Abstracts

Both acute and chronic mechanisms are believed to mediate the anticonvulsant effect of Vagus Nerve Stimulation (VNS). Animal data, and anecdotal evidence from patients suggest the presence of an immediate effect related to device activation. However, possible modulation of anticonvulsant effect by ON and OFF phases of chronic intermittent stimulation has not been explicitly looked for. Any such effect would have implications for the optimal choice of VNS duty cycles. Long-term Video-EEG monitoring studies performed in patients on chronic VNS therapy offer an opportunity to study seizure probabilities in relation to the phase of the VNS duty cycle., We retrospectively reviewed Video-EEG data from 17 studies performed in patients undergoing active VNS therapy for medically refractory partial epilepsy. Time of seizure onset relative to VNS duty cycle was determined based on the VNS artifact present in the EKG channel which was viewed at high gain after high-pass filtering. VNS parameters, including the durations of ON versus OFF periods, as well as epilepsy variables were recorded. The data was analyzed using a proportional hazard model that controlled for subject effects. The model was used to estimate the relative risk of a seizure during ON versus OFF periods. A simpler analysis using the Mann-Whitney non-parametric test was also performed to compare seizure counts across the two conditions in the patient group., There were a total of 199 seizures available for study. Of these, 129 (64.8%) occurred during OFF periods of the VNS duty cycle, while 70 (35.2%) occurred during ON periods. The ratio of ON to OFF periods ranged from 0.1 to 1.16 (0.64[plusmn]0.57). The number of seizures available per patient ranged from 2 to 69 (10[plusmn]18.7). Assuming an unconditional fixed subject-effect in the proportional hazard model, the hazard-ratio was 0.948 (95% CI= 0.624-1.440, p=0.802). Assuming a conditional fixed subject-effect the hazard ratio was 0.929 ( 95% CI = 0.607-1.424, p= 0.733). Mann-Whitney non-parametric test comparing the ON and OFF seizure rates between patients yielded a p = 0.53. These results point to the absence of any significant difference in seizure probability between ON and OFF periods of the VNS duty cycle. There were four patients who experienced all of their recorded seizures exclusively during VNS OFF periods. However, in all four instances, the number of seizures was too few for this to be a statistically significant observation., Our data suggest that the anticonvulsant effect of VNS is not modulated by the device turning ON or OFF during chronic intermittent stimulation. This questions the utility of magnet-activation of the device at the onset of seizures during chronic VNS therapy. It is possible, however, that there are individual differences in response to VNS that this study did not have sufficient power to discern.,