Artikel
Graph theoretical network changes in presurgical epilepsy patients are independent of the underlying pathology
Graphentheoretische Netzwerkveränderungen in prächirurgischen Epilepsiepatienten sind unabhängig von der zugrundeliegenden Pathologie
Suche in Medline nach
Autoren
Veröffentlicht: | 26. Juni 2020 |
---|
Gliederung
Text
Objective: Several aspects of abnormal reorganization have been elucidated using a graph theory approach in describing the cortical network in presurgical epilepsy patients. While most studies concentrate on temporal lobe epilepsy, less is known about epilepsies caused by other pathologies, such as tumor, dysplasia, cavernoma, etc. The present study uses the principles of graph theory to support the hypothesis that the epileptogenic hemisphere shows similar graph theoretical network changes independent of the underlying pathology.
Methods: This retrospective study enrolled 25 patients (31.7 ± 25.6 years) with epilepsy deriving from various pathologies. These were divided in two groups: patients with tumour-induced epilepsy (n=15) and patients with epilepsy due to other type of pathology (n=10). Individual 3D MR images were coregistered to a brain atlas covering 88 region-of-interest (ROI). Finally, DTI connectivity between these ROIs and graph theory parameter (i.e. density and betweenness centrality) were calculated.
Results: The epileptogenic hemisphere shows higher density (0.33±0.06 and 0.29±0.06, p<0.05) but lower betweenness centrality (0.057±0.008 and 0.060±0.006, p<0.05) in comparison to the non-epileptogenic hemisphere. There was no group effect of the underlying pathology.
Conclusion: Our preliminary results show chronic epileptic attacks causing graph theoretical network changes within the epileptogenic hemisphere. These changes are independent of the underlying pathology.