We are honored to present an interview with the world-renowned epileptologist Dr. John Ebersole on the occasion of his retirement from the University of Chicago.
Brief bio: Dr. Ebersole is an internationally recognized authority on epilepsy and EEG with a main clinical interest in treating patients with intractable epilepsy. He completed his MD degree and neurology residency at Yale University. He did fellowship training at NIH and National Hospital of London. As the medical director of the Illinois MEG Center, he is developing techniques of functional imaging and localization using EEG and MEG. In addition to having published extensively, Dr. Ebersole is the editor-in-chief of the Journal of Clinical Neurophysiology and co-editor of the textbook Current Practice of Clinical Electroencephalography
Dr. Rui Guan: What are some basics about EEG / epilepsy that every clinician should know?
Dr. John Ebersole: Every clinician should have neurophysiology training. Clinical neurophysiology is being minimized and sometimes made an elective. So [some] physicians have no [formal] training in EEG when they go out in practice. So sometimes EEG reading in the community is not as high it is should be. Let’s face it, clinical neurophysiology are the only procedures that are ours [as] neurologists, and we should be expected to do them well. Neurologists that go out into [community] practice are expected to do [EEGs], wanted to do them, and should do them well. A poor reading is worse than no reading at all. Most of the time if an outside physician says an EEG is normal, it is normal. The usual error is over-reading — particularly with benign EEG variants like temporal theta waves.
RG: And sometimes false positive can cause more harm than a false negative.
JE: Absolutely. Once a tentative diagnosis of epilepsy is made, it sticks with them. And the patient has no more spells because the drugs are working, not because they do not have epilepsy to begin with.
RG: When would a primary physician refer to a neurologist and when would a general neurologist refer to a epileptologist?
JE: The good thing is that about 2/3 of seizure disorders can be controlled with one or two medications. And an internist can treat these patients without complications. Once a person has failed one medication, or if they have failed two medications, then they need to refer to a neurologist. And when that doesn’t work, then they need to be referred to an epileptologist. So it depends on on how well the seizures are controlled. When you’re an epileptologist, all you see are those patients who aren’t well controlled.
RG: Like intracranial [EEG monitoring] patients!
Dr. John Ebersole: Absolutely.
RG: When those patients come to you, because of their very complex disorders, how do you counsel them on their treatments and do you advise them about any clinical trials?
JE: Usually by the time I see the patients, they have not been on all the medications available, or their seizure disorders have not been characterized properly. So in someone who is supposedly intractable. So what we typically do is characterize the seizure disorders better – someone diagnosed with has generalized [seizures] actually has partial. Or maybe they don’t at all — and only have behavioral seizures. So characterization is the first thing. And then it is just a matter of trying various combinations [of mediations] that tend to work for more complex disorders.
Monitoring [is] to determine if their seizures are focal. To determine a source that is amenable to surgery. Because you know the famous study, that if you fail two medications, the likelihood that the third one is going to work is very small. Surgery, however, in select patients who have focal seizures from a location that can be removed can be a very good form of therapy.
RG: Should every patient who undergoes pre-op planning get a MEG so the resection location(s) are very accurately mapped out?
JE: MEG like all the other tests is helpful in localization. Particularly if you have a patient where you are considering implanting electrodes because you have insufficient localization, then another localizing test like MEG is more useful. MEG is a little more accurate than the EEG for localization, but it is principally used for localizing interictal spikes, so you need to know that the person’s seizures are coming from the same general area to take advantage of the localization accuracy of the MEG. So the answer is that everyone who is going to be implanted needs to have a MEG. For everybody who is not going to be implanted because they have classic mesial-temporal epilepsy with hippocampal atrophy with a history of febrile seizures — no they don’t necessarily need to have one if they go directly to surgery.
RG: You get a lot of referrals from neurologist and internists — what are some common mistakes when they first come to you?
Dr. John Ebersole: Not enough medication. Or switching medications without reaching one of two endpoints. The bottom line is you pick the medication that is most appropriate. You increase the medications until one of two things happens: The seizures stop or the patient develops toxicity. The therapeutic range is just a guideline. So some people can get away with less medication; and some people need more. When all else fails, look and talk to the patient.
Oftentimes people come in on several medications, but you are not sure whether the optimal benefit was reached with those medications, so sometimes you have to go back and try again.
RG: That makes sense. What are some big events have you you seen in the development of epilepsy management in your career?
JE: In my career, which is a long career, certainly epilepsy surgery is a big advance. And also advancing of intracranial EEG. When I was a resident, you initially had an 8-channel [EEG], and then a 16-channel, paper written! And initially, when we had intracranial originally, we used a 16-channel EEG, and every time the patient had a seizure, we would change which channel we were recording to try to localize it better. So going from an 8-channel EEG to 128-channel EEG made a big difference, at least in intracranial. And the other big interesting thing, on the flip side, since I started, there have been 10-12 new drugs introduced. But the unfortunate thing is that we still have a fair bit of patients that are intractable. So there have been no magic bullets pharmacologically that have wiped out seizures.
RG: Do you think there are anythings you are doing better now than when you were a more junior attending?
JE: Our electro-diagnostic abilities and EEG abilities are much better now. Remember when I started as a resident, we didn’t have CT let alone MRI. We had to use angiography and pneumoencephalopraphy! [chuckles] So the development of structural imaging is fantastic. And functional imaging as well. So we have many more tools than we had at one time.
RG: I know that you have pioneered some of the recordings of intra-cranial alongside extra-cranial EEG leads — are there any current projects that you are currently working on?
JE: Currently the projects that I am working on include determining the relationships between EEG and MEG. Unfortunately in most MEG centers in the country, let alone in the world, MEG is recorded by itself. They may record EEG to help them find spikes, but they don’t model the two. EEG and MEG both have weaknesses and strength. But the good thing is the the strengths are complementary. And they make up for the weaknesses they have. So to do both together and to develop some kind of combined or unified solution is much better than either alone.
RG: Dr. Ebersole, thank you so much for all your time! Any closing remarks?
JE: No closing remarks other than clinical neurophysiology is important for everybody to learn. Regardless of you’re going to do.
About the Author: Dr. Rui Guan is a neurology resident at the University of Chicago and a regular contributor to DocNeuro.com.
