T132. Connectivity analysis of glioma patients with language disturbances. (May 2018)

• Record Type: Journal Article
• Authors: Leote, Joao; Loução, Ricardo; Nunes, Rita G.; Viegas, Catarina; Lauterbach, Martin; Silvestre, Ana; Monteiro, Joana; Pérez-Hick, Antonio; Ferreira, Hugo A.
• Is Part Of: Clinical neurophysiology. Volume 129(2018)Supplement 1
• Page Start: e53
• Page End: e54
• Publication Date: 2018-05
• Subjects: Neurophysiology -- Periodicals; Electroencephalography -- Periodicals; Electromyography -- Periodicals; Neurology -- Periodicals; 612.8
• DOI: 10.1016/j.clinph.2018.04.133
• Languages: English
• ISSNs: 1388-2457

Abstract

Abstract : Introduction: Language production involves a complex brain network that should be taken into account in pre-surgical planning of patients with brain tumors. Connectivity analysis (functional and structural) of magnetic resonance imaging data (cMRI) enables the study of the functional interaction and structural relationship between brain regions. Our goal was to correlate cMRI data with clinical manifestations of language disturbances Methods: We recruited sixteen patients (mean age of 47 years old; 8 female) diagnosed with low and high grade brain gliomas (47%) located in frontal (57%), temporal (17%), parietal (13%) lobes and in the insular region (13%). They performed pre-operative clinical language tests, structural (diffusion tensor imaging – DTI), and functional MRI (fMRI) using a battery of language paradigms. We extracted connectivity metrics (i.e. node degree, betweeness centrality, partition coefficient, and clustering coefficient) from thirteen brain regions and normalized the data (i.e. using mean and standard deviation) within each region and across patients. For a set of brain regions involved in language processing (i.e. opercular, supramarginal and superior temporal gyrus), we evaluated the correlation between local connectivity metrics and brain tumor location, language clinical tests and the outcome of cortical stimulation during awake craniotomy (i.e. whether it produced language disturbances). Results: We analyzed seventeen MRI datasets (i.e. one patient received treatment twice and three others had previously performed craniotomies) including DTI data in only twelve patients. Clinically, three presented conduction aphasia (two with temporal and one with a frontal lobe tumor), four had anomic aphasia (two with frontal, one with insular and another with temporal lobe tumor) and two with expressive aphasia with frontal lobe tumors. The extracted metrics failed to correlate with tumor location. Language tests correlated with higher partition ( r = 0.75) coefficient of opercular or supramarginal gyrus. Cortical stimulation at opercular or supramarginal gyrus correlated either with a higher clustering ( r = 0.68) or lower partition ( r = −0.64) coefficient in supramarginal or opercular gyri, respectively. Conclusion: cMRI from DTI and task-based language paradigms correlates with clinical evaluation language tests as expected. Also, connectivity measures suggest a homo-lateral compensation in the language network when opercular or supramarginal gyri are affected besides contralateral as reported in literature.