Abstracts

Rationale: Partial epilepsy can produce several types of deficits in neuropsychological functions and attention in children. The fronto-parietal superior longitudinal fasciculus (SLF) and cingulum connections play an important role in modulating attention. Diffusion Tensor Imaging (DTI) abnormalities of these tracts have been documented in patients with attention deficit hyperactivity disorder (ADHD). To our knowledge, DTI studies of the SLF have not been performed in pediatric epilepsy. One DTI study of the cingulum in pediatric epilepsy reported abnormalities in apparent diffusion coefficient (ADC), but no difference in fractional anisotropy (FA). We hypothesized that DTI would reveal abnormalities in the SLF and cingulum in children with partial epilepsy that are not detectable on conventional MR imaging. Methods: 7 children (3 boys) between the ages of 5 and 17 years with a definitive diagnosis of partial epilepsy underwent MR imaging on a 3T scanner. The mean duration of epilepsy was 2.9 2.24 years. 6 had medication-resistant epilepsy and 1 had controlled epilepsy. Children with frontal lobe pathology were excluded. Following clinical imaging, all children underwent a DTI sequence using the following parameters: single-shot echo-planar imaging, TE = 90 ms, TR = 8800 ms, 128 x 64 matrix, FOV of 220 mm, and 3-mm slice thickness. Images were acquired with diffusion weighting in each of 6 different directions, all with b-values of 0, 1000. DTI data was analyzed using DtiStudio, version 3.0.1. The SLF and cingulum were first identified using tractography and then regions of interest were placed in the mid-portion of the tract in each hemisphere by a single observer. ADC and FA data were compared to normal age- and gender-matched controls from a pre-existing DTI normative database obtained using the same sequence and field strength. Mean values for both hemispheres as well as values for individual hemispheres were compared between patients and controls. Results: Mean SLF FA of patients was 0.365 0.045 and for controls was 0.410 0.026 (Wilcoxon Signed Rank p = 0.0034 by hemisphere, p = 0.031 by patient). Mean SLF ADC was 77.7 6.8 x 10-5 for patients and 79.0 5.1 x 10-5 for controls (p = 0.22 by hemisphere, p = 0.69 by patient). Mean FA of the cingulum in patients was 0.411 0.062 and 0.477 0.048 in controls (p = 0.0061 by hemisphere, p = 0.047 by patient). Mean cingulum ADC in patients was 84.4 9.8 x 10-5 and 88.3 4.8 x 10-5 in controls (p = 0.25 by hemisphere, p = 0.38 by patient). Conclusions: Our findings indicate abnormal water diffusion in the SLF and cingulum of children with partial epilepsy in otherwise normal-appearing white matter regions. These abnormalities likely reflect microstructural white matter changes that potentially could contribute to neuropsychological co-morbidity by diminished integrity of a white matter tract that is important in attention. DTI appears to be a useful tool for identifying white matter pathology in children with partial epilepsy that cannot be depicted on conventional MR imaging.