Language Mapping in Insular Gliomas: A Lesional Disconnectome Study

Introduction: The insula hosts one of the most complex patterns of cerebral functional connectivity. Patients with insular gliomas thus provide a unique opportunity to understand the relationship between higher cognitive functions, particularly language, and the underlying pattern of brain connectivity, formed by white matter tracts. This study aimed to link specific preoperative language deficits to distinct patterns of white matter disconnection.

Methods: Twenty-four patients with insular gliomas were included in this study. Manual lesion segmentation was performed on high-resolution MRI images using 3D Slicer. All patients underwent an extensive preoperative neuropsychological evaluation, including tests for verbal and object naming, phonemic and semantic fluency, repetition, and verbal learning. A voxel-based tumor disconnection analysis was conducted to identify a common pattern of tumor-induced disconnection for each specific language deficit. The results of the analyses were corrected using 5000 permutations and false discovery rate, with corrected p-values < 0.05 considered significant.

Results: Verbal naming deficits were independently associated with the disconnection of the left IFOF and arcuate fasciculus, while object naming deficits were also associated with the disconnection of the left uncinate fasciculus, ILF, and SLF III. Patients with deficits in phonemic fluency showed significant disconnection of the left arcuate fasciculus and left thalamo- and striato-cortical connections. The same disconnection pattern was identified in patients with deficits in repetition and verbal learning. No specific disconnection pattern was identified in patients with deficits in semantic fluency.

Conclusion : The results of this study suggest the presence of specific tumor disconnection patterns underlying various language deficits, with the potential to enhance our understanding of the relationship between function and white matter anatomy, and to support targeted and coordinated intraoperative stimulation and testing.