Abstracts

RATIONALE: The exact mechanism of action of vagal nerve stimulation (VNS) for the treatment of epilepsy is unknown. In animal models, changes have been demonstrated in the frequency of interictal spike activity and duration of seizures. No changes have been reported during direct visual analysis of EEG activity of awake, asleep or anesthetized human subjects or in the background EEG rhythms using quantitative methods. Previous studies from our institution have shown that there is an average decrease in the spike amplitude and duration and an increase in average sharpness following successful treatment with antiepileptic drugs (AED). The change can be quantified as a single value (also called [dsquote]Composite Spike Parameter[dsquote] or CSP) and is thought to reflect the active epileptogenic neuronal population coincident with seizure control. We applied this technique to patients undergoing VNS to see if improvement in seizure control produced similar changes.
METHODS: Patients with medically intractable partial seizures who were not candidates for resective surgery were prospectively enrolled from the Baylor Comprehensive Epilepsy Center at The Methodist Hospital. CSP anaylsis was performed on EEG data (first NREM period of night time sleep) acquired from the first night of video EEG monitoring (prior to VNS implantation) and was compared to EEG data acquired one month after initiating therapy with VNS. No changes were made to AED doses for this period. Seizure frequency was recorded using seizure calanders. VNS settings at 1 month: current strength-0.25-0.50 mA, On time-30 seconds, Off time-3 minutes.
RESULTS: See table.
CONCLUSIONS: Our preliminary analysis shows that improved seizure control following initiation of VNS is not associated with a decrease in CSP, as was observed following successful AED therapy. Only one patient exhibiting improvement in seizure control demonstrated a decrease in CSP while the other three patients exhibited a significant increase in CSP. This finding suggests that improved seizure control associated with VNS may involve a different neurophysiological mechanism than that underlying AED efficacy. Previous human PET studies have shown changes in thalamic glucose metabolism with VNS. It is possible that VNS affects pathways of spread of seizure activity rather than affecting the epileptogenic neuronal population. A larger data set is required to confirm our findings.[table]
Support: Research Grant from The Methodist Hospital Foundation and The Peter Kellaway, PhD Endowment For Research
Disclosure: Honoraria - Dr. Verma has recieved honoraria for speaking