Abstracts

Rationale: Patients with childhood absence epilepsy (CAE) experience 3 - 10 second, petit mal seizure episodes. These seizures are characterized by 3 - 4 Hz spike and wave discharges that cause impaired consciousness and a behavioral phenotype commonly described as staring spells. Approximately 10% of all children with epilepsy are diagnosed with childhood absence epilepsy, and the disease represents a significant cause of impaired attention and social dysfunction in school-age children. Even if seizure control is possible through medication, many children with CAE suffer from impaired attention between absence episodes. The mechanism of impaired attention between absence seizures (i.e. interictal impairment) is not known. Methods: In this study, we use a Continuous Performance Task (CPT) of attentional vigilance to examine interictal attention in children with CAE. During CPT, subjects observe a string of letters and are asked to press a button whenever they see the stimulus letter (the letter ‘X’). 32s periods of letter presentation are alternated with 32s epochs of cross-hatch fixation in a block design that is repeated 10 times per run of CPT. Our cohort consisted of 34 subjects with CAE (age 7 - 18) and 21 controls that were matched for age, sex, intelligence quotient, and socioeconomic status. Each of these subjects underwent simultaneous 3 Tesla fMRI-EEG and CPT testing. Behavioral performance during CPT (i.e. omission and commission errors as well as reaction time) was compared between CAE subjects and controls. Using Statistical Parametric mapping software, fMRI BOLD signal changes during CPT were modeled in both patient and control groups. A total of 100 runs of CPT in CAE subjects and 44 runs in control subjects were analyzed. In CAE and control subjects, fMRI signal changes during CPT runs were compared using a 2-tailed t-test (FDR, p=0.05, k = 3). CAE subjects were monitored for seizure episodes during CPT testing using EEG and all seizure-related fMRI signal changes during CPT were regressed out of the model. Results: Here, we report that, in the interictal period, CAE subjects made more omission errors during the CPT task than normal controls. In both patients and controls, CPT task fMRI increases were observed in medial frontal cortex as well as medial and bilateral prefrontal cortices. CPT-associated fMRI decreases were observed in bilateral medial and lateral parietal cortices. Conclusions: In the context of these findings, we hypothesize that the CPT task engages brain networks involved in attention which can be studied by fMRI. With further analysis, we hope to explain the interictal deficits in CPT performance observed in this patient population and identify differences in attention networks between CAE patients and controls. This work was supported by NIH R01 NS055829 (HB), by the Betsy and Jonathan Blattmachr family (HB), and by a Howard Hughes Medical Institute fellowship (MV). Special requests: No Saturday Poster Presentation (religious reasons)