Network-behavior mapping of lasting executive impairments after low-grade glioma surgery.
Jérôme CochereauAnne-Laure LemaitreMichel WagerSylvie Moritz-GasserHugues DuffauGuillaume HerbetPublished in: Brain structure & function (2020)
Executive functions (EF) may be significantly impaired following low-grade glioma (LGG) surgery, especially in the event of white matter (WM) disruption. The aim of this study was to identify the connective tracts associated with EF impairments after LGG surgery, and to provide new insights into the WM network architecture of EF. EF measurements were collected in 270 patients at the chronic postoperative phase. This comprised cognitive flexibility, verbal inhibition and fluency abilities (phonological and categorical). The scores were z-corrected for age and educational level, and further submitted to a principal component analysis (PCA). Tracwise and disconnectome-behavior analyses were then performed using EF measures independently but also the extracted components from PCA. For the first analyses, 15 tracts of interest were selected. Two principal components were extracted from the behavioral data, interpreted as 'EF' and 'language' components. Robust, bonferroni-corrected correlations were established between the EF component and Layers II and III of the left superior longitudinal fasciculus, and between phonological fluency/inhibition and the same tracts. Less powerful but still significant correlations were also observed with the left frontal aslant and fronto-striatal tracts. These results were confirmed by disconnectome-behavior analyses. Our results indicate that surgically-related disruption of the fronto-parietal and the frontal cortico-subcortical connectivity, and of the frontal aslant tract, is related to long-lasting EF impairments. In addition to providing new insights into the WM pathways supporting EF, these findings are especially useful for both surgical planning and the predictive approach of neuropsychological disorders in the context of LGG surgery.
Keyphrases
- working memory
- low grade
- minimally invasive
- coronary artery bypass
- white matter
- high grade
- functional connectivity
- surgical site infection
- multiple sclerosis
- high resolution
- mild cognitive impairment
- atrial fibrillation
- machine learning
- coronary artery disease
- artificial intelligence
- percutaneous coronary intervention