Abstract
Background: In Wilson's disease (WD), demyelination, rarefaction, gliosis, and iron accumulation in the deep gray matter cause opposing effects on T2-weighted MR signal. However, the degree and interplay of these changes in chronically treated WD patients has not been quantitatively studied.
Purpose: To compare differences in brain multiparametric mapping between controls and chronically treated WD patients with neurological (neuro-WD) and hepatic (hep-WD) forms to infer the nature of residual WD neuropathology. Study Type: Cross-sectional.
Population/Subjects: Thirty-eight WD patients (28 neuro-WD, 10 hep-WD); 26 healthy controls. Field Strength/Sequence: 3.0T: susceptibility, T2*, T2, T1 relaxometry; 1.5T: T2, T1 relaxometry.
Assessment: The following 3D regions of interest (ROIs) were manually segmented: globus pallidus, putamen, caudate nucleus, and thalamus. Mean bulk magnetic susceptibility, T2*, T2, and T1 relaxation times were calculated for each ROI.
Statistical Tests: The effect of group (neuro-WD, hep-WD, controls) and age was assessed using a generalized least squares model with different variance for each ROI and quantitative parameter. A general linear hypothesis test with Tukey adjustment was used for post-hoc between-group analysis; P < 0.05 was considered significant.
Results: Susceptibility values were higher in all ROIs in neuro-WD compared to controls and hep-WD (P < 0.001). In basal ganglia, lower T2 and T2* were found in neuro-WD compared to controls (P < 0.01) and hep-WD (P < 0.05) at 3.0T.
Much smaller intergroup differences for T2 in basal ganglia were observed at 1.5T compared to 3.0T. In the thalamus, increased susceptibility in neuro-WD was accompanied by increased T1 at both field strengths (P < 0.001 to both groups), and an increased T2 at 1.5T only (P < 0.001 to both groups).
Data Conclusion: We observed significant residual brain MRI abnormalities in neuro-WD but not in hep-WD patients on chronic anticopper treatment. Patterns of changes were suggestive of iron accumulation in the basal ganglia and demyelination in the thalamus; 3.0T was more sensitive for detection of the former and 1.5T of the latter abnormality.