Moyamoya disease (MMD) is a progressive cerebrovascular disorder, with an unknown pathogenesis and aetiology. MMD is characterized by steno-occlusive changes at the terminal portion of the internal carotid artery (ICA), which is accompanied by variable development of the basal collaterals, also known as moyamoya vessels. Patients with MMD show variable patterns of brain damage and may experience recurrent multiple transient ischaemic attacks, intracranial bleeding and cerebral infarction. In this study, we investigate the potential for structural T1 magnetic resonance imaging (MRI) to help characterize abnormal cortical development in MMD clinically, with an analysis of both average and variability of regional cortical thicknesses. This study also included a machine learning analysis to assess the predictive capacity of the cortical thickness abnormalities observed in this research. This study included 993 MRI examinations from neurotypical controls and 269 MRI examinations from MMD patients. Results demonstrate abnormal cortical presentation of the insula, caudate, postcentral, precuneus and cingulate regions, in agreement with previous literature cortical thickness findings as well as alternative methods such as functional MRI (fMRI) and digital angiography. To the best of our knowledge, this is the first manuscript to report cortical thickness abnormalities in the middle temporal visual area in MMD and the first study to report on cortical thickness variability abnormalities in MMD.
Keywords: cortical thickness; machine learning; magnetic resonance imaging; moyamoya; paediatric.
© 2021 International Society for Developmental Neuroscience.