Utility of diffusion tensor imaging parameters for diagnosis of hemimegalencephaly

BACKGROUND: Hemimegalencephaly is a rare hamartomatous entity characterised by enlargement of all or part of the cerebral hemisphere ipsilaterally with cortical dysgenesis, large lateral ventricle and white matter hypertrophy with or without advanced myelination. Although conventional magnetic resonance imaging (MRI) is useful for detecting these diagnostic features, hemimegalencephaly is not always easily distinguished from other entities, especially when hemimegalencephaly shows blurring between the grey and white matter. Diffusion tensor imaging (DTI) is a functional MRI technique commonly used to assess the integrity of white matter. The usefulness of DTI in assessing hemimegalencephaly has not been fully elucidated. In this study, we clarified the characteristics of hemimegalencephaly with regard to DTI and its parameters including fractional anisotropy and apparent diffusion coefficient. METHODS: Three patients with hemimegalencephaly underwent MRI including DTI. We first visually compared fractional anisotropy mapping and conventional MRI. Next, we quantitatively measured the fractional anisotropy and apparent diffusion coefficient values in the subcortical white matter of the hemisphere with hemimegalencephaly and corresponding normal-appearing contralateral regions and analysed the values using the Mann-Whitney U test. RESULTS: On fractional anisotropy mapping, we could clearly distinguish the junction of grey and white matter and observed thicker white matter in the hemisphere with hemimegalencephaly, which was unclear on conventional MRI. The white matter in the hemisphere with hemimegalencephaly showed significantly higher fractional anisotropy (P<0.0001) and lower apparent diffusion coefficient (P=0.0022) values than the normal contralateral side. CONCLUSION: DTI parameters showed salient hemimegalencephaly features and could be useful in its assessment.