Cranial MR findings in Wilson's disease

Transcription

1 Acta Radiologica ISSN: (Print) (Online) Journal homepage: Cranial MR findings in Wilson's disease I. Saatci, M. Topcu, F. F. Baltaoglu, G. Köse, K. Yalaz, Y. Renda & A. Besim To cite this article: I. Saatci, M. Topcu, F. F. Baltaoglu, G. Köse, K. Yalaz, Y. Renda & A. Besim (1997) Cranial MR findings in Wilson's disease, Acta Radiologica, 38:2, To link to this article: Published online: 04 Jan Submit your article to this journal Article views: 218 View related articles Full Terms & Conditions of access and use can be found at Download by: [ ] Date: 08 January 2018, At: 20:26

2 Actu Rudiologicu 38 (1997) Printed in Denmark All rights reserved Copyright 0 Artu Rudiologica 1997 A CTA RA DIOLOGI CA ISSN CRANIAL MR FINDINGS IN WILSON S DISEASE I. SAATCI, M. TOPCU, F. F. BALTAOGLU, G. KOSE, K. YALAZ, Y. RENDA~ and A. BESIM Departments of Radiology and *Pediatric Neurology, Hacettepe University Hospital, Ankara, Turkey. Abstract Purpose: To define various cranial MR appearances in Wilson s disease (WD). Material and Methods: MR examinations of 30 patients (944 years old) with WD were retrospectively reviewed. Six patients were asymptomatic siblings. Three other patients had isolated hepatic involvement, one with no symptoms. The remaining 21 patients had neurological involvement, 7 of whom had the mixed form of the disease. Nine patients had hepatic dysfunction, the 3 with isolated hepatic involvement and 6 of the 7 with the mixed form. Results: All symptomatic patients (n=23) had abnormal MR examinations. Atrophy was present in the majority of them. The most frequently involved sites were putamen (18/21) and pons (18/21) in patients with neurological abnormality. The putaminal lesions showed a consistent pattern of symmetric, bilateral, concentriclaminar T2 hyperintensity. Putaminal lesions were lacking in only 3 patients with neurological involvement, all of whom were relatively old and had had the disease for a longer duration. Most of the patients with hepatic dysfunction (8/9) had increased T1 signal intensity in the basal ganglia, particularly in the globus pallidus. Pontine involvement always included the dorsal aspect of the pons, however, in some cases the central portion of pons was also affected but ventrolateral longitudinal fibers were spared. Midbrain (16/21), thalamic (10/21) and caudate nucleus lesions (9/21) were also encountered. In a few patients cortical and subcortical white matter lesions were present with a predilection to the frontal lobe, particularly the precentral region. In one patient, a hemorrhagic focus was identified within the white matter lesion. Conclusion; On T2weighted images, WD is suggested by: atrophy; putaminal lesions with a pattern of symmetric, bilateral, concentriclaminar T2 hyperintensity; and the involvement of the pars compacta of the substantia nigra, periaqueductal gray matter, the pontine tegmentum and the thalamus. The hepatic component of WD may cause increased T1 signal intensity in basal ganglia. In the adult age group, the basal ganglia lesions may be different from those in the pediatric group; the putaminal lesions may not be present; the globus pallidus and substantia nigra may show increased hypointensity on T2weighted images. Cortical and subcortical lesions may also be present with a predilection to the frontal lobe. Key words: Central nervous system, Wilson s disease; MR; hepatolenticular degeneration; pediatric. Correspondence: Isil Saatci, Radiology Department, Hacettepe University Hospital, Sihhiye, Ankara, Turkey. FAX Accepted for publication 13 August Wilson s disease (WD) or hepatolenticular degeneration is a disorder of copper metabolism characterized by abnormal deposition of copper in various tissues, most pronouncedly in the liver and the brain. The cause of copper accumulation is still unknown, but it is postulated that the intrahepatic processing of ceruloplasmin may be disturbed, thus affecting both secretion into blood and excretion into the bile in the form of a proteaseresistant ceruloplasmin (9). In the brain, WD may involve primarily the corpus striatum but also the thalami, brain stem nuclei, cerebral cortex, cerebral and cerebellar white matter, and dentate nuclei (9). Neurological abnormalities may result directly from the deposition of copper in the brain, indirectly from hepatic dysfunction, or from a combination of the two (22, 30). Although it is an uncommon disease, early diagnosis is important as prompt therapy can prevent neurological sequelae. 250

3 MR IMAGING IN WILSON S DISEASE Table 1 Clinical features of and MR$ndings in our 30patients Pat/ Ageltypea Time Atrophy Pons Mid Subs. Periaq. Putamen Globus Caudate Thalami White Cortex Follow Sex at exam since brain nigra gray pallidus matter UP ination onset of matter syrnpt.b 1M 2M 3/F 4/F 5/F 6/F 7M 8/F 9M 1 OM 11/F 12N 13M 14M 15M 16M 17M 18F 19/F 20M 21M 22M 23F 24M 25M 26/M 27F 28M 29M 30M 12/N 2 years 17/N 4years 12lN 1 112y 12lA 12lA 13/N 3 1/2y 13lN 1/12y 14lN 2years 13lN 1112y 14lA 11M y 17M 10years 16lN 8 years 12/H 1 year 17lN 3 years 14/N 2years 13lA 21lA 12lA 15M 1 year 17/N 3years 13lN 5112~ 27/N 5 years 16/H 8112~ 44N### 5years 20M 5 years 9m 6/12 y 16N 8 years 33M 17 years 41lN 11 years # a N neurologic, H hepatic, A asymptomatic, M mixed. binterval (in years) since the onset of symptoms or findings regarding neurological and/ or hepatic involvement. * Lesions hyperintense on TIW. ** Lesions hyperintense on both TlW and T2W. *** Pandaface appearance. # Only cerebellar atrophy. ## Containing foci of hemorrhage. ### Hepatic involvement shown by biopsy current liver function tests within normal limits. i Increased hypointensity on T2W. In this report, the findings in 42 cranial MR examinations of 30 patients with WD are described. Material and Methods Cranial MR imaging of 30 patients with WD examined in the period were analyzed retrospectively. The material included followup MR examinations in 12 cases. The diagnosis was based on low serum ceruloplasmin and copper levels with increased 24h urinary copper excretion. The patients ages at the time of examination ranged from 944 years (mean 17.2 years, male/female 20/10), the majority of the patients being in the pediatric group, i.e. below the age of 18 years (24/30). Twentyfive patients had KaiserFleischer rings. Twentyone patients were neurologically sympto matic, including 7 with associated hepatic dysfunction, referred to here as the mixed form. Six patients were asymptomatic siblings, detected during family screening. The remaining 3 patients presented with the hepatic form alone, one of whom had only abnormal liver function tests with no symptoms. Altogether 9 patients had hepatic dysfunction, the 3 with isolated hepatic involvement and 6 of the 7 with the mixed form; the seventh was diagnosed by biopsy but had no hepatic dysfunction at the time. Family history was positive in 20 patients and 23 patients were children of consanguineous parents. Except for the asymptomatic siblings, all had been on penicillamine therapy when examined. MR examinations were performed on a 0.5 T Philips Gyroscan T5 system and a 0.5 T MR Max GE system. In all patients sagittal and axial T1

4 I. SAATCI ET AL. Fig. 1. Frequency of lesions according to location in 21 neurologically symptomatic patients. I 0 m c 07 P weighted (TlW) spinecho (TWTE /1325 ms), and axial T2W (TR/TE /20 and ms) images were acquired. Additional sagittal or coronal images were obtained with short or long TR sequences. The section thickness was 57 mm. Number of excitations (NEX) was 2 in T1W and one in T2W images. The time interval between initial and followup MR examinations varied between 4 and 18 months (mean 8 months). The MR images were evaluated by 2 radiologists (one of them specialized in neuroradiology) with respect to presence of atrophy, and focal signal changes in the cerebellum, pons, midbrain, putamen, globus pallidus, caudate nucleus and thalamus. Areas of involvement in the cortex and subcortical white matter were noted. In the followup examinations, the interval change (both in the intensity and in the extension of previously noted signal abnormality) was noted and, if present, new lesions. The examiners were not aware of the patient's clinical response to therapy. did not show any abnormality on the cranial MR. In all symptomatic patients (23/23), the cranial MR examinations were abnormal (Fig. 1). Of 21 patients with neurological involvement, 18 had diffuse atrophy and one had cerebellar atrophy. Two patients did not have atrophy. No cerebellar focal lesion was found in any patient. The majority of the focal lesions were hyperintense on T2W images, and hypo or isointense on T1W images except for some basal ganglia lesions. The most frequent focal finding was the bilateral Results The findings in the initial MR examinations are given in Table 1. In all 6 asymptomatic patients detected during family screening, MR examinations were unremarkable (patients 4,5, 10, 17, 18, 19). The asymptomatic patient (case 27) with abnormal liver function tests 252 Fig. 2. Patient 16, neurological type. T2W (TRITE 2002/80 ms) axial images showing a) the bilateral symmetric thalamic, lentiform and caudate nucleus lesions, and a) and b) bilateral frontal, parietal and left occipital cortical and subcortical white matter involvement. Note the concentriclaminar pattern of the thalamic and putaminal lesions. A focus of hemorrhage is present in the white matter lesion on the left () (b).

5 MR IMAGING IN WILSON'S DISEASE Fig. 3. Patient 26, mixed type. a) T2W (TRITE 2002/80) axial image shows the bilateral putaminal lesions with concentriclaminar pattern and caudate nuclei involvement. b) T1W (TR/TE 561/13) axial image at the same level as (a) demonstrates cystic areas in the putamen and slightly increased intensity in the globus pallidus (*). c) T2W (TRiTE 2002/80) axial image at the level of the pons discloses the involvement of the pontine tegmentum and an area of increased signal intensity in trapezoid configuration centrally, sparing a band of tissue anteriorly and laterally. symmetric involvement of the putamen with an invariable pattern on T2W images; the lesions were always hyperintense with a concentriclaminar appearance (Figs 2a, 3a). On T1W images, the putaminal lesions were either laminar or consisted of multiple millimetric foci, resembling the appearance of VirchowRobin spaces (Fig. 4a). In a few cases, single cavity formation was noted (Fig. 3b). The lesions were always hypointense except in patient 11 who had abnormal hepatic functions associated with the neurological involvement (Fig. 5b). The putamen was abnormal in all patients with neurological involvement except for 3 of the adult patients (18/ 21) (patients 23,25, 30) (Fig. 6a). Fig. 4. Patient 15, neurological type. a) T1W (TR/TE 561/13) and b) T2W (TR/TE 2002/80) axial images at the level of basal ganglia. The putamen is partially involved at the lateral aspect. On T1 W images there are punctate areas resembling the appearance of VirchowRobin spaces (). Involvement of the caudate nuclei (Figs 2a, 3a, 5a) and thalami (Figs 2a, 6a) were present in respectively 9 and 10 of the neurologically symptomatic patients, always bilateral and symmetric. The globus pallidus was affected in 15 patients. Globus pallidus lesions, which were hyperintense on't1w and isointense on T2W images, were the only abnormality in 2 patients (cases 14, 24) who had the hepatic form of the disease. In the remaining patients, all with neurological involvement, globus pallidus lesions accompanied the putaminal lesions, except in the adult patients (cases 25, 30) with normal putamen (Fig. 6a). The lesions were hyperintense on T2W images in patients with isolated neurological involvement. The patients, who also had abnormal liver function tests, showed increased TI signal intensity with the lesions either hyperintense or, less often, isointense on T2W images (Fig. 3b). Two patients in the adult group (patients 25, 30), who showed decreased signal intensity on T2W images, constituted the exceptions to the above generalizations (Fig. 6a). Pontine involvement was present in 18 of the neurologically symptomatic patients. Two patterns were recognized: the increased signal intensity confined to the dorsal aspect of the pons, the pontine tegmentum; and additional involvement of the central region of the pons always sparing a band of tissue anteriorly and laterally (Fig. 3c). In the latter pattern, unless the central pontine lesions were diffuse, a trapezoid configuration was recognized. Dorsal pontine involvement was sometimes isolated; 253

6 I. SAATCI ET AL Table 2 Interval changes in the clinical status and MRjndings in followup Pat. Change in clinical status Interval change in followup MR investigation 3 Dysarthria decreased Dystonia no longer present Decrease in putaminal, thalamic, brain stem and corticalsubcortical lesions 6 Neurological symptoms disappeared Brain stem and ~ucleus caudatus lesions improved 8 Dystonia and paresis decreased Decrease in basal ganglia and thalamic lesions 9 Unchanged New lesions in frontal subcortical white matter, substantia nigra 11 Dysarthria no longer present Pontine and caudate lesions disappeared 12 Dystonia decreased Unchanged Putaminal lesions regressed (decrease in size, TI hyperintensity disappeared) Hyperintensity of globus pallidus disappeared 14 Hepatic form Unchanged 15 Dystonia developed Unchanged 20 Minimal general improvement Unchanged 21 Rigidity developed, dystonia in the right arm Bilateral cogwheel sign New lesion in left globus pallidus 22 Minimal general improvement Unchanged however, the central involvement always accompanied the pontine tegmentum lesion. The substantia nigra was abnormal in 14 patients. The pars compacta was hyperintense on T2W images, with the pars reticulata spared, in all but the abovementioned 2 patients (cases 25, 30). These 2 patients showed increased hypointensity in the pars reticulata of the substantia nigra while the pars compacts was normal (Fig. 6b). Periaqueductal gray matter was involved with increased T2 signal in 12 of the patients, all of whom were neurologically symptomatic. The giant panda face appearance was recognized in 8 of the patients (Fig. 7). Six patients had subcortical white matter lesions. In 5 of them the adjacent cortex was also involved. Frontal and parietal lobes, particularly the precentral area, were affected in all (Figs 2b, 8). In one of these patients, temporal and occipital regions were also involved (Fig. 2b). Millimetric foci of hyperintensity on T2W images, equivocal in etiology, were not taken into consideration. Followup examinations were performed in 12 patients depending on the changes in the patient s symptomatology or, mostly, at the patient s own request. In some patients with interval changes in the clinical status, followup MR examination could not be obtained. Interval changes in the clinical presentation and followup MR of the 12 cases are summarized in Table 2. In 4 patients (cases 3,6,8, 11) out of 10 neurolog Fig. 5. Patient 11, mixed type. a) T2W (TR/TE 2007/80) axial image at the level of basal ganglia showing the putaminal and caudate nucleus lesions, bilaterally. b) T1W (TR/TE 520/20) axial image at the same level demonstrating the increased signal intensity of the caudate nucleus and putaminal lesions. c) Corresponding T2W (TR/TE 2001/80) axial images of the followup MR examination after 4 months of therapy showing the decrease in the size of the lesions. Note that the putaminal lesions are regressed rather medially. 254

7 MR IMAGING IN WILSON S DISEASE Fig. 6. Patient 2.5, mixed type. T2W (TWE 2203/90) axial images depicting a) the increased hypointensity of the globus pallidus () and b) the substantia nigra (). There are bilateral thalamic lesions (a). The ventricles and cortical sulci are enlarged owing to cerebral atrophy. Note that the putamen is unremarkable. Fig. 7. Patient 6, neurological type. T2W (TR/TE 2200/80) axial image demonstrating the involvement of mesencephalon with increased signal intensity in the periaqueductal gray matter and pars compacta of the substantia nigra giving an appearance of giant panda face. ically symptomatic patients who had clinical and MR followup, the MR lesions regressed in accordance with the notable clinical improvement (Fig. 5). Discussion WD may cause neurological abnormality either directly owing to copper accumulation in the brain, or indirectly as a consequence of liver dysfunction, or both (22, 30). When the brain is directly involved, the microscopic examination reveals edema, neuronal loss, gliosis, laminar necrosis, spongy degeneration, cavitation, iron and copper accumulation, and atrophy involving particularly basal ganglia but also the brain stem, cerebral and cerebellar white matter, and cortex (9). The pathological changes are presumed to result from excess copper or from ischemia secondary to vascular changes which can range from swelling of capillaries or arterial endothelium to perivascular thickening and proliferation of capillaries (30). The MR findings noted in our group of neurologically symptomatic patients (21/30) included atrophy (90.5%), pons (85.7%) and midbrain involvement (76.2%), putaminal (85.7%), thalamic (47.6%), caudate (42.9%) and globus pallidus lesions (61.9%) as well as cortex (23.8%) and white matter (28.6%) abnormalities (Fig. 1). In previous reports (1, 2, 5, 7, 10, 12, 14, 1723, 2830, 32, 33) the most frequently described abnormality was the involvement of the putamen, mostly showing bilateral symmetric increased T2 signal intensity and, less commonly, a decreased T2 signal resulting from copper accumulation or iron deposition secondary to basal ganglia destruction. The stage of the disease and the field strength of the imager by which the study was performed may have had an influence on the appearance of the lesions. In our series, except for the 3 adult patients who had been on penicillamine therapy for a long time, all patients with neurological symptoms had putaminal involvement. ROH et al. (24) demonstrated that the focal lesions may regress, more frequently when located in the thalami and the brain stem than in other locations. In their group of patients, who had not received any medical treatment previously, the involvement of thalami and brain stem was more frequent than that of the putamen. In the series of THUOMAS et al. (30), the patients who were examined while on therapy showed no or mild abnormality of lentiform nuclei whereas the patients investigated before treatment had lentiform nuclei involvement that regressed with treatment. Our initial observation of the patients in the pediatric age group was the invariable involvement of the putamen (31). However, among the older patients there were a few with putamen of normal appearance. In our group, except for some asymptomatic patients, all had been on therapy for various time periods when examined by MR. Therefore the putamen of normal appearance may have resulted from the treatment of previous lesions or the putamen may never have been involved. We recognized a constant pattern in the putaminal lesions on T2W images that were always bilateral, symmetric and hyperintense with concentriclaminar appearance (Figs 2, 3). The increased T2 signal in putamen most likely reflects the predomi 255

8 I. SAATCI ET AL. Fig. 8. Patient 3, neurological type. a) T1W parasagittal (TWE 520/20) and b) T2W (TR/TE 2001/80) axial images showing the cortical and subcortical white matter involvement of the frontal lobe, the precentral area in particular. The lateral ventricles and cortical sulci are enlarged. nant effect of edema, gliosis and laminar necrosis which may result in cavitation. Preferential involvement of the lateral part was noted when the putamen was partially affected, probably reflecting the centripetal direction of the glial response (21) (Fig. 4). Likewise, when the putaminal lesion regressed, the medial part was the first region to return to normal signal intensity (patient 11) (Fig. 5). On T1W images, we noted bilateral symmetric small foci of decreased T1 signal, giving an appearance similar to exaggerated VirchowRobin spaces in some cases. This may be due to the fact that a mild cavitation is primarily perivascular (9) (Fig. 4a). In some cases cavitation was so obvious that a slitlike appearance emerged (Fig. 3b). Five patients with the mixed form of WD (patients 12, 20, 26, 28, 29), who had involvement of both putamen and globus pallidus on T2W images, had T1 hyperintensity confined to the globus pallidus (Fig. 3). Two patients with the hepatic form also showed TI hyperintensity in the globus pallidus. An increased TI signal was seen in the putamen in one particular patient (patient 11) who had the mixed form of WD (Fig. 5). It is possible that this T1 shortening may be the first manifestation of copper deposition resulting from its paramagnetic effect (25, 30). However, excess copper is claimed to affect T2 relaxation time rather than T1 relaxation time (7) and in neither of the previous studies (5, 20, 21, 28, 33) reporting decreased T2 signal, was T1 shortening seen. Moreover, in WD the putamen is more likely to be affected than the globus pallidus. In acquired chronic hepatic dysfunction, an increased TI signal in the globus pallidus (3, 6, 11, 16, 26) and less frequently in the putamen (6) has been described. We believe that the observed T1 256 shortening in the basal ganglia in the mixed form was more likely due to the hepatic dysfunction than to copper deposition. Two older patients (cases 25, 30) showed increased hypointensity in the globus pallidus and substantia nigra (Fig. 6). It is known that the globus pallidus and pars reticulata of the substantia nigra are the 2 sites of preferential iron accumulation (8). The histopathological examinations revealed that phagocytes containing iron pigment were commonly seen in the globus pallidus and substantia nigra in WD (9). The thalami were involved in 10 patients with neurological involvement, showing hyperintense lesions with a laminar appearance on T2W images (Figs 2a, 6a). The giant panda face appearance was described by HITOSHI et al. as owing to increased T2 signal in the periaqueductal gray matter, in the pars compacta of the substantia nigra, and around the red nuclei which retained their normal hypointensity (12). In our series, 8 patients showed the giant panda face appearance (Fig. 7). We consider that the giant panda face appearance may be related to the severity of brainstem involvement and may appear in the course of the disease. The involvement of the pontine tegmentum (2, 7, 12, 29), base of the pons (14) and individual pyramidal and extrapyramidal tracts (32) has been reported. IMIYA et al. described hyperintensity in the base of the pons on T2W images in 3 WD patients with hepatic dysfunction and inquired whether these changes were related to central pontine myelinolysis resulting from the liver disease (14). In our series the pontine tegmentum was affected in all 18 patients with pontine abnormality. In some of them, the central part of pons was also involved either dif

9 MR IMAGING IN WILSON S DISEASE fusely or as a trapezoid configuration with the ventrolateral longitudinal fibers always spared. Although the pattern of the central pontine lesions resembled that of central pontine myelinolysis (1 3, 15), the patients with this pattern did not necessarily have accompanying liver abnormality in our group, i.e. some of the patients with the neurological form of the disease also showed this particular pattern. Moreover, patients with the isolated hepatic form did not have any pontine lesions. Thus we are convinced that the pontine lesions are the result of the direct involvement of the neurological WD rather than a consequence of central pontine myelinolysis secondary to liver dysfunction. Five patients had cortical and subcortical white matter involvement. It has been reported that new lesions, particularly those involving the cortex and subcortical white matter, may develop following the initiation of penicillamine therapy accompanying clinical deterioration (4, 27). However, we did not attribute these corticalsubcortical lesions to the therapy (although the patients did not have pretreatment studies and were on therapy when examined) because the related symptoms started before the initiation of the therapy and showed no further clinical worsening during the treatment. One of the patients showed foci of hemorrhage in the subcortical white matter (Fig. 2b): to our knowledge, hemorrhage within the white matter in WD has not been demonstrated before. In accordance with previous reports (2, 29), all 5 patients with cortical and subcortical white matter involvement showed frontal lobe predilection. What has not previously been described is the preferential involvement of the precentral area (Fig. 8) that we noted in all patients. Correlation between the neurological symptoms and MR findings has varied (2, 17, 22, 23, 29, 30). In accordance with some of those reports (2, 17,22, 29), all patients with neurological disorders in our group had abnormal MR examinations and the asymptomatic patients had unremarkable MR examinations. We conclude that the following characteristic MR findings should strongly suggest the diagnosis of WD. These include: atrophy; symmetric bilateral lenticular nucleus involvement (putamen rather than globus pallidus) with a concentriclaminar pattern of increased T2 signal; hyperintensity in the pars compacts of the substantia nigra, periaqueductal gray matter and pontine tegmentum on T2W images; and the less common involvement of the thalamus with a similar pattern to that of putaminal lesions. We propose that normal cranial MR examination in a neurologically symptomatic patient excludes WD. The putamen is affected in a centripetal fashion and, if regression of lesions takes place in response to therapy, it tends to start medially. The hepatic component of WD may cause similar changes in basal ganglia to those noted in acquired hepatocerebral degeneration due to chronic hepatic dysfunction which differs from those of neurological WD. In the adult age group, the basal ganglia lesions may be different from those in the pediatric group. The putaminal lesions may not be present, most likely due to the therapy. The globus pallidus and substantia nigra may show increased hypointensity on T2W images, possibly reflecting the augmented iron content. The cortical and subcortical involvement show frontal lobe predilection, particularly for the precentral area. White matter lesions may occasionaly contain areas of hemorrhage. REFERENCES 1. ABDOLLAH A., TAMPIERI D. & MELANSON D.: Wilson s disease. Computed tomography and magnetic resonance imaging findings. Can. Assoc. Radiol. J. 42 (1991), AISEN A. M., MARTEL W., GABRIELSENT. 0. et al.: Wilson s disease of the brain. MR imaging. Radiology 157 (1985), BRAFFMAN B. H., TROJANOWSKI J. Q. & ATLAS S. W.: The aging brain and neurodegenerative disorders. In: Magnetic resonance imaging of the brain and spine, p Edited by S. W. Atlas. Raven Press, New York BREWER G. J., TERRY C. A., AISEN A. M. & HILL G. M.: Worsening of neurologic syndrome in patients with Wilson s disease with initial penicillamine therapy. Arch. Neurol. 44 (1987), BRUGIERES P., COMBES C., RICOLFI F., DEGOS J. D., POIRIER J. & GASTON A,: Atypical MR presentation of Wilson s disease. A possible consequence of paramagnetic effect of copper? Neuroradiology 34 (1992), BRUNBERG J. A., KANAL E., HIRSCH W. & THIEL D. H. V.: Chronic acquired hepatic failure. MR imaging of the brain at 1.5 T. AJR 157 (1991), DE HAAN J., GROSSMAN R. I., CIVITELLO L. et al.: Highfield magnetic resonance imaging of Wilson s disease. J. Comput. Tomogr. 11 (1987), DRAYER B., BURGER P., DARWIN R., RIEDERER S., HERFKENS R. & JOHNSON G. A,: Magnetic resonance imaging of brain iron. AJNR 7 (1986), DUCHEN L. W. & JACOBS J. M.: Nutritional deficiencies and metabolic disorders. In: Greenfield s neuropathology, 5th edn., p Edited by J. H. Adams & L. W. Duchen. L. W. Arnold, London GRIMM G., PRAYER L., ODER W. et al.: Comparison of functional and structural brain disturbances in Wilson s disease. Neurology 41 (1991), HANNER J. S., LI K. C. P. & DAVIS G. L.: Acquired hepatocerebral degeneration. MR similarity with Wilson s disease. J. Comput. Assist. Tomogr. 12 (1988), HITOSHI S., MAKOTO I. & YOSHIKAWA K.: Midbrain pathology of Wilson s disease. MRI analysis of three cases. J. Neurol. Neurosurg. Psychiatry 54 (1991), Ho V. B., FITZ C. R., YODER C. C. & GEYER C. A.: Resolv 257

10 I. SAATCI ET AL. ing MR features in osmotic myelinolysis (central pontine and extrapontine myelinolysis). AJNR 14 (1993), IMIYA M., ICHIKAWA K., MATSUSHIMA H., KAGEYAMA Y. & Fu JIOKA A.: MR of the base of the pons in Wilson s disease. AJNR 13 (1992), KOROGI Y., TAKAHASHI M., SHINZATO J. et al.: MR findings in two presumed cases of mild central pontine myelinolysis. AJNR 14 (1993), KULISEVSKY J., RUSCALLEDA J. & GRAU J. M.: MR imaging of acquired hepatocerebral degeneration. AJNR 12 (1991), LAWLER G. A., PENNOCK J. M., SEINER R. E., JENKINS W. J., SHERLOCK S. & YOUNG I.R.: Nuclear magnetic resonance (NMR) imaging in Wilson s disease. J. Comput. Tomogr. 7 (1983), LENNOX G. & JONES R.: Gaze distractibility in Wilson s disease. Ann. Neurol. 25 (1989), MAGALHAES A. C. A., CARAMELLI P., MENEZES J. R. et al.: Wilson s disease. MRI with clinical correlation. Neuroradiology 36 (1494), MIRONOV A.: MRI in Wilson s disease of the brain. Neuroradiology (1991), Suppl. 33, p MIRONOV A.: Correspondence. Decreased signal intensity of the putamen and the caudate nucleus in Wilson s disease of the brain. Neuroradiology 35 (1993), NAZER H., BRISMAR J., ALKAWI M. Z., GUNASEKARAN T. S. & JORULF K. H.: Magnetic resonance imaging of the brain in Wilson s disease. Neuroradiology 35 (1993), PRAYER L., WIMBERGER D., KRAMER J., GRIMM G., ODER W. & IMHOF H.: Cranial MRI in Wilson s disease. Neuroradiology 32 (1990), ROH J. K., LEE T. G., WIE B. A., LEE S. B., PARK S. H. & CHANG K. H.: Initial and followup brain MRI findings and correlation with the clinical course in Wilson s disease. Neurology 44 (1994), RUNGE V. M., CLANTON J. A., SMITH F. W. et al.: Nuclear magnetic resonance of iron and copper disease states. AJR 141 (1983), SAATCI I., CILA A. & DINCER F. F.: Hyperintense basal ganglia lesions on TIweighted MR images in asymptomatic patients with hepatic dysfunction. Eur. Radiol. 5 (1995), SENER R. N.: Wilson s disease. Contrast enhancement of cerebral lesions on MR images after penicillamine therapy. AJR 163 (1994), SINGOHAROEN T., CHAKKAPHAK K. & UDOMPANICH 0.: Unusual magnetic resonance findings in Wilson s disease. Br. J. Radiol. 64 (1991), STAROSTARUBINSTEIN S., YOUNG A. B., KLUIN K. et al.: Clinical assessment of 31 patients with Wilson s disease. Correlations with structural changes on magnetic resonance imaging. Arch. Neurol. 44 (1987), THUOMAS K. A., AQUILONIUS S. M., BERGSTROM K. & WEST ERMARK K.: Magnetic resonance imaging of the brain in Wilson s disease. Neuroradiology 35 (1993), Topcu M., SAATCI I., DIREN B. et al.: MR imaging in Wilson s disease. J. Magn. Reson. Imaging 3 (P) (1993), VAN WASSENAERVAN HALL H. N., VAN DEN HEUVEL A. G., JANSEN G. H., HOOGENRAAD T. U. & MALI W. P. T. M.: Cranial MR in Wilson s disease. Abnormal white matter in extrapyramidal and pyramidal tracts. AJNR 16 (1995), YUH W. T. C. & FLICKINGER F. W.: Unusual MR findings in CNS Wilson s disease. AJR 151 (1988),