| | Liver imaging findings of Wilson's diseaseReceived 8 October 2006; accepted 2 November 2006. Abstract Wilson's disease is a rare, autosomal-recessive inherited disorder of copper metabolism resulting in accumulation of copper in liver. The form of liver disease varies, depending on the severity of the disease at the time of diagnosis and pathological findings include fatty changes, acute hepatitis, chronic active hepatitis, cirrhosis and occasionally fulminant hepatic necrosis. Liver imaging findings reflect a wide range of physiopathological processes of the disease and also demonstrate the associated findings of cirrhosis in cases with advanced disease. 1. Introduction  Wilson's disease or hepatolenticular degeneration is a rare, autosomal-recessive inherited disorder of copper metabolism characterized by copper accumulation in hepatocytes and in extrahepatic organs such as the brain and the cornea. An average adult diet contains approximately 5 mg of copper, 40% of which is absorbed in the upper gastrointestinal tract. Almost all of the absorbed copper is excreted in the bile as a complex that has no enterohepatic circulation. In Wilson's disease, copper accumulates in hepatocytes due to inability to secrete the copper–ceruloplasmin complex into plasma and free copper into bile. The form of liver disease varies, depending on the severity of disease at the time of diagnosis. Pathologic changes, in the basic sequence of their evolution, include fatty change, acute hepatitis, chronic active hepatitis, cirrhosis and occasionally fulminant hepatic necrosis [1], [2], [3], [4]. Ultrasound (US), computed tomography (CT), and magnetic resonance (MR) imaging findings of liver usually reflect these wide range of the physiopathological processes of the disease and also demonstrate the associated findings of cirrhosis in cases with advanced disease [5], [6]. 2. Genetics and pathophysiology In 1953 an evidence of autosomal recessive mode of inheritance was presented by Bearn who confirmed this inheritance by an extended genetic analysis of 30 families [7], [8]. The gene for Wilson's disease is distributed worldwide and the prevalence of the disease is approximately 1 in 30,000 live births, with an incidence rate ranging from 15 to 30 per million [9]. In 1985 Frydman et al. [10] established linkage of Wilson's disease gene to the esterase D locus on chromosome 13 and the gene responsible for Wilson's disease (ATP7b) was cloned independently in three different laboratories in 1993 [3]. Thus far, more than 200 distinct mutations have been described [3]. ATP7b gene encodes a metal-transporting P-type ATP-ase, which is expressed mainly in hepatoyctes and resides in hepatocytes in the trans-Golgi network, transporting copper into the secretory pathway for incorporation into apoceruloplasmin and biliary copper excretion. Absence or reduced function of ATP7b protein leads to decreased hepatocellular excretion of copper into bile resulting in hepatic copper accumulation and injury, as well as copper deposits and damage of other organs, notably the brain cornea and kidneys [4], [11], [12]. There is considerable phenotypic variation in Wilson's disease. Phenotypic variability among Wilson's disease patients could potentially be in some measure due to the type of mutation and resultant effect on the ATP7b protein. However, it is unknown why some patients with the same genotype present with hepatic disease during the first decade of life and others with neurological degeneration in adolescence or adult life. Some environmental factors such as amount of copper in the diet, epigenetic and other genetic factors will probably play a role in this issue [13], [14]. The pathogenesis of Wilsonian liver disease is a direct consequence of copper accumulation in hepatocytes. It is postulated that, the harmful effects of excess copper are mediated by generation of free radicals. Initially, it causes mitochondrial damage with alteration of lipid oxidation, resulting in marked hepatic steatosis. Severe mitochodrial dysfunction was shown in the liver with diminished enzyme activity [4], [9]. Accumulation of prooxidant copper within the hepatic mitochondria leads to premature oxidative aging of mitochondrial DNA and release of copper from necrotic hepatocytes result in an oxidative stress leading to further damage of hepatocytes, inflammation and fibrogenesis [4]. 3. Pathology Liver disease may mimic all forms of common liver conditions, ranging from asymptomatic transaminasemia, acute or chronic hepatitis, fulminant hepatic failure, and cirrhosis [4]. The earliest changes detectable on light microscopy include glycogen deposition in the nuclei of periportal hepatocytes and moderate fatty infiltration [9]. The lipid droplets which are composed of triglycerides, progressively increase in number and size, in some cases resembling the steatosis induced by ethanol [9], [15]. The rate of progression of the liver histology from fatty infiltration to cirrhosis is variable and some Wilson's disease patients develop a histologic picture that is indistinguishable from chronic active hepatitis. If remains untreated, this may evolve into macronodular cirrhosis, or may progress rapidly into fulminant hepatitis [9], [16]. Changes in the precirrhotic stages of Wilson's disease may be minimal, but in many cases there are hepatocellular degenerative changes, fatty metamorphosis, periportal atypical lipofuscin, and vacuolated nuclei; the latter are commonly clustered about portal areas. Each of these findings can be regarded as nonspecific, but their combination; particularly in a child or young adult should suggest the possibility of Wilson's disease. Histochemical demonstration of copper can strengthen the evidence for diagnosis, but negative stain does not rule out either Wilson's disease or an increased hepatic copper concentration. Stromeyer and Ihsak defined copper deposition patterns in Wilson's disease. In early Wilson's disease, the presence of copper is classified as Pattern A in which the copper is localized in periportal areas, Pattern B in which the copper is found near residual portal tracts and fibrous septa and also there are macronodular cirrhosis with pseudolobules. Pattern C in which mixed micronodular–macronodular cirrhosis with absence of copper in regenerative nodules and finally Pattern D which is very similar to Pattern C except all nodules contain copper in Pattern D [17]. 4. Imaging findings 4.1. Structural changes One of the most important objectives of liver imaging is to assess the existence of cirrhosis. Parenchymal necrosis, regeneration and scarring which develops secondary to cirrhosis in the liver parenchyma, can cause morphologic changes. Harbin et al. reported that the caudate to right lobe ratio greater than or equal to 0.65 was 84% sensitive and 100% specific in the diagnosis of cirrhosis [18]. However, the normal caudate to right lobe ratio in patients with cirrhosis secondary to Wilson's disease is a usual finding in contrary to cirrhosis due to other reasons (Fig. 1). Therefore, Wilson's disease should be strongly considered in the etiology of the cirrhotic patients without caudate lobe hypertrophy in childhood and young adult group. In addition to this, if there is caudate lobe hypertrophy in patients with hepatic Wilson's disease, then a secondary etiologic cause such as hepatitis B infection should also be investigated [5], [19]. Contour irregularity, increased periportal thickness were also found in US, CT and MRI examinations (Fig. 2) [5], [19], [20], [21]. These findings were non-specific for Wilson's disease and they demonstrated the parenchymal fibrosis due to cirrhosis. 4.2. Parenchymal changes and nodules Increased hepatic echogenecity in Wilson's disease has been reported in the literature and this may be due to fatty change, or due to fibrosis in the long-term period [19], [20], [22], [23]. Corda et al. have determined increased liver echogenicity in three cases of eight patients with Wilson's disease [23]. Akpinar has determined similar findings in 10 of 28 cases [19]. Increased liver echogenecity could only be a positive finding in Wilson disease with fatty changes (Fig. 3). Therefore, hepatic presentation of Wilson's disease should be considered in patients with increased hepatic echogenecity during childhood with appropriate clinical scenario. There have been difficulties in diagnosing Wilson's disease when it is presented as an isolated liver disease with increased echogenecity in childhood. In these symptom-free children, Kayser–Fleischer corneal rings are usually absent, serum ceruloplasmin can be normal in up to a third, and the diagnostic sensitivity of 24-h urinary copper storage excretion, even enhanced by d-penicillamine challenge, is low [24]. The presence of microvesicular steatosis associated with various degree of portal fibrosis, in non-overweight children with asymptomatic elevation of aminotransferase activity, should suggest Wilson disease [24], [25], [26]. And only mutation analysis allowed a definitive diagnosis of Wilson's disease in selected cases. This is particularly important when facing a child with features of non-alcoholic fatty liver disease [24]. Liver parenchymal heterogeneity in US could be observed in most of the patients with Wilson's disease. Hepatic parenchymal changes were observed in 29 out of 33 cases in sonography in the study performed by Cancado et al. [22]. The parenchymal changes in US were classified under three main patterns which were: (1) parenchymal heterogeneity, (2) parenchymal heterogeneity with multiple hypoechoic nodules (Fig. 4), and (3) parenchymal heterogeneity with multiple hyperechoic and hypoechoic nodules (Fig. 5). Vogel et al. have stated that multiple hypoecoic nodules were found in three cases out of five patients with Wilson's disease. Vogel et al. defined these parenchymal changes as increased echogenecity with numerous roundish foci of decreased echogenecity resembling metastatic disease [27]. This group of patients have responded well to decoppering therapy and ultrasound might have helped to define Wilson's patients with good prognosis [27], [28]. Akpinar had determined multiple hypoechoic nodules in 4 of 28 cases. These nodules were not visible on CT (Fig. 4) [19]. Vogl et al. observed multiple low-intensity nodules surrounded by high-intensity septa in 10 out of 16 patients on T2-weighted images. Five patients had also low-intensity nodules on T1 weighted images in their study. Common feature of this patient group was marked inflammatory cell infiltration into fibrous septa, increase of copper concentration in liver parenchyma and distinct pathological changes of laboratory data. In the remaining 6 patients, no pathological changes of liver morphology was demonstrated by MRI corresponding to slight histopathological changes of parenchyma and normal laboratory data. The authors also claimed that bad prognosis may be diagnosed by the presence of low-intensity nodules surrounded by high-intensity septa [29]. Akpinar showed similar nodules in 5 of 12 cases examined by MRI. Multiple low-intensity nodules surrounded with hyperintense septa form a honeycomb pattern (Fig. 6, Fig. 7). Patients with honeycomb pattern on MRI had multiple hyperechoic and hypoechoic nodules in US [19]. Chu et al. showed multiple hypointense nodules ranged from 2mm to 1cm in diameter in the liver on the T2-weighted images and determined elevation in the phosphomonoester (PME) resonance and a reduction in the phosphodiester (PDE) resonance in 31P MR spectra. Follow-up MRI study after a daily regimen of oral penicillamine and vitamin K revealed marked reduction in the number of regeneration nodules and showed normalization of PME and PDE resonance. Simultaneous liver biopsy showed a reduction of necroinflammatory activity. And they suggested that 31P MR spectroscopy had a potential role for assessing disease severity and for measuring the response of the liver to treatment. This noninvasive technique may reduce the necessity for liver biopsy in monitoring liver involvement in pediatric patients and those at risk of complications (e.g., coagulopathy) from biopsy [30]. The metallic deposition may result in diffuse increased attenuation on CT owing to high atomic number of copper. This finding is variable, however, in part because the associated fatty change causes decreased attenuation and prevents any appreciable increase in attenuation [6]. Dixon and Walshe performed upper abdominal CT in 24 patients with Wilson's disease in order to determine hepatic attenuation values. They found hepatic attenuation values (range 44.7–69.3, mean 58.4 HU) within normal limits [31]. Nodules observed on CT may appear as hypodense nodules which become more prominent in the portal venous phase or hyperdense nodules on unenhanced axial images (Fig. 5, Fig. 6) [19], [32]. Hyperdense nodules on unenhanced CT were hyperintense on T1-weighted and hypointense on T2-weighted MR images (Fig. 6). Ko et al. claimed that this might be ascribed to the paramagnetism of copper deposited in liver at a relatively early stage of the disease before severe cirrhosis had evolved [32]. US demonstrates parenchymal involvement better than CT and MRI in the early period of the disease. However, in advanced disease there is an excellent correlation among US, CT and MRI. Zachoval and Glaser presented CT findings of a patient with Wilson's disease as small hyperdense nodules and surface irregularity. Follow-up CT imaging 5 years after d-penicillamine therapy showed disappearance of small hyperdense nodules and surface irregularity [21]. Hepatocellular carcinoma is rarely associated with Wilson's disease. It has been suggested that a protective effect of hepatic copper against oncogenesis may account for the infrequent development of HCC in Wilson's disease [33]. However cases of Wilson's disease complicated by HCC have also been reported [33], [34]. Walshe et al. retrospectively reviewed the case records of 363 patients with Wilson's disease to assess the frequency of abdominal malignant disease in long-term follow-up. No cancers were seen in patients followed for <10 years. For patients in the 10–19 years follow-up group, the frequency was 4.2%; 20–29 years, it was 5.3% and 30–39 years, it was %15. The malignancies were including hepatomas, cholangiocarcinomas, and poorly differentiated adenocarcinomas of undetermined origin. The authors stated that patients with Wilson's disease appear to be vulnerable to the formation of aggressive malignant intra-abdominal tumors during long-term follow-up, irrespective of treatment [34]. Lee et al. stated arterial phase enhancement of nodular lesions in a cirrhotic liver at CT is diagnostic of hepatocellular carcinoma; however, in a later series, of more than 500 patients with cirrhosis, findings revealed that, 2% of arterial-phase-enhancing masses were not hepatocellular carcinoma but rather were benign lesions including transient hepatic attenuation difference, hemangioma, hepatic peliosis, fibrosis, splenic lobule and cryptogenic causes [35], [36]. Akhan et al. reported dysplastic nodules in a cirrhotic liver with Wilson's disease (Fig. 8) [5]. Ultrasound scanning of the abdomen seems to be a useful screening method [34]. Additionally, contrast enhanced dynamic CT and/or MRI can be used for detection and characterization of nodular lesions in cirrhotic patients. 4.3. Perihepatic subcapsular fat layer Akhan et al. firstly defined the perihepatic fat layer in cirrhosis due to Wilson's disease with correlative CT and MRI findings [5]. Perihepatic fat layer was noted as perihepatic hyperechogenic zone in US, perihepatic hypodense (isodense to subcutaneous fat) zone on CT and perihepatic T1W and T2W hyperintense layer on MRI examinations. Akpinar showed perihepatic fat layer in 8 of 28 patients (Fig. 1, Fig. 5, Fig. 6, Fig. 8) [19]. Hepatic subcapsular steatosis in response to intraperitoneal insulin delivery might mimic this finding however clinical history and other associated findings can be helpful in the differential diagnosis [37]. Further studies are needed to understand the importance of this feature. 4.4. Other findings Cholelithiasis is not rare in patients with Wilson's disease and can cause cholesystitis [38]. Cancado et al. reported cholelithiasis in 8 of 33 patients and Akpinar reported in 6 of 28 patients [19], [22]. Therefore, young patients with Wilson's disease should undergo routine investigation for the presence of stones and cholelithiasis should also be considered in the differential diagnosis of abdominal pain in such patients [38]. Akpinar showed splenomegaly in 22 out of 28 cases and collateral veins due to the portosystemic shunts in 11 of 28 (Fig. 9) [19]. Uno et al. defined color Doppler findings as intrahepatic vascular narrowing and nodular coarse parenchymal texture, with multiple very-high-echo spots along the portal vein [39]. We think that the routine imaging is needed to determine the presence of portal hypertension and portosystemic shunts. 5. Conclusion Imaging findings of liver involvement in Wilson's disease show variation considering different stages of the disease. Increased liver echogenicity, parenchymal heterogeneity with multiple nodules, hepatic contour irregularity, normal caudate to right lobe ratio and presence of perihepatic fat layer are considered as imaging findings of Wilson's disease. Although some of these findings are not pathognomonic, the latter two (normal caudate to right lobe ratio and presence of perihepatic fat layer) seem very important to obtain the correct diagnosis in the differential diagnosis of Wilson's disease. References [1]. [1]Lipscomb MF. Wilson's disease. In: Kumar V, Cotran RS, Robbins SL editor. Basic pathology. 5th ed.. Philadelphia: Saunders; 1992;p. 557. [2]. [2]Ferenci P, Caca K, Loudianos G, et al. Diagnosis and phenotypic classification of Wilson disease. Liver Int. 2003;23(3):139–142. MEDLINE [3]. [3]ATP7b gene and Wilson's disease. J Gastroenterol Hepatol. 2004;19(3):343. MEDLINE |
CrossRef
[4]. [4]Ferenci P. Pathophysiology and clinical features of Wilson disease. Metab Brain Dis. 2004;19(3–4):229–239. MEDLINE |
CrossRef
[5]. [5]Akhan O, Akpinar E, Oto A, et al. Unusual imaging findings in Wilson's disease. Eur Radiol. 2002;12(Suppl. 3):S66–S69. [6]. [6]Mergo PJ, Ros PR, Buetow PC, Buck JL. Diffuse disease of the liver: radiologic–pathologic correlation. Radiographics. 1994;14(6):1291–1307. MEDLINE [7]. [7]Bearn AG. Genetic and biochemical aspects of Wilson's disease. Am J Med. 1953;15(4):442–449.
Full-Text PDF (740 KB)
|
CrossRef
[8]. [8]Bearn AG. A genetical analysis of thirty families with Wilson's disease (hepatolenticular degeneration). Ann Hum Genet. 1960;24:33–43. MEDLINE |
CrossRef
[9]. [9]Zucker S, Gollan JL. Wilson's disease and hepatic copper toxicosis. In: Zakim L, Boyer TD editor. Hepatology: a text book of liver disease. 3rd ed.. Philadelphia: Saunders; 1996;p. 1405–1439. [10]. [10]Frydman M, Bonne-Tamir B, Farrer LA, et al. Assignment of the gene for Wilson disease to chromosome 13: linkage to the esterase D locus. Proc Natl Acad Sci U S A. 1985;82(6):1819–1821. MEDLINE |
CrossRef
[11]. [11]Gheorghe L, Popescu I, Iacob S, et al. Wilson's disease: a challenge of diagnosis. The 5-year experience of a tertiary centre. Rom J Gastroenterol. 2004;13(3):179–185. MEDLINE [12]. [12]Roberts EA, Schilsky ML. A practice guideline on Wilson disease. Hepatology. 2003;37(6):1475–1492. MEDLINE |
CrossRef
[13]. [13]Vrabelova S, Letocha O, Borsky M, Kozak L. Mutation analysis of the ATP7B gene and genotype/phenotype correlation in 227 patients with Wilson disease. Mol Genet Metab. 2005;86(1–2):277–285. MEDLINE |
CrossRef
[14]. [14]Panagiotakaki E, Tzetis M, Manolaki N, et al. Genotype–phenotype correlations for a wide spectrum of mutations in the Wilson disease gene (ATP7B). Am J Med Genet A. 2004;131(2):168–173. MEDLINE [15]. [15]Scheinberg IH, Sternlieb I. Wilson's disease. In: Smith LH editors. Major problems in internal medicine. vol. 23:Philadelphia: W.B. Saunders Co.; 1984;p. 1–179. [16]. [16]Davies SE, Williams R, Portmann B. Hepatic morphology and histochemistry of Wilson's disease presenting as fulminant hepatic failure: a study of 11 cases. Histopathology. 1989;15(4):385–394. MEDLINE |
CrossRef
[17]. [17]Stromeyer FW, Ishak KG. Histology of the liver in Wilson's disease: a study of 34 cases. Am J Clin Pathol. 1980;73(1):12–24. MEDLINE [18]. [18]Harbin WP, Robert NJ, Ferrucci JT. Diagnosis of cirrhosis based on regional changes in hepatic morphology: a radiological and pathological analysis. Radiology. 1980;135(2):273–283. MEDLINE [19]. [19]Akpinar E. Liver imaging findings of Wilson's disease. Thesis. Ankara: Hacettepe University Press; 2003. p. 1–49. [20]. [20]Erden A. The prognostic value of a rare sonographic pattern in Wilson's disease. A case report. Acta Radiol. 1998;39(2):179. MEDLINE |
CrossRef
[21]. [21]Zachoval R, Glaser C. Wilson's disease before and after 5 years of treatment with d-penicillamine. J Hepatol. 1998;29(3):489.
Full-Text PDF (128 KB)
|
CrossRef
[22]. [22]Cancado EL, Rocha Mde S, Barbosa ER, et al. Abdominal ultrasonography in hepatolenticular degeneration. A study of 33 patients. Arq Neuropsiquiatr. 1987;45(2):131–136. [23]. [23]Corda R, Nurchi AM, Corrias A, et al. Importance and significance of liver echotomography in Wilson's disease during childhood. J Nucl Med Allied Sci. 1987;31(3):231–239. MEDLINE [24]. [24]Caprai S, Loudianos G, Massei F, et al. Direct diagnosis of Wilson disease by molecular genetics. J Pediatr. 2006;148(1):138–140. Abstract | Full Text |
Full-Text PDF (83 KB)
|
CrossRef
[25]. [25]Sanchez-Albisua I, Garde T, Hierro L, et al. A high index of suspicion: the key to an early diagnosis of Wilson's disease in childhood. J Pediatr Gastroenterol Nutr. 1999;28(2):186–190. MEDLINE |
CrossRef
[26]. [26]Gitlin JD. Wilson disease. Gastroenterology. 2003;125(6):1868–1877. Full Text |
Full-Text PDF (332 KB)
|
CrossRef
[27]. [27]Vogel W, Kathrein H, Dietze O, Judmaier G. Sonography of the liver in Wilson's disease. Sonographic studies of the liver in Wilson's disease—significance for assessing prognosis?. Ultraschall Med. 1988;9(6):270–273. MEDLINE |
CrossRef
[28]. [28]Vogel W, Kathrein H, Dietze O, Judmaier G . Ultrasonography in Wilson disease [letter]. Ann Intern Med. 1988;108:769–770. MEDLINE [29]. [29]Vogl TJ, Steiner S, Hammerstingl R, et al. MRT of the liver in Wilson's disease. Rofo. 1994;160(1):40–45. MEDLINE |
CrossRef
[30]. [30]Chu WC, Leung TF, Chan KF, et al. Wilson's disease with chronic active hepatitis: monitoring by in vivo 31-phosphorus MR spectroscopy before and after medical treatment. AJR Am J Roentgenol. 2004;183(5):1339–1342. [31]. [31]Dixon AK, Walshe JM. Computed tomography of the liver in Wilson disease. J Comput Assist Tomogr. 1984;8(1):46–49. MEDLINE |
CrossRef
[32]. [32]Ko S, Lee T, Ng S, Lin J, Cheng Y. Unusual liver MR findings of Wilson's disease in an asymptomatic 2-year-old girl. Abdom Imaging. 1998;23(1):56–59. MEDLINE |
CrossRef
[33]. [33]Iwadate H, Ohira H, Suzuki T, et al. Hepatocellular carcinoma associated with Wilson's disease. Intern Med. 2004;43(11):1042–1045. MEDLINE |
CrossRef
[34]. [34]Walshe JM, Waldenstrom E, Sams V, Nordlinder H, Westermark K. Abdominal malignancies in patients with Wilson's disease. QJM. 2003;96(9):657–662. MEDLINE |
CrossRef
[35]. [35]Lee HM, Lu DS, Krasny RM, et al. Hepatic lesion characterization in cirrhosis: significance of arterial hypervascularity on dual-phase helical CT. AJR Am J Roentgenol. 1997;169(1):125–130. [36]. [36]Baron RL, Marsh W, Oliver JH, et al. Screening cirrhosis for hepatocellular carcinonam (HCC) with helical contrast CT: specificity [abstract]. Radiology. 1997;205P:143. [37]. [37]Khalili K, Lan FP, Hanbidge AE, et al. Hepatic subcapsular steatosis in response to intraperitoneal insulin delivery: CT findings and prevalence. AJR Am J Roentgenol. 2003;180(6):1601–1604. [38]. [38]Rosenfield N, Grand RJ, Watkins JB, Ballantine TV, Levey RH. Cholelithiasis and Wilson disease. J Pediatr. 1978;92(2):210–213. Abstract |
Full-Text PDF (1353 KB)
|
CrossRef
[39]. [39]Uno A, Ishida H, Konno K, et al. Portal hypertension in children and young adults: sonographic and color Doppler findings. Abdom Imaging. 1997;22(1):72–80. MEDLINE |
CrossRef
Hacettepe University, Faculty of Medicine, Department of Radiology, 06100 Ankara, Turkey  Corresponding author. Tel.: +90 312 305 11 88; fax: +90 312 311 21 45.
PII: S0720-048X(06)00446-3 doi:10.1016/j.ejrad.2006.11.006 © 2006 Elsevier Ireland Ltd. All rights reserved. | |
|
FULL TEXT
