Abstracts

Whole-brain multimodal MRI phenotyping of periventricular nodular heterotopia.

Abstract number : 2.214
Submission category : 5. Neuro Imaging / 5A. Structural Imaging
Year : 2017
Submission ID : 345658
Source : www.aesnet.org
Presentation date : 12/3/2017 3:07:12 PM
Published date : Nov 20, 2017, 11:02 AM

Authors :
Francesco Deleo, Montreal Neurological Institute and Hospital, McGill University; Hong Seok-Jun, Montreal Neurological Institute and Hospital, McGill University; Anton Plavski, Montreal Neurological Institute and Hospital, McGill University; Benoit Caldai

Rationale: Periventricular nodular heterotopia (PVNH) is an epileptogenic malformation of cortical development mainly related to abnormal neuronal migration 1. Electrophysiological studies suggest an interplay between the heterotopic grey matter, the overlying cortex and the hippocampus in the generation and spread of seizures 2. While MRI has been instrumental in the classification and diagnosis of PVNH, no study has so far systematically assessed whole-brain integrity. Here we investigated markers of neocortical morphology, sulco-gyral patterns and local function, as well as indices of hippocampal maldevelopment. Methods: We studied 40 patients (mean age 29±8 years) with PVNH (22 unilateral, 18 bilateral) and 43 age- and sex-matched healthy controls. Based on a 3D T1-weighted MRI, we measured surface-based indices of cortical morphology (thickness) and folding complexity (absolute curvature). To define the overlying cortex, manually labelled PVNH nodules were projected onto the normotopic cortical surface along synthetic streamlines spanning from the ventricular wall. Hippocampal malrotation features 3 (medial positioning, curvature and longitudinal rotation) were combined into a multivariate score. In a subgroup of 16 patients and 18 controls who had resting-state functional MRI, we calculated vertex-wise local and global features (regional homogeneity, ReHo; degree centrality, DC), which were combined into a composite score. Analyses were corrected using random field theory, controlling the family-wise error at p < 0.05. Results: The PV nodules load was higher in the right (125 cm3) than the left hemisphere (169 cm3). Compared to controls, patients presented with increased cortical thickness of the right fronto-opercular, temporal (polar, lateral and basal) and the temporo-parietal cortices (p fig. 1A). In clusters of findings, there were no differences between patients with unilateral versus bilateral PVNH. The extent of anomalies largely overlapped with cortical areas overlying the nodules (fig. 1B). Mapping of local function revealed reduced ReHo and increase DC, particularly in the right temporo-parietal junction (largely overlapping with areas of cortical thickening), and the mesial temporal lobe; notably, there was a positive correlation between thickness and functional derivative changes (r=0.33, p fig. 1C) and positively correlated with hippocampal malrotation (r= 0.49, p=0.001). We did not observe any associations between nodular volume and cortical thickness, curvature and function.  Conclusions: This study provides evidence for widespread alterations of cortical and hippocampal morphology and function in PVNH. The severity of these anomalies relates to the location rather than the overall load of heterotopic grey matter. The multiplicity of abnormal imaging markers likely reflects both atypical migration, triggering nodular formation and increased cortical thickness in the overlying cortex (possibly related to delayed pruning), and abnormal post-migration phase, affecting cortical folding and hippocampal development. Funding: Canadian Institutes of Health Research References1.       Barkovich, A. J. et al. Brain 135, 1348–69, 20122.       Valton, L. et al. Clin. Neurophysiol. 119, 212–23, 20083.       Kim, H., et al. Front. Neurol. 4, 1–7, 2013
Neuroimaging