NC-01 * ATTENTIONAL AND LANGUAGE FUNCTIONING IN ADULT PATIENTS WITH GLIOMA: A VOXEL-BASED LESION-SYMPTOM MAPPING STUDY

Abstract

OBJECTIVE: We statistically examined the relationship between lesion presence and neurocognitive performance on tasks of attention and language on a voxel-by-voxel basis in adult brain tumor patients using voxel-based lesion-symptom mapping (VLSM). VLSM is an innovative technique that has primarily been used in studies of stroke patients. METHODS: 98 right-handed adult patients with glioma were administered a neuropsychological battery comprised of 10 tasks: Digit Span Forward & Backward from the Wechsler Adult Intelligence Scale, Verbal Fluency (FAS, Animals), Woodcock-Johnson III Word Attack, and subtests of the Boston Diagnostic Aphasia Examination: Boston Naming Test, Complex Ideational Material, Commands, Responsive Naming, and Reading Sentences & Paragraphs. T2-weighted images were registered to a T1-weighted MNI atlas and the tumor region was contoured using a semi-automated thresholding technique. For a voxel, a general linear model was run to determine the relationship between cognitive performance and lesion presence, with medical and demographic characteristics specified as covariates. This analysis was repeated for each voxel with each subtest. Suprathreshold cluster-based analysis was used to correct for multiple comparisons. RESULTS: VLSM analysis identified several clusters of voxels with a significant relationship (p