When Drugs and Surgery Don't Work, an Option in Epilepsy

Andrew N. Wilner, MD; Barbara C. Jobst, MD, PhD


June 26, 2017

Andrew N. Wilner, MD: Welcome to Medscape. I'm Dr Andrew Wilner and today I have the pleasure of speaking with Dr Barbara Jobst. Dr Jobst is an epileptologist and professor of neurology at Dartmouth Medical School in Hanover, New Hampshire. Dr Jobst, thanks for joining us.

We have the opportunity today to talk about a recent publication[1] of yours in Epilepsia about the responsive neurostimulator (RNS). Let's start by talking about what RNS is.

Barbara C. Jobst, MD, PhD: The RNS is actually an implanted device for epilepsy. It is implanted into the skull and has two leads that go into the brain where the patients' seizure onset is located. The doctor who implants the RNS needs to know where the seizures start in each particular patient. The device will then record and screen the EEG in the brain [for activity that could lead to a seizure]. Whenever a possible seizure is recorded, it will stimulate the brain [to disrupt the] abnormal brain activity.

The device has been in testing for a long, long time. We started testing it in 2005 and ran initial studies where it was shown that the RNS reduces seizure frequency in those patients. It was finally approved in 2013 by the US Food and Drug Administration (FDA) for general commercial use. Since then it has been commercially available and is approved for patients who have partial-onset seizures that are not controlled by regular antiseizure medications.

Dr Wilner: How did you learn so much about RNS? What's your experience?

Dr Jobst: We participated in the initial trials, which were conducted in many centers around the United States, to test the device in these patients, to study whether it is effective or not. And later on we also participated in follow-up studies. This is how I learned about the device.

Dr Wilner: What has the device taught us about seizures?

Dr Jobst: The device records intracranial EEG constantly, but it only saves snippets of the data captured. It has taught us a great deal about what an intracranial EEG looks like in patients who are walking around in their normal life. Previously, we would only receive this data in patients while they were strapped into the video-EEG monitoring unit and had just had surgery. At that time, we could only monitor patients for a few days; now we can monitor the patient for years. We can now also observe the seizures in patients' natural life and environment.

We now receive data over a period of years, and with that we have also learned interesting things. Historically, when we monitored this activity in the epilepsy unit, we could only monitor for 10 or 14 days. For example, we might have only seen seizures in a patient's left brain. Now, with prolonged monitoring, we may see patients who, all of a sudden, are also developing seizures out of the right side of the brain or vice versa. Because we were able to monitor patients for so long, we have been able to learn a great deal about the EEG data and what actual seizures look like in patients. We also learned that seizures often are not reported by patients, even though we see them on the EEG.

Dr Wilner: Sometimes patients will tell you, "I have this feeling that I am going to have a seizure soon."—not an aura where they are going to have it right away, but a prodrome. Is there such a thing? Can you see the prodrome on the intracranial recordings? Are there changes or is this still something that needs to be further defined electrically?

Dr Jobst: I think it needs to be further defined. It is always difficult to say what patients really feel. It could be that patients who have more epileptiform activity hours before their seizures may feel something. We haven't run a study where we've looked systematically at that within those patients. I can't give you a definite answer on that.

Dr Wilner: This sounds like a major process to install the responsive neurostimulator in a particular patient. Who would be a proper candidate?

Dr Jobst: In the decision to implant the RNS, first, we need to ensure that the patient has focal epilepsy. That means that the seizures start in a certain area in the brain. The patient should not have more than two seizure foci. The seizures should not start in more than two places somewhere in the brain, and we need to have knowledge of these two places.

The patient who has failed at least two medications, at sufficient doses, should be referred to an epilepsy center to be evaluated for epilepsy surgery. The epilepsy center will ensure that the patient has partial seizures and one or more seizure foci. If the patient has only one seizure focus and it can be resected with epilepsy surgery, then surgery should be performed.

However, there may be limitations associated with a surgical decision. For example, the seizure focus cannot be resected because it is in the language area or the motor area, the patient has two seizure focus rather than one, or the seizures come out of the left and right temporal lobe. These are ideal candidates for the RNS implant. They should go through a surgical workup, and if surgery cannot be performed, then the RNS should be considered.

Dr Wilner: If a physician has a patient with intractable epilepsy who's failed medical therapy, I guess step one should be to find a comprehensive epilepsy center nearby and let them see if the patient is a surgical candidate, which is certainly preferable. If you can cure the patient with surgery, that's where you want to go. This is a backup plan for patients who are not surgical candidates but who need something to help them control their seizures.

Dr Jobst: Or for patients who have already had surgery and it was not effective. Quite a significant number of patients in our study actually had had previous surgery.

Dr Wilner: The next question is, how well does it work?

Dr Jobst: RNS definitely works; it reduces seizure frequency, and that is the conclusion reported in the two studies that were just published. We've now published the long-term data on the initial patients who were in the study. One publication is about seizures that originate in the neocortical area—in the frontal lobes, in the lateral areas of the temporal lobe, and the occipital and parietal lobes.[1] The other was a study where we looked at mesial temporal lobe seizures only.[2] In the first group, which included the patients who had seizures from the neocortical areas, we followed 126 patients for 6.1 years. And if we divide them up, both of those brain lobes, the median seizure reduction was 70% in the frontal and parietal seizure patients.

Dr Wilner: That is a whole lot better than epilepsy drugs. Usually you don't get a 70% reduction in these intractable patients.

Dr Jobst: Usually adding on a new medication doesn't give you a 70% seizure reduction. The temporal neocortical and the multilobar onsets were 58% and 51%, respectively. It is really important that we not only reduce seizures but that we make patients seizure free. About a quarter of the patients were seizure free for 6 months and about 14% were seizure free for about a year. For patients who had a lesion on the MRI, we had information about where the seizures were located. These patients had a better outcome than those with a normal MRI. The patients with lesions experienced [over] a 70% seizure reduction; [those with normal MRI findings] had a 45% seizure reduction. This was in patients with neocortical onset.

We did a similar study in patients with a mesial temporal lobe onset with a similar follow-up of 6 years.[2] The results were very similar in that the median percent seizure reduction was 70%, and also [over] a quarter of these patients were seizure free for more than 6 months. This outcome is quite good.

Dr Wilner: I think it is interesting that the antiseizure effect seems to be enduring over time. Often with medications, we find that the patients do well initially, but then over time they need a different medication. With this neuromodulation therapy, the patients actually seem to do better over time.

Dr Jobst: We have observed that in several studies, and it's actually a general finding in any neuromodulation study that over time the seizures actually decrease. The longer the RNS is in place, the more the seizures decrease. It could be that over time the neuromodulation changes the physiology of the brain and overall decreases the seizure threshold. With this device, as it is a very sophisticated device, you have to program it to the patients' specific seizures. There is also a learning curve, initially, on how best to do this, so that may also have contributed to the better outcomes observed in these long-term studies. We still have to study that further—why it is that neuromodulation has better outcomes with time.

Dr Wilner: You're not off the hook. You're going to be writing the 10-year outcome and the 15-year outcome papers.

Dr Jobst: Generally, not off the hook, because ideally we want patients to be 100% seizure free. We have to learn more and we can certainly learn more from what we record—how to stimulate the seizure foci and how to do so differently. That's the next step.

Dr Wilner: Dr Jobst, I want to thank you very much for teaching us about RNS, which is an FDA-approved therapy for intractable epilepsy. Thank you for sharing your experience with Medscape.

Dr Jobst: Thank you very much.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.