About 150000 children and adolescents seek evaluation of a newly developed seizure disorder each year in the United States. And the incidence of epilepsy (recurrent unprovoked seizures) in this age group seems relatively consistent across all populations, ranging from 50 to 100/100000.1 Among these, 25% of patients continue to experience seizures despite receiving the best medical management availble, and may be defined as medically intractable. In these cases, if the seizures originate in a particular area of the brain and secondarily spread to involve other brain regions, a variety of surgical treatments may be beneficial, including focal cortical resection, corpus callosotomy, functional hemispherectomy, and multiple subpial resections.2

In order for an epilepsy surgery to be successful, one must not only localize the epileptogenic lesion, but also determine the anatomic localization of the eloquent cortex and functional white matter tracts in the brain in order to spare these areas during any planned cortical excision of the epileptogenic cortex.2 Accordingly, conventional structural imaging, magnetic resonance spectroscopy, and other advanced MR imaging techniques are widely used in epilepsy surgery; however, they cannot detect all causes of pediatric epilepsy.3

Diffusion tensor tractography (DTT) enables the graphical reconstruction of the white matter pathways in the brain and spinal cord of living humans. This technique has many potential clinical applications, including the investigation of stroke, multiple sclerosis, epilepsy, neurodegenerative diseases, and spinal cord disorders.4 There are even reports of its use in the alternation of DTT in cortical dysplasia and tumors.5-7 However, there has been limited study in regards to the usefulness of DTT in pediatric epilepsy surgery in the literature. In this study, we aimed to assess the clinical relevance of DTT in pre- and post-operative evaluations of childhood epilepsy according to types of surgeries performed.

Since the late 1990's, diffusion tensor technology has been utilized to visualize human white matter tract in vivo, as a routine clinical procedure for contemporary neuroimaging. DTT was found to be effective in neuro-oncology, neurodegenerative disease, and stroke imaging, but its clinical relevance in epilepsy surgery has still not been established. Although this study was not a well designed prospective study, we insist that the results of this review could be proof of the clinical usefulness of DTT in daily neuroimaging of epilepsy patients.

Diffusion tensor imaging captures anisotropic diffusion of water to delineate microstructural tissue organization. In brain imaging, the diffusion of water in white matter is greater in directions parallel to fiber tracts, allowing for axonal fiber delineation. In DTT, voxels are color coded according to the most prominent direction of water diffusivity in the axial, sagittal, or coronal plane. Conventionally, red is used for tracts in the left-to-right direction, green for the anterior-to-posterior direction, and blue for the superior-to-inferior direction. Three-dimensional image formation is also possible.9

Cortical dysplasia is one of the most common epileptogenic substrates responsible for intractable epilepsy in children, with a nearly 80% incidence in children younger than 3 years of age.10,11 Circumscribed lesionectomy of focal dysplastic lesions provides seizure relief in medically intractable patients with temporal and extratemporal epilepsy.12 Accordingly, precise lesion localization on presurgical imaging is important in these patients. Decreased anisotropy of white matter adjacent to the malformed cortex and an aberrant course of major fiber pathways due to dysplastic white matter are common findings in cortical dysplasia. And DTT can identify these microstructural changes based on elevated mean diffusivity and reduced anisotropy in the subcortical white matter of MR imaging-visible cortical dysplasia.5,6,11 In our study, 21 of 32 patients with cortical dysplasia had approximal proximity with major white matter tracts, i.e. located along the course of the corticospinal tract, dominant hemispheric deep white matter, etc. DTT clearly depicted a spatial relationship between the lesion and white matter tract and preoperative information aided neurosurgeons in avoiding injury during lesion resection.

Neoplasms of the central nervous system in pediatric patients often manifest clinically as seizures, and these are the second most common causes of epilepsy in pediatric patients.13 These tumors include ganglioglioma, gangliocytoma, desmoplastic infantile ganglioglioma, DNET, and pleomorphic xanthoastrocytoma.14 Our study also included cases of DNET and ganglioglioma. In regards to performing DTT in these patients, Yu, et al.7 showed that DTT allowed for visualization of the exact location of tumors relevant to eloquent tracts and was found to be beneficial in neurosurgical planning and postoperative assessment. Our study also showed the usefulness of DTT in four out of seven patients.

Hippocampal sclerosis is rare in pediatric patients.13 On DTT, diffusion anisotropy may reveal abnormalities in mesial temporal lobe epilepsy in adult patients.15,16 However, in children with a shorter duration of epilepsy, such changes may have not yet emerged.17 Our study also showed no change in DTT for patients with hippocampal sclerosis. A recent study showed that Meyer's loop of the optic radiation could be damaged after anterior temporal lobectomy and that the heterogenous anterior extent of Meyer's loop could be evaluated with DTT.18 However, there was no significant anterior extent of Meyer's loop in our patients and no visual field deficits were observed after operation. We deduced that DTT is not useful in pediatric patients with hippocampal sclerosis.

Corpus callosotomy, while not curative, may reduce the rapid generalization of discharges from the ictal onset zone and prevent harmful drop attacks from occurring, finally improving the quality of life for children.19,20 DTT and fiber tractography can reveal the hemispheric fiber connections of the corpus callosum.5 Accordingly, we applied DTT in pre- and post-surgical evaluation of corpus callosotomy, and the fiber connections were well-demonstrated in all cases, including the one case of incomplete resection. Therefore, we discerned that DTT is a useful sequence in the MR imaging evaluation of patients with corpus callosotomy for both before and after surgery.

In cases with functional hemispherectomy, lesions showed a loss of white matter tract on pre-operative study and DTT was not utilized for the evaluation of major white matter tract or complete resection on both pre- and post-operative imaging.

There are some limitations of our study. Firstly, DTT was not performed in all patients who underwent pediatric epilepsy surgery. DTT was not performed in the emergency unit. And if there was a tumor in some epilepsy patients, tumor protocol was performed without DTT. If DTT was not performed within 6 months before and/or one week after operation, these data were excluded. Additionally, as children grow, brain lesion can change during follow up. Therefore, we felt that performing imaging studies more than 6 months before operation is too long. We also thought that early postoperative DTT could be more accurate in evaluating complete resection. However, there may be selection bias. Secondly, we included not only pediatric patients, but also young adult patients (who were older than 18 years old, n=6). However, these patients suffered from epilepsy for a long duration, since childhood (duration range 10-20 years, mean 13.7 years). Moreover, the final diagnoses of these patients were cortical dysplasia in three patients, hippocampal sclerosis in two, and normal brain tissue in one. Therefore, we believe that these inclusions are appropriate. Thirdly, we did not evaluate DTT quantitatively. We only evaluated the relationship of the major white matter tracts and epileptogenic lesions, but did not measure the fiber volume or fractional anisotropy. Quantitative image analysis could provide more objective information and suggest a guideline for tissue characterization and surgical strategy. Accordingly, further study is needed to address this point.

Neuroimaging study in pediatric epilepsy surgery should be multimodal, and DTT is a useful technique for the evaluation of epilepsy surgery. DTT was shown to be more helpful in cases of focal epilepsy caused by cortical dysplasia and low grade brain tumors. Pre- and post-operative tractography was also shown to be useful in callosotomy procedure. In conclusion, DTT is a suitable routine pulse sequence for performing proper epilepsy surgery in children.