Abstracts

Rationale: IGEs, genetically-mediated disorders, are relatively easily controlled with standard antiepileptic drugs (AEDs). A minority of patients with IGEs continue to have seizures despite appropriate treatment. Various factors are suggested to contribute to AED resistance in IGEs but detailed, large sample examinations of the contributing factors including EEG assessment are scarce. The purpose of this study was to determine the clinical and EEG predictors of medication response in patients with idiopathic generalized epilepsies (IGEs) and a subset of patients with juvenile myoclonic epilepsy (JME). Methods: We identified all patients with IGEs evaluated by epilepsy specialists between 11/17/08 and 5/17/09. We collected information on seizure freedom (dependent variable), EEG (all available past EEG reports and the last EEG recording were reviewed) and MRI characteristics, medication use, demographic information and seizure history (predictors). Fisher’s Exact tests or Student’s t-tests were used to compare predictors between those seizure-free and not seizure-free in the two groups: IGE patients (excluding JME) and JME. Logistic regression was used to estimate odds ratios for seizure freedom for significant predictors. Results: There were 1631 individual patient charts reviewed of which 208 were for patients with IGEs. Four subjects had incomplete records and were excluded. Of the 204 included patients, 69 had JME. JME patients were more frequently seizure-free compared to other IGEs types (79.7% vs. 68.9%; p=0.05). The presence of asymmetric GSWDs (30% amplitude difference; p<0.001), abnormal most recent EEG (p=0.05), and lack of complete seizure control with valproic acid (VPA; p<0.001) were associated with lack of seizure freedom. Gender, family history, presence/absence of GSWDs on the last EEG or other EEG abnormalities were not associated with seizure freedom. Later age of onset was associated with higher chance of seizure-freedom both for patients with JME and for patients with other IGEs (p<0.001 and p=0.013, respectively). Logistic regression indicated that factors associated with seizure freedom include later age of onset (p=0.002), diagnosis of JME (p=0.026) and initial response to VPA (p<0.001) and symmetric GSWD (p=0.004) while focal EEG abnormalities tended to be associated with lack of seizure control (p=0.060).