Abstracts

Rationale: The fornix and frontal white matter circuits are major white matter tracts whose structural integrity can be quantified using diffusion tensor imaging (DTI). However, structure-function associations for these tracts have not been mapped in children with epilepsy. In this preliminary study, we hypothesized that the structural integrity of the fornix and frontal white matter tracts (ventral and dorsal anterior forceps) as measured by DTI would reflect functional integrity of memory and executive circuits, respectively, as measured by neuropsychological tests in children with epilepsy. Methods: Children with focal epilepsy were recruited through the pediatric neurology program at Alberta Children s Hospital. All underwent magnetic resonance imaging (MRI) with DTI as part of a clinical protocol on a Siemens Avanto 1.5T MRI scanner (45 slices, 6 diffusion gradient directions, b=1000 s/mm2, voxel size 2 2 3 mm3). White matter integrity was measured by fractional anisotropy (FA) and apparent diffusion coefficient (ADC) using DTI Studio (Jiang et al. 2006). Tracts were computed after anatomically guided manual placement of ROIs by seeding voxels with FA greater than 0.25 and angle less than 70 degrees. All children were also administered tests of verbal and visual memory (California Verbal Learning Test, Children s Version; CVLT-C; Faces subtest of the Children s Memory Scale) and a parent-rated executive functioning scale (Behavior Rating Inventory of Executive Function; BRIEF). Analyses consisted of partial correlations between DTI (FA, ADC) and neuropsychological variables (CVLT Total, Faces Delay, BRIEF Global Executive Composite) controlling for age. Results: Participants were 18 children with focal epilepsy [8 girls, 12 boys; age = 11.4 years (4.6); age at epilepsy onset = 5.1 years (4.7); number of AEDs = 1.9 (0.8); number of prior failed AEDs = 1.7 (1.5); seizure frequency in last month = 8.4 (17.7); IQ = 80.1 (21.2)]. In all, 9, 5 and 4 children had a left-hemisphere, right hemisphere or bilateral focus, respectively. Left fornix ADC was significantly related to verbal memory (ADC, r = -.51, p = .05). There was a non-significant trend for an association between FA and verbal memory (r = .46, p = .08), but no significant associations between DTI measurements of the left fornix and visual memory or executive functioning. Right fornix FA and ADC were not associated with any cognitive domains. Dorsal but not ventral frontal white matter FA was highly related to executive functioning (r = -.66, p = .005). Frontal white matter FA values were associated with verbal memory, but not visual memory. Neither fornix nor frontal white matter FA or ADC were related to any epilepsy severity variables or to laterality of seizure focus. Conclusions: This preliminary study suggests a specific association between left fornix and verbal memory, and between frontal white matter and executive functioning. The study also indicates that the use of DTI as part of a clinical imaging protocol allows mapping of specific structure-function relationships in children with epilepsy.