Globus pallidus MR signal abnormalities in children with chronic liver disease and/or porto-systemic shunting.

OBJECTIVES: Detection of subclinical hepatic encephalopathy in children is difficult. We aimed to assess the changes in imaging of the central nervous system in children with chronic liver disease using MR imaging, diffusion, and 1 H -spectroscopy.

METHODS: Forty three children with chronic liver disease and/or porto-systemic shunting (111.4±56.9 months) and 24 controls (72.0±51.8 months) underwent brain MRI/spectroscopy on a 1.5T to examine T1, T2, ADC, Cho/Cr, ml/Cr, Glx/Cr ratio spectroscopy in the globus pallidus. Patients were divided into 3 groups according to the ratios of globus pallidus/putamen T1 signal : isointense (i), hyperintense (h), much more hyperintense (h+). The relationship with clinical and biological data was analyzed.

RESULTS: T1 signal intensity and ml/Cr were significantly different between controls and group h+ (p=0.001). ADC did not differ significantly between groups. Age correlated strongly with the presence of a T1 signal ratio (p > 0.001). There was no correlation between imaging findings and biological parameters.

CONCLUSIONS: In children with chronic liver disease and/or porto-systemic shunting, the presence of a hyperintense T1 signal in the globus pallidus correlated strongly with age. Biological and clinical parameters were not predictive of these changes. MRI may become a useful screening tool for hepatic encephalopathy in children.

KEY POINTS: • Children with chronic liver disease should undergo brain MRI during their follow-up • T1 hyperintensity of globus pallidus is suggestive of liver-related CNS involvement • MRS mI/Cr is decreased in children with chronic liver disease • Biological parameters (ammonium) were not predictive of hepatic encephalopathy • Duration of chronic liver disease may be causative the hepatic encephalopathy.