Abstracts

Rationale: Brain-responsive neurostimulation with the RNS® System provides treatment by means of a neurostimulator that detects epileptiform activity including  ictal onset patterns and subsequently  triggers  a small stimulus directly to the seizure focus though intracranially placed leads.  Following the RNS implantation the device is programed to detect epileptic patterns specific to an individual patient.   These patterns are represented in the reporting program in the form of electorcorticographpy (ECoG) and other descriptive ways eg  “long episodes”- defined as device-identified patterns lasting longer than a specified time period.  Once the detection is optimized the device is programed to deliver appropriate therapies with the  goal of reducing seizure frequency and severity.  Detections and treatments as well as patient initiated magnet swipes are collected by the device and are available for analysis by a clinician.  The purpose of our study is to identify signature differences in device-recorded parameters between responders and non-responders with the goal of early identification and  adjustment of treatment parameters. Methods: Twenty three consecutive patients treated with the RNS System at University of California, Los Angeles (UCLA) were analyzed.   Those with an appropriate amount of data to perform the analysis were selected (n=7). We examined  changes over time in individual ECoGs, number of patient initiated magnet swipes, device initiated recordings of long episodes and number of delivered therapies (per episode and per day). Results: Our data indicates that patients who report clinical improvement following implantation of RNS, showed a downward trend in the number of long episodes identified by the device, as well as downward trend in number of therapies delivered per day.   This trend was evident in the first 2-3 months following initiation of treatment.  Interestingly, ECoG recordings did not change overtime and neither did number of therapies delivered per episode. Conclusions: Our findings suggest that there are a few parameters that differentiate responders and non-responders to RNS treatment and that these differences can be seen as early as 2-3 month following stabilization of detection parameters. Thus, in order to achieve full benefit, optimizing detection/stimulation parameters in non-responders within the first few months following it’s activation is recommended. Funding: none