EASL Clinical Practice Guidelines

Diagnostic methods

Typically, the combination of Kayser–Fleischer rings and a low serum ceruloplasmin (<0.1 g/L) level is sufficient to establish a diagnosis. When Kayser–Fleischer rings are not present (as is common in the hepatic manifestation of Wilson's disease), ceruloplasmin levels are not always reliable because they may be low for reasons other than Wilson's disease (e.g. autoimmune hepatitis, severe hepatic insufficiency in advanced liver disease, celiac disease, familial aceruloplasminemia) [43] or in heterozygous carriers of ATP7B mutations who do not show copper overload disease. On the other hand, inflammation in the liver or elsewhere may cause the ceruloplasmin concentration to rise to normal levels, reflecting its identity as an acute phase protein. This is also true for treatment with estrogens. Thus, for many patients, a combination of tests reflecting disturbed copper metabolism may be needed. Not a single test is per se specific and, thus, a range of tests has to be applied (Table 4). A diagnostic score based on all available tests was proposed by the Working Party at the 8th International Meeting on Wilson's disease, Leipzig 2001 [44] (Table 5). The Wilson's disease scoring system provides a good diagnostic accuracy [45]. The diagnostic algorithm based on this score is shown in Fig. 1.

Table 4
Routine tests for diagnosis of Wilson's disease.

Table 5
Scoring system developed at the 8th International Meeting on Wilson's disease, Leipzig 2001 [44].

∗If no quantitative liver copper available, ∗∗or typical abnormalities at brain magnetic resonance imaging. KF, Kayser–Fleischer; ULN, upper limit of normal.

Fig. 1 Diagnostic algorithms for Wilson's disease based on the Leipzig Score [44]. ∗In children the cut off can be lowered to 0.64 μmol/d.

View Large Image | View Hi-Res Image | Download PowerPoint Slide

Serum ceruloplasmin

Ceruloplasmin is the major carrier of copper in the blood. It contains six copper atoms per molecule (holoceruloplasmin) but may be present just as the protein without the copper (apoceruloplasmin). Ceruloplasmin is an acute phase reactant possessing a ferroxidase activity [46]. Levels of serum ceruloplasmin may be measured enzymatically by its copper-dependent oxidase activity towards specific substrates, or by antibody-dependent assays such as radioimmunoassay, radial immunodiffusion, or nephelometry. Immunologic assays may overestimate ceruloplasmin concentrations since they do not discriminate between apoceruloplasmin and holoceruloplasmin. The normal concentration of ceruloplasmin measured by the enzymatic assay varies among laboratories (with a lower limit between 0.15 and 0.2 g/L). In Wilson's disease, it is usually lower than 0.1 g/L. Serum ceruloplasmin concentrations are elevated by acute inflammation, in states associated with hyperestrogenemia such as pregnancy and estrogen supplementation. Serum ceruloplasmin is typically decreased in patients with neurologic Wilson's disease, but may be in the low normal range in about half of patients with active Wilson's liver disease. On the other hand, serum ceruloplasmin may be low in other conditions with marked renal or enteric protein loss, malabsorption syndromes or with severe end-stage liver disease of any etiology. Approximately 20% of heterozygotes have decreased levels of serum ceruloplasmin [[1], [47]]. Patients with aceruloplasminemia lack the protein entirely due to mutations in the ceruloplasmin gene on chromosome 3. These patients may exhibit hemosiderosis but do not have copper accumulation [48]. Thus, serum ceruloplasmin alone is not sufficient to diagnose or to exclude Wilson's disease. A prospective study on serum ceruloplasmin, as a screening test for Wilson's disease in patients referred with liver disease, showed that subnormal ceruloplasmin had a positive predictive value of only 6%. In children with Wilson's disease, 15–36% had ceruloplasmin in the normal range [[14], [49]]. In one series, 12 out of 55 Wilson's disease patients had normal ceruloplasmin and no Kayser–Fleischer rings [12]. The predictive value of ceruloplasmin for diagnosis of Wilson's disease in acute liver failure is poor [50]. In one recently published study, measurement of serum ceruloplasmin oxidase activity was superior to immunologic assays for diagnosing Wilson's disease, but these assays are generally not available in routine labs [51].

Serum copper

Although a disease of copper overload, the total serum copper (which includes copper incorporated in ceruloplasmin) in Wilson's disease is usually decreased in proportion to the decreased ceruloplasmin in the circulation. In patients with severe liver injury, serum copper may be within the normal range, independent of whether serum ceruloplasmin levels are elevated or low. In the setting of acute liver failure due to Wilson's disease, levels of serum copper may even be markedly elevated due to the sudden release of the metal from liver tissue stores. Normal or elevated serum copper levels, in the face of decreased levels of ceruloplasmin, indicate an increase in the concentration of copper which is not bound to ceruloplasmin in the blood (non-ceruloplasmin-bound copper). Non-ceruloplasmin-bound copper (or “free copper”) can be calculated by subtracting ceruloplasmin-bound copper (3.15 × ceruloplasmin in mg/L equals the amount of ceruloplasmin-bound copper in μg/L) from the total serum copper concentration (in μg/L; serum copper in μmol/L × 63.5 equals serum copper in μg/L) [52]. The serum non-ceruloplasmin-bound copper concentration has been proposed as a diagnostic test for Wilson's disease [53]. In most untreated patients, it is elevated above 200 μg/L. The serum non-ceruloplasmin copper concentration may be elevated in acute liver failure of any etiology, in chronic cholestasis [54], and in cases of copper intoxication. The major problem with non-ceruloplasmin-bound copper as a diagnostic test for Wilson's disease is that it is dependent on the adequacy of the methods for measuring both serum copper and ceruloplasmin. It is of more value in monitoring pharmacotherapy than in the diagnosis of Wilson's disease.

Urinary copper excretion

The amount of copper excreted in the urine in a 24-hour period may be helpful for diagnosing Wilson's disease and for monitoring treatment. In untreated patients, the 24-hour urinary excretion of copper reflects the amount of non-ceruloplasmin-bound copper in the circulation. The exact urine volume and the total creatinine excretion per 24 h are important for accurate determination of urinary copper excretion. In case of renal failure, the test is not applicable. In untreated symptomatic patients, “baseline” copper excretion greater than 1.6 μmol/24 h (100 μg/24 h) is taken as diagnostic of Wilson's disease [5]. However, basal 24-hour urinary copper excretion may be less than 1.6 μmol/24 h at presentation in 16–23% of patients, especially in children and asymptomatic siblings [[12], [14], [55]]. Since urinary copper excretion is negligible in healthy individuals [56], a urinary copper excretion above 0.64 μmol/24 h can be suggestive of Wilson's disease in asymptomatic children. The problems of measuring 24-hour copper excretion include incomplete urine collection, and, on the other hand, copper contamination of the collection device (this being less problematic with the advent of disposable containers). Interpreting 24-hour urinary copper excretion can be difficult due to the overlap with findings in other types of liver disease (e.g. autoimmune hepatitis, chronic active liver disease or cholestasis and in particular during acute hepatic failure of any origin). Heterozygotes may also have higher copper excretion than controls, rarely exceeding the normal range levels [57].

Urinary copper excretion with D-penicillamine administration was thought to be a useful diagnostic test. This test has only been standardized in a pediatric population in which 500 mg of D-penicillamine was administered orally at the beginning and again 12 h later during the 24-hour urine collection, irrespective of body weight [58]. Compared with a spectrum of other liver diseases, including autoimmune hepatitis, primary sclerosing cholangitis, and acute liver failure, a clear differentiation was found when more than 25 μmol/24 h was excreted. A reassessment of this test in paediatric patients reconfirmed the value in the diagnosis of Wilson's disease with active liver disease, but was unreliable to exclude the diagnosis in asymptomatic siblings [59]. In comparison to children with other liver diseases, the D-penicillamine test had only a sensitivity of 12.5%. However, data by Dhawan et al. and by Nicastro et al. now suggest that using a lower threshold for urinary copper excretion (without D-penicillamine stimulation) of only 0.64 μmol/24 h increases sensitivity of the test and eliminates the need for the stimulation testing with D-penicillamine [[41], [45]].

The penicillamine challenge test has been used in adults, but many of the reported results of this test utilized different dosages and timing for administration of the D-penicillamine [[12], [53], [56]]. Thus, this test is not recommended for diagnosis of Wilson's disease in adults.

Hepatic parenchymal copper concentration

Hepatic copper accumulation is the hallmark of Wilson's disease. However, specific stains like rhodamine or orcein reveal focal copper stores in less than 10% of patients because they detect only lysosomal copper depositions. Thus, hepatic copper overload cannot be excluded by histochemical evaluation of a liver biopsy alone. Therefore, the measurement of hepatic parenchymal copper concentration is the method of choice for the diagnosis of Wilson's disease. Biopsies for quantitative copper determination should be placed dry in a copper-free container. Shipment for quantitative copper determination does not require special precautions like freezing. In general, the accuracy of measurement is improved with adequate specimen size: at least 1 cm of biopsy core length should be submitted for analysis [62]. Paraffin-embedded specimens may also be analyzed for copper content, but may be less reliable if the specimen is small. Hepatic copper content >4 μmol/g dry weight is considered as the best biochemical evidence for Wilson's disease. Lowering the threshold from 4 μmol/g dry weight to 1.2 μmol/g dry weight improved sensitivity from 83.3% to 96.5%, while specificity remained acceptable (95.4% vs. 98.6%) [28]. The major problem with hepatic parenchymal copper concentration is the inhomogeneous distribution of copper within the liver in later stages of Wilson's disease. Thus, the concentration can be underestimated due to sampling error. In about 18% of adult patients, hepatic copper concentrations are only between 0.8 and 4 μmol/g dry weight with a few even in the normal range [28]. In a pediatric study, sampling error was sufficiently common to render this test unreliable in patients with cirrhosis [60]. On the other hand, in long-standing cholestatic disorders, hepatic copper content may also be increased. Markedly elevated levels of hepatic copper may also be found in idiopathic copper toxicosis syndromes such as Indian childhood cirrhosis [61].

Liver histology

For diagnostic purposes, a liver biopsy is only required if the clinical signs and noninvasive tests do not allow a final diagnosis or if there is suspicion of other or additional liver pathologies [62].

The earliest histologic abnormalities in the liver include mild steatosis (both microvesicular and macrovesicular), glycogenated nuclei in hepatocytes, and focal hepatocellular necrosis [[62], [63]]. Frequently, these changes are misdiagnosed as nonalcoholic fatty liver disease (NAFLD) or nonalcoholic steatohepatitis (NASH). The liver biopsy may show classic histologic features of autoimmune hepatitis (the so-called “chronic active hepatitis” picture). With progressive parenchymal damage, fibrosis and subsequently cirrhosis develop. About half of the patients have cirrhosis at the time of diagnosis [28]. There are a few older patients with Wilson's disease who do not have cirrhosis or even signs of liver disease [[5], [12]]. In the setting of acute liver failure due to Wilson's disease, there is a marked hepatocellular degeneration and parenchymal collapse, typically on the background of cirrhosis. Apoptosis of hepatocytes is a prominent feature during the acute injury [64].

Detection of copper in hepatocytes by routine histochemical evaluation is highly variable. Especially in early stages of the disease, copper is mainly present in the cytoplasm bound to metallothionein and is not histochemically detectable [65]. The amount of copper varies from nodule to nodule in the cirrhotic liver and may vary from cell to cell in pre-cirrhotic stages. The absence of histochemically identifiable copper does not exclude Wilson's disease. Lysosomal copper complexes can be stained by various methods, including the rhodanine or orcein stain.

Ultrastructural analysis of liver specimens at the time steatosis is present reveals specific mitochondrial abnormalities [66]. Typical findings include variability in size and shape, increased density of the matrix material, and numerous inclusions including lipid and fine granular material that may be copper. The most striking alteration is increased intracristal space with dilatation of the tips of the cristae, creating a cystic appearance [66]. In the absence of cholestasis, these changes are considered to be essentially pathognomonic of Wilson's disease. At later stages of the disease, dense deposits within lysosomes are present. Ultrastructural analysis may be a useful adjunct for diagnosis.

Neurologic findings and radiologic imaging of the brain

Neurologic evaluation should be performed also on patients with presymptomatic and hepatic Wilson's disease. Consultation with a neurologist should be sought for evaluation of patients with evident neurologic symptoms before treatment or soon after treatment is initiated.

Neurologic disease may manifest as motor abnormalities with Parkinsonian characteristics of dystonia, hypertonia and rigidity, choreic or pseudosclerotic, with tremors and dysarthria. Due to the great variability of neurological signs, differences in their severity and concomitant presence of different signs in one patient, clinical description is very difficult. There is not yet a commonly accepted scale which describes neurological signs and their severity. One recent proposal is the Unified Wilson's disease Rating Scale (UWDRS) [[67], [68]].

Magnetic resonance imaging (MRI) or computerized tomography of the brain may detect structural abnormalities in the basal ganglia [69]. The most frequent findings are an increased density on computerized tomography or hyperintensity on T2 MRI in the region of the basal ganglia. MRI may be more sensitive in detecting these lesions. Abnormal findings are not limited to this region, and other abnormalities have been described. A characteristic finding in Wilson's disease is the “face of the giant panda” sign [[70], [71]], but is found only in a minority of patients. Besides this sign, hyperintensities in tectal-plate and central pons (CPM-like), and simultaneous involvement of basal ganglia, thalamus, and brainstem are virtually pathognomonic of Wilson's disease [72]. Significant abnormalities on brain imaging may even be present in some individuals prior to the onset of symptoms [69].

Other neuroimaging techniques as magnetic resonance spectroscopy [70] and single-photon emission computed tomography (SPECT) might be useful in detecting early brain damage in Wilson's disease, not only in the perspective of assessing and treating motor impairment but also in better evaluating the less investigated disorders in the cognitive domain [73]. Transcranial brain parenchyma sonography (TCS) detects lenticular nucleus hyperechogenicity even when in MRI no abnormalities are observed [74], but it must be confirmed in further studies [75].

Auditory-evoked brainstem potentials are helpful to document the degree of functional impairment and the improvement by treatment [[76], [77]].

Genetic testing

Direct molecular-genetic diagnosis is difficult because of more than 500 possible mutations; except for a few more frequent mutations, each of which is rare [78]. Furthermore, most patients are compound heterozygotes (i.e. carry two different mutations). Comprehensive molecular-genetic screening takes several months, which makes this an impractical method. Nevertheless, it is reasonable to perform molecular analysis of the ATP7B gene in any patient who has a provisional diagnosis of Wilson's disease, both for confirmation purposes and to facilitate the subsequent screening of family members.

By contrast, allele-specific probes allow direct identification of a mutation and this can be rapid and clinically very helpful. However, this can only be accomplished if a mutation occurs with a reasonable frequency in the population (e.g. H1069Q in Central Europe [79], –441/–427 del. in Sardinia [[80], [81]], R778L in the Far East [[82], [83], [84]]). In those cases, identification of a mutation can support the diagnosis, while identification of two mutations will confirm the diagnosis. With the advancement of DNA-based diagnostics, such as the development of a single chip that is able to identify the most common mutations, these recommendations may change.

Acute liver failure due to Wilson’s disease

The most challenging aspect is the diagnosis of acute liver failure due to Wilson's disease, since mortality without emergency liver transplantation is very high. Readily available laboratory tests, including alkaline phosphatase (AP), bilirubin, and serum aminotransferases, provide the most rapid and accurate method for diagnosis of acute liver failure due to Wilson's disease [85]. Combination of an AP elevation/total bilirubin elevation ratio <4 and an AST:ALT ratio >2.2 yielded a diagnostic sensitivity and specificity of 100% [86]. However, these findings were challenged by other authors. Therefore, these parameters should be considered in case acute Wilson's disease is suspected, but should be used in combination with other signs and symptoms suggesting Wilson's disease. The combination of clinical symptoms and the conventional Wilson's disease diagnostic parameters (ceruloplasmin, serum or urinary copper) are less sensitive and specific but important for the diagnosis [86]. The diagnosis has to be ascertained by liver biopsy if possible or at least after transplantation (hepatic copper content, mutation analysis) to enable screening of asymptomatic siblings.