The primary finding of this study is the presence of DTI abnormalities in the white matter of children with very recent onset idiopathic epilepsies compared to control children. These diffusion abnormalities occur in the context of comparable volumes of lobar gray and white matter in the frontal, parietal, temporal and occipital lobes. Volumetric analysis of the corpus callosum also shows no difference for the total corpus callosum as well as for the anterior, posterior or middle subregions in particular. Taken together, these findings imply an early vulnerability of the structural integrity of cerebral white matter in new-onset epilepsy to which DTI, but not volumetric analysis, is sensitive. Our results extend previous DTI investigations of children with epilepsy (Govindan et al., 2008
; Kimiwada et al., 2006
; Nilsson et al., 2008
) by demonstrating that not only are these abnormalities evident in children with a relatively short duration of disorder, but that they can be seen near to the time of diagnosis of epilepsy, and occur in other epilepsy syndromes.
The corpus callosum has been previously implicated in adults with chronic temporal lobe epilepsy and to incur both DTI (Arfanakis et al., 2002
; Kim et al., 2008
; Thivard et al., 2005
) and volumetric (Hermann et al., 2003b
; Weber et al., 2007
) abnormalities. In the present study the posterior corpus callosum exhibits reduced FA, increased Drad
and increased MD in epilepsy patients compared to controls. Interestingly, the finding of abnormal DTI took place in the absence of volumetric change in the same callosal region. This finding is in direct support of the idea that DTI is able to detect abnormalities not evident using conventional MRI techniques (Duncan, 2002
). Other MRI studies have shown DTI abnormalities in the presence of volumetric differences (Gong et al., 2008
), however to our knowledge this is the first study of epilepsy patients to exhibit a structure specific abnormality in diffusion but not volume.
Structural studies of the normal development of white matter in children suggest that the anterior corpus callosum matures before the mid or posterior regions (Giedd et al., 1999
; Hasan et al., 2009
). Our finding of DTI abnormality in the posterior but not anterior corpus callosum may indicate a preferential damage to later myelinating callosal regions. Furthermore, radial but not axial diffusivity was altered in the posterior corpus callosum. Although there is no definitive microstructural change certain to underlie this pattern, there are a growing number of DTI studies relating altered Drad
to myelination abnormalities (Budde et al., 2007
; Song et al., 2005
) and showing a reduction of Drad
during normal white matter development in humans thought to correspond to compacting fibers and myelination (Alexander et al., 2007
; Hasan et al., 2009
; Snook et al., 2005
). It is possible that the myelination of axons in the patients is slowed by the epileptogenic process or that there is seizure-related damage to the posterior corpus callosum myelin sheath surrounding the onset of epilepsy. In support of the view of myelin as a primary target, there is no observed group difference in gray matter volume in any lobar region to suggest axonal degeneration secondary to neuronal loss, although the limitations of volumetric measurements are recognized as specific layers of the neuropil may be affected but not detected by these volumetric measurements.
The other region that displays abnormal DTI values is the cingulum, in which FA is reduced and Drad
is increased. This finding is in agreement with several studies of chronic temporal lobe epilepsy populations (Concha et al., 2005
; Concha et al., 2009
; McDonald et al., 2008
) and suggests a vulnerability of the cingulum that may accompany the onset of epilepsy. The internal capsule, external capsule and fornix were not significantly different from control values. It is possible that there is no change in these structures; however, they should not be ruled out of future studies without first addressing the limitations of this study.
Future studies will be needed to more fully characterize the structural state of the newly epileptic brain. This study provides an important initial characterization of DTI indices in this group, however there are several limitations that must be addressed. These include the small sample size and mixed epilepsy syndromes of the patient group and the imaging limitations of resolution and acquisition gap thickness, which was large and precluded the use of tractography or analysis of small white matter structures as well as measurement of complete white matter structures.
The finding of white matter vulnerability in this study is in conceptual agreement with previous MRI studies of patients with epilepsy. Diffuse abnormalities in cerebral white matter in focal epilepsy have been reported using both traditional volumetrics and voxel based morphometry (Hermann et al., 2003b
; McMillan et al., 2004
; Mueller et al., 2006
; Seidenberg et al., 2005
; Yu et al., 2008
). Vulnerability of frontal-temporal connections including the uncinate and arcuate fasciculi have been described in temporal lobe epilepsy (Lin et al., 2008
; Rodrigo et al., 2007
) with verification of this abnormal connectivity and demonstration of their important linkages to well known cognitive abnormalities (Diehl et al., 2008
; Flugel et al., 2006
; Lui et al., 2005
; McDonald et al., 2008
; Powell et al., 2007
; Yogarajah et al., 2008
). Understanding the cause, course, and consequences of microstructural abnormalities of cerebral white matter should help to understand the etiology of neurobehavioral comorbidities of epilepsy.