Abstracts

Rationale: VNS can reduce the severity of multifocal epilepsies as well as decrease the number and dosage of antiepileptic drugs (AEDs). Despite being in use for nearly two decades in the treatment of epilepsy, its mechanism remains largely unknown. Given that some patients with multifocal epilepsy demonstrate unifocality or even cure after VNS insertion, we postulate that VNS may provide neuromodulation which we aimed to quantify via spectral analysis of EEGs. Methods: The study population was identified by querying the electronic medical record at University of Chicago Medical Center for patients who underwent VNS insertion between 2007 and 2015. After excluding patients for insufficient medical history, incomplete EEG recordings or those with major artifact, we identified the 4 most recent subjects who fit our criteria. Using long-term monitoring video EEG recordings from the pre-VNS and post-VNS phases, we identified three types of activity ?" interictal spikes (IS), seizures (SZ), and awake background activity (BG) ?" in each phase. All EEG epochs were classified by a board certified physician in clinical neurophysiology. We then generated multitaper power spectra for each epoch and compared the standard EEG frequency bands between 1 and 70 Hz. Results: The four selected patients ranged from 4 to 14 years old, and each had unique clinical outcomes following VNS insertion. P1 demonstrated a shift from multifocal to unifocal left-sided epilepsy and underwent subsequent seizure focus ablation. P2 showed a decrease in seizure frequency post-VNS and later underwent a functional hemispherectomy. P3 had minimal clinical response to VNS insertion. P4 had intractable partial complex seizures and has been seizure-free for 2 years after VNS implantation. Spectral analysis revealed the principal changes in the lower EEG frequency bands, specifically the delta and theta ranges. Of note, our preliminary results indicate that a decrease in delta rhythm and increase in theta rhythm relate to clinical benefit. In the case of P1, who developed left-sided focal seizures, pre-VNS to post-VNS IS activity showed right-sided loss and left-sided gain in delta power during IS and theta gain during SZ. The patient who became seizure-free (P4) showed delta power loss and theta power gain in the BG activity. Patients with the least clinical benefit from VNS (P2, P3) did not demonstrate such changes in their delta and theta frequency ranges. Conclusions: Our results demonstrate that the majority of power distribution change in the EEG between VNS insertion took place in the lower frequency ranges, with those patients who had clinical benefit demonstrating a decrease in delta power and an increase in theta power. Funding: None