Abstracts

Rationale: Despite recent advances in pharmacological and surgical treatments, approximately 20-25% of epilepsy patients remain refractory. There are more than 7 new anti-epileptic drugs (AEDs) currently under clinical trials for patients with refractory epilepsy. These trials provide patients with the opportunity to try the newest pharmacological treatments. We reviewed refractory patients in clinical trials to evaluate if vagal nerve stimulation (VNS) has a significant impact on seizure prognosis while participating in clinical trials. Methods: We retrospectively reviewed epilepsy patients who previously participated in phase II and III clinical trials at the Barrow Neurological Institute from 2003 to 2008. Source documents were reviewed in order to obtain information regarding sex, age, age of seizure onset, number of previously failed AEDs, the use of VNS, and seizure frequencies during baseline, blinded, and open-label phases. Patients were divided into two groups: those with active VNS (VNS group), and those without (non-VNS group). Results: There were seven clinical trials with a total of 32 participants. Fifty nine percent were male and the median age was 39 years (ranged from 21 to 66 years). Thirty one percent received VNS prior to their enrollment and continued VNS without setting changes. The mean age of seizure onset for the VNS group was 7.2 years, compared to 13.5 years in non-VNS group. Before entering a clinical trial, the median number of previously failed AEDs for VNS group was 8 (4 to 11), while non-VNS group failed 6 (2 to 13). During baseline, VNS group experienced a mean of 64.6 seizures per month in comparison to non-VNS group, 29.9 seizures. For VNS group, the mean seizure reduction was 53.6% in blinded-phase and 38.2% in open-label phase. For non-VNS group, the mean seizure reduction was 23.4% in blinded-phase and 37.5% in open-label phase. When comparing VNS and non-VNS mean seizure reduction during both the blinded and open-label phases, no statistical differences were observed (p=0.28 and p=0.13, respectively). Conclusions: Patients with VNS tend to have more frequent seizures when they enter clinical trials. However, presence of VNS did not significantly alter the response to trial medications although there was a trend favoring patients with VNS in mean seizure reduction. Although a larger study is needed, the presence of VNS should be considered in the analysis of future clinical trials.