Abstracts

RATIONALE: Malformations of cortical development (MCD) are a common aetiology of epilepsy. Band heterotopia, bilateral subependymal heterotopia and lissencephaly have a genetic basis. Focal cortical dysplasias do not appear to be genetic. There have been no quantitative MRI studies of relatives of MCD patients. The aim of this study was to use quantitative morphometric tools to demonstrate gray matter abnormalities in relatives of patients with MCD and epilepsy.
METHODS: We identified 19 relatives of 10 probands with MCD (1 band heterotopia, 4 subependymal heterotopia and 5 focal cortical dysplasias). The control group of 58 healthy individuals were age and sex matched to the probands and relatives.
All controls, probands and relatives had high resolution MRI, comprising a T1-weighted conventional spin echo, a coronal oblique PD & T2 weighted conventional spin echo, a coronal T1 weighted 3D IR prep FastSPGR, acquired in a plane perpendicular to the temporal lobe and a coronal oblique FastFLAIR sequence. The relatives and controls had no neocortical abnormalities on visual inspection by two experienced neuroradiologists.
All 3D-data sets were corrected for non-uniformity, reoriented, spatially normalized, segmented, smoothed and analysed using in-house software (MRreg) and SPM99. Total numbers of gray and white matter voxels and ratios of ipsilateral cortical to subcortical matter in volumes of interest (VOI=10 equal thickness blocks for left and right hemisphere respectively) and ratios of ipsi- to contralateral volumes of homologous volumes of interest were also calculated. 3 probands and 1 control had to be excluded from the VOI comparisons due to processing difficulties with their data.
RESULTS: Voxel-based morphometric comparison showed significant increases of gray matter in 8 out of 10 probands and in 4 out of 19 relatives compared to 5 out of 58 controls. Fisher[scquote]s Exact Test (2-tailed) comparing relatives and controls p=0.2. Comparison of total numbers of gray and white matter voxels in the VOI method showed abnormalities beyond 3 standard deviations (SD) of the mean in 6 out of 7 probands, in 8 out of 19 relatives and 9 out of 57 controls. Abnormalities were significantly more common in the relatives of MCD patients than in the controls; Fisher[scquote]s Exact Test (2-tailed) p=0.026.
CONCLUSIONS: Voxel- and VOI-based methods are complementary. The data suggest subtle gray matter abnormalities exist in relatives of patients with MCD, implying a genetic component, even when relatives appear unaffected.
Support: MBM was supported by the Institut fuer Diagnostik der Epilepsien; TNM and SLF were supported by a grant from the Medical Research Council and the National Society of Epilepsy