Abstracts

Rationale: The basal ganglia play an important role in the modulation and propagation of seizure activity. Recent neuroimaging studies have found that caudate volumes are reduced in juvenile myoclonic epilepsy. However, the precise location of atrophy pattern within the caudate nucleus has not been studied. The goal of the current study is to define the distribution of caudate atrophy in recent-onset JME. Using three-dimensional shape analysis, the study aims to test two competing hypotheses. On one hand, if the overall pattern of volume reduction in JME is nonspecific, then the distribution of caudate atrophy would be spread throughout the nucleus. On the other hand, if the caudate atrophy pattern reflects disruption of specific frontostiatal circuits, then atrophy should be localized to specific regions of the caudate. Methods: T1 weighted SPGR images were obtained on a 1.5 Telsa GE Signa scanner in 21 individuals with recent-onset JME (age = 15.8 /- 3.0 years; epilepsy duration = 8.5 /- 3.7 months) and 54 healthy controls (HC, age = 13.3 /- 3 years). Automated segmentation of the caudate nucleus was performed using FIRST (part of FSL, http://www.fmrib.ox.ac.uk/fsl)), which is a model-based segmentation tool that searches through linear combinations of shape modes of variation for the most likely shape, based on T1 image intensity. First, the automated segmentation process produces binarized masks of each subject s caudate nucleus. Total caudate volumes from each subject were calculated from these masks and then normalized to total intracranial volume (TIV). Second, the automated segmentation also generates a deformable mesh of vertices composed of a set of triangles. The relative position of each corresponding vertex is then compared between JME subjects and healthy controls in order to determine the specific regions of atrophy. A one-tail t-test was performed to compare TIV normalized caudate volumes between JME and HC subjects. Caudate three-dimensional shape analysis was carried out using F-statistics and multiple comparisons were corrected with false discovery rate. Results: Individuals with JME had significantly smaller bilateral caudate volumes when compared to HC (Left 5064 /- 361mm³ vs. 5368 /- 570, p=0.026; Right 5318 /-502 vs. 5631 /- 594; p=0.036). Three-dimensional shape analysis showed atrophy is located predominately in the head and body of the caudate nucleus. After correcting for multiple comparisons, the most consistent region of atrophy was located in the left dorsal and ventromedial head of the caudate. Conclusions: The current study demonstrates that caudate atrophy in JME is selective for dorsal and ventromedial head of the caudate. These two regions have critical connections to the dorsolateral and orbital prefrontal regions, areas that are important for executive functioning and mood regulation as well as seizure propagation. Further, disturbances in frontostriatal circuits are evident in this group of JME with very recent onset seizures, suggesting that caudate atrophy is less likely due to cumulative effects of seizures.