation, the outcome is fatal unless orthotopic liver transplantation (OLT) is performed (10). Given that chelating agents require weeks to appreciably reduce the plasma copper concentration, modalities such as plasmapheresis have been added (10). By attenuating hemolysis and stabilizing the clinical condition with a combination of chelation and plasma exchange, OLT may be averted or postponed (10). Early and definitive diagnosis of WD-FHF is therefore essential, but it is not accomplished without difficulty. Kayser–Fleischer rings, although present in >90% of individuals with neuropsychiatric WD, occur in only 50% of those with liver disease (7, 9). There were no obvious Kayser–Fleischer rings in our patient, but slit-lamp examination was not performed.
The biochemical assessment of copper status in FHF is unreliable, and the tests are subject to long turnaround times. Ceruloplasmin assays based on oxidase activity are preferred for WD diagnosis, because immunonephelometric methods measure both holo- and apoceruloplasmin (8, 9). Nonpathologic concentrations may be seen in WD-FHF, because ceruloplasmin is an acute-phase reactant and hypoceruloplasminemia is not uncommon in non-WD FHF, owing to impaired hepatic synthesis (4, 7). Total copper concentrations in serum, which are reduced in chronic WD, are frequently increased in WD-FHF because of copper release from necrotic hepatocytes (4, 7). Hypercupremia occurs in non-WD FHF for the same reason, but concentrations >31.4 μmol/L (>200 μg/dL) are highly suggestive of WD (7). Urinary copper excretion is often dramatically increased in WD-FHF and is less so in non-WD FHF [mean (SD), 56.6 (9.8) μmol/day or 3602 (622) μg/day, vs 9.2 (3.6) μmol/day or 584 (231) μg/day, in one study], but assessment may be hampered by oliguria (4). Estimating the tissue copper concentration is of limited value because hepatic copper is increased in several chronic liver diseases. Copper concentrations in WD are not consistently high and can vary between biopsies because of the patchy distribution of copper in cirrhosis (8, 9). In patients with FHF, the presence of coagulopathy may preclude liver biopsy.
Liver disease in WD is hypothesized to be due to oxidative damage caused by reactive oxygen species produced by the copper excess. This hypothesis is supported by studies of hepatic tissues from affected patients, which show decreased ratios of reduced to oxidized glutathione, increased lipid peroxidation products, and lower activities of antioxidant enzymes (11). FHF in WD is believed to be initiated by extensive apoptosis of hepatocytes triggered by extremely high loads of oxidant stress (11). The sudden massive release of hepatic copper precipitates hemolysis via direct damage to erythrocyte membranes, by inhibiting erythrocyte enzymes, and oxidative stress (4, 5). Low ALP activity in WD-FHF has similarly been ascribed to oxidative damage by hydroxyl free radicals generated during Cu2+-catalyzed ascorbate oxidation (6). Another proposed mechanism is that transient marked hypercupremia leads to incorporation of Cu2+ instead of Zn2+ at the ALP active site, yielding an enzyme with reduced activity; however, attempts to simulate this phenomenon in vitro have been unsuccessful (5, 6).
The patient recovered sufficiently after treatment with penicillamine and plasmapheresis to permit discharge after 12 days. She continues to do well on chelation therapy, and listing for OLT is not currently being considered. Genetic testing will be offered to the proband and her first-degree relatives.
POINTS TO REMEMBER
The combination of Coombs-negative hemolysis and low serum ALP activity in a patient with FHF should arouse suspicion of WD. Most of the other causes of low ALP activity are easily excluded and unlikely to be associated with FHF.
Pronounced hyperbilirubinemia, low ALP activity, and relatively mild increases in aminotransferases are suggestive of WD as the cause of FHF. ALP/total bilirubin and AST/ALT ratios have high sensitivity and specificity for diagnosing WD-FHF, particularly in combination.
Of the biochemical indicators of copper status, serum total copper is the best predictor of WD in the setting of FHF. Dramatic increases in urinary copper excretion are likewise suggestive, but assessment may be hampered by renal insufficiency. Ceruloplasmin measurement and tissue copper estimation remain useful in the workup of chronic WD.
Rapid reduction of plasma copper levels by plasmapheresis may be lifesaving in WD-FHF, allowing OLT to be averted or postponed.
Confirmation of the heterozygote status is essential in asymptomatic individuals to prevent inappropriate lifelong chelation therapy.
I am grateful to Dr. Lani Frantzen for introducing me to the case and thank Lancet Laboratories and Vermaak and Partners' Pathologists for making laboratory data available to me. I thank Dr. Robert Bond, Dr. Theunis Botha, and Prof. Ernie Song for sharing clinical case details.
↵2 Nonstandard abbreviations:
- upper reference limit;
- reference interval;
- aspartate aminotransferase;
- alanine aminotransferase;
- alkaline phosphatase;
- fulminant hepatic failure;
- hemolysis, elevated liver enzymes and low platelets;
- Wilson disease;
- autoimmune hepatitis;
- FHF associated with WD;
- dry weight;
- orthotopic liver transplantation.
Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.
Authors' Disclosures or Potential Conflicts of Interest: No authors declared any potential conflicts of interest.
Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or preparation or approval of manuscript.
- Received for publication June 24, 2010.
- Accepted for publication September 1, 2010.
- © 2010 The American Association for Clinical Chemistry