Abstracts

Fronto-Temporal Phenotypes in Pediatric Epilepsy

Abstract number : 2.135
Submission category : 5. Neuro Imaging
Year : 2010
Submission ID : 12729
Source : www.aesnet.org
Presentation date : 12/3/2010 12:00:00 AM
Published date : Dec 2, 2010, 06:00 AM

Authors :
Prabha Siddarth, B. Hermann, J. Levitt, S. Gurbani and R. Caplan

Rationale: Volumetric abnormalities are related to psychopathology, impaired cognition, and linguistic deficits in children with and without epilepsy. However, in pediatric epilepsy these comorbidities are differentially related to epilepsy variables, to each other, and to brain volumes. To begin to delineate the brain/behavior profiles of comorbidities found in pediatric epilepsy, we used cluster analysis to determine if these children have specific fronto-temporal volume combinations compared to healthy children. We then examined if these volume clusters were related to the comorbidity profile of the epilepsy subjects. Methods: 116 children, 45 with complex partial seizures (CPS), 21 with childhood absence epilepsy (CAE), and 50 healthy children underwent MRI scans at 1.5T. Tissue was segmented and total brain, frontal lobe, frontal parcellations (inferior frontal gyrus (IFG), orbital frontal gyrus (OFG), dorsolateral prefrontal cortex (DLPFC)), and temporal lobe (TEMP) volumes were computed. Parents and medical charts provided epilepsy-related information. Psychiatric interviews, IQ, language and achievement tests, parent and child report questionnaires on behavior problems, depression, and anxiety were administered to all subjects. Agglomerative hierarchical clustering was performed on the basis of Euclidean distances computed from the frontal and temporal gray and white matter volumes for each diagnostic group. The resulting clusters were compared on demographic, seizure, psychopathology, cognitive, linguistic, and achievement variables using MANCOVAs. Results: For both epilepsy groups, the cluster analysis identified two distinct clusters of similar size. One cluster had lower OFG and DLPFC gray and white matter volumes and higher IFG and TEMP gray and white matter volumes (Low OFG/DLPFC). The other cluster showed higher OFG and DLPFC volumes and lower IFG and TEMP volumes (High OFG/DLPFC). The two clusters were not different on demographic, seizure, and cognitive variables except age (Low OFG/DLPFC: 10.9 2.5 vs. High OFG/DLPFC: 9.0 1.9, p<.001). Controlling for age, the Low OBG/DLPFC cluster was associated with significantly more children with a psychiatric diagnosis (61% vs. 37%, p<.05), Child Behavior Checklist Total and Social Problems (52% vs. 23%, p<.02), suicidal ideation (45% vs. 7%, p<.02), language (53% vs. 24%, p<.01), and math achievement deficits (18% vs. 3%, p<.04) than the High OFG/DLPFC cluster. The controls formed a cluster with higher IFG and OFG volumes and lower DLPFC and TEMP volumes (High OFG/IFG) and another cluster with higher DLPFC and TEMP volumes and lower IFG and OFG volumes (Low OFG/IFG). These clusters did not differ on demographic, psychopathology, cognitive, linguistic, or achievement variables.
Neuroimaging