Welcome to our live stream, Ask the Expert webinar. How do we know we have missed the SEEG seizure onset zone? Thank you for joining us. My name is Ryan and I'll be the operator for the presentation today. We are joined today by our moderator, Dr. Jaysing Singh and our speaker, Dr. Philip Kahan. At this time, I'd like to turn the microphone over to our moderator, Dr. Singh for opening remarks. Thank you. Thank you, Ron. Welcome to everyone for this live webinar. Ask the Expert webinar. How do we know we have missed the SEEG seizure onset zone? My name is Jaysing Singh. I'll be moderating today's webinar. I'm a surgical epileptologist and a medical director of epilepsy surgery program at Ohio State University and a part of an AES online education committee. I have no disclosure relevant to this talk and it's my pleasure to introduce today's speaker, Dr. Philip Kahan. Dr. Kahan is a professor of neurology at Grand Rapids Hospital and University of France. He's also a section head there. He's an international expert, does not need any introduction on this topic of SEEG monitoring for pre-surgical evaluation of a drug-resistant epilepsy in adults and children. The goal or the objectives of today's webinar, hopefully we'll be able to achieve is to understand why SEEG may miss the seizure onset zone. Also recognize the main situation suggesting that electrodes are not at the right place and determine what to do in those scenarios when we do miss out SEEG seizure onset zone. Please feel free to enter your questions in your Q&A box. I'll try to answer or respond to them with Dr. Kahan as we receive them during the presentation. And now I will turn over the presentation to our presenter, Dr. Kahan. Thank you very much, Jay, for this very kind introduction. So today I'm gonna talk to a very strange topic, which means how can we understand that we have missed the seizure onset zone when doing SEEG. So I have nothing to disclose for that presentation. And this is the outline of my talk. A few words on SEEG, because you probably all know the technique. And then I will give you a few example on how we know that we probably have missed the seizure onset. I will give you just a few words on the perspectives and then it will be your turn to work a bit on a few cases. Okay, so SEEG, it works. Definitely, it's a good technique for epilepsy surgery patients, but there are some pitfalls that you have to be aware of. So it's clear that SEEG is a method that seems to be safer than subdural electrode recordings, especially regarding bleedings. And also it seems that SEEG might offer a greater chance of seizure freedom than when using subdural electrodes. It has not been proven definitely, but if you're looking at these two studies, for instance, you can see that indeed the chance of becoming seizure-free seems to be a bit higher when using SEEG electrodes than when using strips or grids. But of course, and this is a major issue, SEEG does not guarantee at all surgery nor seizure freedom, which means that there are substantial number of patients for which after SEEG recording, we do not do anything. And also even when operated on, you see that the rates of patients that become seizure-free is not 100%. So definitely the method is not a magic method and there are reasons for that. The first is that we are using only a very small number of electrodes and the number of contacts that are sampling the whole brain and to cut a long story short, you can see that when using a mean number of 12 electrodes, which represents something like 150 contacts within the gray matter, we are sampling only 0.5% of the whole cortex. And of course, this is a crucial point. And with years, you can see here on the right, with years, the number of electrodes that have been implanted in patient has increased only from 12, mean of 12 electrodes per patient, to 16 electrodes per patient, which is already a big number of electrodes. But even with this, we will have a sampling of the gray matter, which is quite low. So what does that mean? It means that SEEG implantation needs strong hypothesis. We have to think before acting. And the way we have to think is to imaging, to imaging based on seizure semiology, video EEG monitoring, MRI, and so on, which could be the epileptogenic zone, which means a region of ectalon set and early seizure spread. And of course, this hypothesis is based on seizure semiology, which is mainly due to the spread of the discharge within the brain. So we are facing with something which could be paradoxical. We are looking at sign and symptoms that are due to the spread of the discharge. And then we have to imagine which and where could be the region that is generating the seizure. Generating all these things. And of course, you have to always bear in mind that there are alternative hypothesis. And of course, if you have this scheme, which is right, and if you put your electrode at the right place, you will win. But unfortunately, there are many situation where your ectalon set and or your early spread is not exactly at the place where you would have think it was. And this is the main problem of SELG, which is how we know when we have put our electrodes in place that indeed we have missed that seizure onset zone. So I will go through four cases that are illustrating four main topic in that field. And then I will give you some perspectives. So the first is of course, that you suspect you have missed your seizure onset because the ectal semiology that you see during your SELG monitoring starts before any SELG change. So this is a first case, a 20 year old left-handed lady. That started having seizures at the age of 13. These are mainly nighttime seizure. And she has no comorbidity, neurological exam is normal, no cognitive deficit. She has a verbal memory, which is a bit better than non-verbal memory. And seizure semiology strongly suggest seizure that could start around insular or insular opercular region, because it starts with an unpleasant sensory sensation of the right arm. Then she has a grimace that could be more marked on the right hemiphase. And she has some bilateral tonic motor signs, more marked on the right than on the left. And she salivates. So this is a semiology, no doubt. It's turning around the insular opercular region. And we had some thoughts that it could start more on the left hemisphere because of the right arm sensory sensation and motor signs that seems to be more right than left. Interictal EEG shows nevertheless bi-temporal spikes, more marked on the left. Ictal EEG could suggest the left temporal central onset, but it was unclear because this is deep seated region. So the discharge was not clearly visible. And this is of course MRI negative, PET negative case that does not help us to decide that. So this is the implantation we did, mainly over the left hemisphere. You can see this large exploration, but with a few sentinel electrodes on the right side, because you never know in a insular opercular seizure, the spread of the discharge is so fast from one side to the other, that it's better to put a few sentinel electrodes in that region. And this is what happened during the seizure. So it clinically stopped there. Here is this grimacing, probably some salivation in the mouth. Here is this tonic posture on the left arm, and then this clear tonic posturing on the right arm with some movements. Now the sticking is bilateral, even more left. Here's some pedaling. You can hear this swallowing, which is very, very typical of that region. And at the end of the seizure, it's clear that the left arm is more tonic than the contralateral one. And now it progressively ends. You see here your discharge. Yes. So she's breathing again. So I think that you will all agree that this is a very typical semiology of insulopercular origin. The main problem is this one. This is the implantation. These are your left electrodes here, or in black, the right one in blue. And you can see that the ACEG onset appears more than 10 seconds after the clinical onset, which is here. And you see here that very nice discharge of spikes that start on the right opercular region. So exactly here. So for sure, it's clear that we missed the side. And the question is what to do in such a situation. So the best way to do is to remove the electrodes and to redo the ACEG again bilateral, but with much more electrodes on the right hemisphere. And this is what we did. Now we have a quite nice sampling over the right because we need to know how much of that cortex is involved at seizure onset. And this is a totally different story. Same seizure, but this time with our electrode at the right place. And the right side is in blue. On the top, the left in black. And you see here the discharge that spread from the central opercular region, the oculopercular region, to the mid part of the insula and to the anterior part. And the posterior insula cortex is totally spared by the discharge. So based on that now, you can plan quietly your surgery. This is what was done. And the patient became seizure free. It was a right anterior insulopercular surgery. And she is a seizure free for more than 24 months. Interestingly, it was an histology negative case. So we will see in the future. Okay, so this is a first really typical situation, not so frequent, but it happens. And you have to recognize it. So the second situation is that when doing your ICG, you record a seizure onset pattern, which is not really convincing. So again, the case, Elise, 70 year old, was started with seizure at the age of 15. Normal neurological exam, no comorbidity, no cognitive deficits. And semiology is quite strange with a cephalic aura. Then you have something which is close to cortical deafness, sometimes tingling on the right or left arm. He may or may not speak. And consciousness is impaired. And then she may or may not have dystonic posture that can be on the right or on the left and plus or minus aversive movement of the head and eyes, which is either right or left. So definitely there are some ambiguities regarding the side of the seizures. And the side is pointing more to the posterior territories. It seems to turn around the temporal pericylgian region with an enterital EEG that show bi-temporal spikes, more on the right than on the left. An ectal EEG that seemed to be more left than right, turning around the pericylgian cortex. And MRI was negative with a PET scan that did not help at all because it was bilateral temporal parietal. So this is the exploration we did. This is almost symmetrical exploration of the posterior temporal pericylgian region. A bit more electrode on the right than on the left. And this is what we got, an SEG onset here, which is not very convincing. You see this activity of spikes which is not really continuous and that grows progressively before going here to those regions. So this activity is located here in the posterior temporal basal cortex, but also in the inferior parietal cortex. And the main spread here is all along the temporal neocortex up to the first temporal convolution, but you see it also very nicely over the insulopercular region in this posterior part. So that's a typical situation where you do not know where the seizure really starts and you cannot perform such an extended resection. This is a situation where definitely you cannot do anything after such a SEG. And I'm not convinced in that case that redoing your SEG will help. So in that situation, we decided not to do anything and to put a VNS. The second situation is this one, Méline, this is a young lady that has started having seizures at 18 months. She had initially at 7 years what was named absences, and then she started having focal to bilateral tonic-clonic seizures at the age of 12. And these were daytime seizures, again normal neurological exam, she's a bit depressed, she has had some psychogenic non-epileptic seizures, she has no cognitive deficit, and the semiology is quite rudimentary, she has no aura, she had a head and eye diversion, and then she went to a focal to bilateral tonic-clonic seizures. In directally anterior left frontal spikes, or B-frontal spikes that predominate over the left, ICTAL-EEG is left, MRI negative, and PEG show hypometabolism which is quite moderate over the left frontotemporal region. So this is a seizure, the onset, it feels probably something, and then look at the head, you have a diversion, a tonic contraction first of the right arm, then bilateral manifestation, and then a tonic-clonic seizure. So all is pointing to the left hemisphere, of course that's not a generalized epilepsy, and this is the SEG, so your frontobasal cortex, frontopolar cortex, dorsolateral frontal cortex, the frontal mesial region here, the opercular region, and the first temporal convolution because of the PET scan, and this is the SEG pattern, quite, quite, quite unusual. Of course you see here that this spike and wave activity, or poly-spike and wave activity that initiates a seizure predominates here over the dorsolateral frontal cortex, but believe me that's quite extended, and the pattern is not a very good pattern that suggests that you are facing with something which is quite focal in the left frontal lobe. So this is the extent of the discharge here, of the frontal lobe, this is a recent case, so we do not have yet decided what to do, but probably we won't go with surgery. So, which are the good patterns of SEG in talonsets? So I would strongly recommend that paper which was done by the Marseille group, where they identified different seizure-onset patterns in drug-resistant epilepsy patients admitted to SEG, and what is striking is that those patterns here that at some point contain fast activity, either low-voltage fast or intermixed with spikes or starting with poly-spike discharge and then low-voltage spikes, all these patterns with fast activities are the most frequent. And more interestingly, they were the patterns that were related to the best seizure outcome in those patients. What does that mean? It means that if you don't get the fast activity when looking at your SEG-onset pattern, then you can suspect that indeed you are not exactly at the right place. Of course, you have to take into account the underlying pathology because the pattern will change depending on the pathology. For instance, the low-voltage fast activity here is much more frequent in polymicrogyra, the fast activity here is more frequent in periventricular nodular heterotopia, this pattern here with poly-spike activity followed by a low-voltage fast activity is quite typical here of focal cortical dysplasia type 2, and so on and so forth. So you have to be aware of the patterns that you are expecting depending on the underlying sub-state. Then, a third situation, which is a bit tricky, is when you see a fast activity that seems to arise from nowhere. I just explained through that case, so a lady who started having seizures at 28 years, daytime and nighttime seizures, no comorbidity, normal neurological exam, some visual and verbal memory difficulties, semiology which is not very informative, no aura, consciousness impairment from the onset with mid-oral alimentary automatisms, and a postictal amnesia. Interictal EEG show right temporal spikes, ictal EEG right temporal, and MRI negative showing left temporal polar hypometabolism. This is the reason why we did an asymmetric bilateral exploration here, and you see that we target mainly the temporal lobe, but also the frontal lobe region, that could explain the fact that the seizure was asymptomatic at onset, and this is what we got. So the right side is in black, the left in blue. You see a very nice discharge here in the anterior part of the frontal lobe, here more frontal polar than frontal basal, but you don't see anything before, no spikes, no slow phase, so you have a strange feeling that you have missed something. So what do we need to do in such a situation? Adding electrodes is probably not the best option because you should add a lot. So the best way to do is to remove the electrodes and to redo ECG. This is what we did. You see that the implantation is quite important, especially in the frontal lobe region. We did not do any left side implantation, and this is a totally different story now. You see this very nice ictal discharge here, which is quite typical on a focal cortical dysplasia with this continuous spiking activity that accelerates, then you have this polyspike discharge and the low voltage fast activity, so it strongly suggests that you are at the right spot, and this is located here in that very small part of the orbitofrontal region that was not explored during the first ECG. And then, knowing this, you can perform a very small resection, now we have a long time follow-up. It was an FCD type 2P and the patient is seizure-free. In that situation, the situation which is not so infrequent, you are close to the seizure onset zone, but not exactly at the right place, and mainly you don't know which are the borders. So again, a case who has started having seizures at the age of 4, mainly night-time seizures, she has a cephalic sensation, grimacing with axial tonic manifestation, a lot of vegetative signs. This could be of insulopercular origin or frontal lobe origin. She has normal neurological exam, she is perfectly normal, cognitively speaking. The EG onset was unclear, with a late ictal right temporal centropietal involvement and the MRI is negative, so she was explored by another group, I show you here what happens during the seizure, so she is awakening, she has this tonic axial contraction, she puts the hands over the face, she is red, she does not breathe, and then she stops. So a semiology which is a bit ambiguous, this could be frontal, but this could be also insulopercular. So she was explored in another center with a subdural implantation, plus death electrode, and they found that indeed this could be of insulopercular origin of the right side, and this is a resection they did, the scar here is observed over the frontal lobe, the dorsolateral frontal convexity due to the grid implantation, and it has absolutely no effect on seizures. Exactly same semiology. Shows here the new evaluation with an EEG that show rare right temporal spikes, an ictal EEG onset that was right frontotemporal, MRI shows insulopercular resection with the dorsolateral premotor scar and PET was not relevant. So this is the SEG that was done in our lab, and this is what we got. So here this is the mesial wall on the top, the lateral convexity, then the opercular region and the temporal lobe here. And this discharge here on J2, M2 are located here on these two electrodes that are the most posterior one on the mesial aspect of the frontal lobe. So what to do in such a situation? So what you can do is to add electrodes during monitoring, and this is certainly the best indication such a situation, which is to expand coverage to better define the resective surgery boundaries. So this is what we did. We had, you see, six electrodes to better define which is the seizure onset region, and this is what we got. So of course, the seizure is involving the territory here that were first recorded, and in red here, you see that there are additional electrodes that are involved at seizure onset, and you can rule out the fact that the most posterior one could be involved. And of course, it changes strategy because you can decide in that situation to perform a resection accordingly with a very good seizure outcome. And of course, if you did not have had the electrode, you would have not been able to decide the extent of your resection. So basically, these are the different situations I wanted to go through. Can we do better in the future? Yes, and especially using signal analysis techniques. I won't go too much into deep in those studies, but indeed, what you can do is to try to classify the HICTL signal using a wavelength-based statistical feature. It can help to disentangle what could be the true onset and what is a spread. You can use a different inter-HICTL-epileptic biomarker and couple them to evaluate the adequacy of your SEG sampling. And this is a very important point to be sure that indeed your electrodes are in the right place or not. These are recent studies that you can have a look on. And you can use tractography, MRI tractography, coupled with SEG features to try to distinguish the seizure onset from the early spread and late spread, and this is interesting. And there are some studies that are in progress in France regarding the so-called virtual epileptic brain modeling to try to predict surgical outcome. So these are just some insights on what could be the future of that topic. But of course, there is no proof at that time that these computers are better than our eyes to decide. So I will stop there. I think we are just on time. And Jay, I don't know what you think, but we can go through a few cases and to ask to the audience what they think about those cases. Sure. Let's do that. Okay. So now this is your turn to work. And we start with the first case. So Johanna, 20 years old, she's left-handed, but she has an fMRI for language which shows a left hemispheric predominance for language. She has started at the age of 14 years. She has daytime seizures. And she has a very discrete right hemiphase paresis. She has no comorbidity. She has both verbal and nonverbal memory difficulties. And semiology is a bit tricky with an epigastric sensation, but sometimes a gustatory feeling. Then she is losing awareness. She is immobile. She shoos a bit with a nose rubbing, which is unlateralized, and she remained confused. So interictal EEG show left spikes, more on the mid-posterior temporal region. Rectal EEG seems to be temporal, a bit more posterior than anterior. And on MRI, she has a left thalamic infarct, that probably explains the right hemiphase paresis, and a PET that shows left temporal entero-mesial hypometabolism. So that's a situation where you could buy a left temporal lobe onset of the seizure. Actually, the big puzzling is the gustatory aura. And what we found in video EEG monitoring was a very long latency before the first EEG changes and the occurrence of the first symptoms. So it means that, indeed, this could be a temporal lobe case, and this is a symptomatogenic area here in blue. But there are different spots in yellow that could be a potential trigger of that semiology. So are we facing with a so-called pseudo-temporal lobe epilepsy case? So our favorite hypothesis was, of course, the insular cortex. But we pay attention to the posterior cortex, especially because the posterior involvement of the temporal territory is inter-ectally and ectally. So this is the implantation, quite large, with a good covering of many regions. And this is what we found. So this is the orbitofrontal region, the insulopercular cortex, mesiotemporal lobe region, temporal lobe cortex, and here, that line, the posterior cingulate gyrus. So you have a very nice discharge in the mesiotemporal lobe region, but of course it is preceded by something which is early in the posterior cingulate cortex. This is a nice pattern with spiking activity followed by low voltage first, and this is localized here, in that very small piece of the brain. So these are my questions. What do you propose? Do we have to add electrodes? Do we have to perform radiofrequency thermocoagulation? Do we need to perform a second SEG? Do we go straight with resective surgery, and is it better not to do anything? And, Ronnie, if you can show them the polling results on the screen? Yes. Yeah, I show it in the chart. So 45% of you propose to add electrodes, 30% to perform thermocoagulation, 12% suggest to go with a second SEG, 9% to go with resection, and only 3% no surgery. So that's very interesting, because probably this is what I would suggest right now, but that case is a very old case. You're right. This is a typical situation where you should put additional electrode in the posterior cingulate region and the surrounding cortex, because you will be sure that, indeed, it starts from there. Doing your RFTC, thermocoagulation, is a good idea, because, indeed, before removing your electrode, you can do that, and if your patient is seizure-free, that's fine. If he's not, it probably suggests that you will have to go with a second SEG. So my favorite would be to add electrodes, but this is not what we did. So what we did is this. So can I move on on my slide? Okay. So we did a small resection, and you see, it was not very easy to do for the surgeon, because he has to remove only this very, very small part of the cortex, and probably we were very lucky, because the patient became seizure-free, and the histology was not very relevant, but it's a so small resection, and interestingly, the patient improved in terms of nonverbal memory. So that's a tricky situation, but again, I would not recommend what we did. Most of you, and probably because of what I said before, suggest to add electrodes, and it was definitely the best option. So we can move on. Ah, we can answer to other Q&A in the chat, yeah? Yes, yes, Dr. Kahn. We'll take a pause here. I'll ask all the audience, please feel free to put your questions in Q&A section, and I'll be happy to post them here. Few questions for you, Dr. Kahn. My two cents is exactly the same, that adding more electrodes in that area would have been a more sense, and you've been lucky to call that surgery, and it also needs the same point that you highlighted, seeing the low-voltage sharps, which was not present in this SEG. You evidently saw the low-voltage, but without any preceding sharp waves. In that case, you mean? In this case, yes. Did you see any, was there any sharp waves? No, no. You have very nice spikes here. Perfect. Perfect. That were continuous all along the SEG recording. Perfect. That's great. So, when you saw this pattern, adding more electrodes is always a luxury, and then the second option, as you suggested, RFA could be one option, but you opted to go for surgery. And any other suggestion or comments, how did you decide on the surgical resection boundaries of this case? Yeah, we decided to do something small to see what happens, and if we did not succeed, then probably we would have done a second SEG. But again, we probably were very lucky, and we should have done first radiofrequency thermocoagulation before removing the electrode, because it helps a lot to decide. And I see a few questions coming up in Q&A, going back on some of the comments that you have made on the previous slide, and this is regarding the planning of the electrode trajectory. Do you have any comments, and how do you, at your center, how do you sample insula, insula opercular region? You prefer more oblique versus orthogonal, one electrode versus two electrode coverage? Okay, that's a good point. So the coverage of insula opercular cortex has to cover not only the insula cortex, but also the opercular region. So do not use either oblique or lateral orthogonal electrode, use both. So typically, if you're quite sure that you're facing with an insula opercular case, you should have, I would say, seven electrodes in that region, which means two obliques, one anterior in the short gyri, one posterior in the long gyri, and three opercular suprasylvian, frontal, central, parietal, and two in the temporal neocortex, anterior, posterior T1. I would suggest, if you're convinced this is insula opercular, to add electrode on the contralateral side, because it goes so fast from one side to the other that you can miss really the side. That takes me to the last question relevant to one of your presentation, insula opercular case, where contralateral side was onset. Is that a standard practice at your center to explore bilateral insula? Yeah, almost always, because we have a few cases where we missed the side. Sometimes PET scans can be mislateralizing, and definitely now, again, when we are sure we are facing with an insula or insula opercular case, we put at least one additional electrode on the contralateral insula cortex, at least one. Fine, okay, well, we can move on to the next case. Okay, so a second case, Kenzo, 12 years old, and he has an epilepsy onset at 9. She is seizing more during the night time than during daytime. Normal neurological exam, no comorbidity, no cognitive deficit, and seizure semiology shows polymorphic aura that can be visual or auditory or right sensory and painful, and involving the whole body. It can be a mixture of all these signs. Sometimes it's more visual, sometimes more auditory, sometimes more sensory, and of course, all these signs are pointing to the posterior temporal pietoccipital region. Then after this aura, the consciousness is impaired. There are some oral alimentary automatisms and plus or minus a version of the head and eyes that seems to be more right-sided. He has also rare focal to bilateral tonic-clonic seizures that were unfortunately not documented during the video EEG monitoring. So during that monitoring, the spikes were definitely left on the left posterior temporal region. Ictal EEG was again temporal, but more mid-temporal on the left side, and with a spread of the discharge posteriorly, MRI was negative. And the PET scan shows a widely extended, though moderate, temporal pietoccipital hypometabolism. So it explained why we did this implantation that was definitely centered over the posterior temporal and temporal insulopercular region. And this is a seizure that we recorded, which is one of the so-called focal to bilateral tonic-clonic seizures. So this is what happens. The kid is opening the eyes. The head turned on the left, the eyes on the right, and you have this contraction of the left hemisphere. And then we got the tonic-clonic generalization. So what happens on SEG? So this is a clinical onset in blue with the eye opening, the head that turned on the left while the eyes are going to the right, and the left face contraction. This is the SEG onset. Here, I let you see this pattern. I don't know if you consider it convincing or not. And these are my questions. Is this SEG onset pattern convincing? Does the semiology start before the SEG onset? Is the semiology that you saw congruent or more discongruent with the SEG data? Do you propose a limited resection of the first temporal convolution, T1, where you saw this change at SEG onset? Or do you think that you need to extend your resection to the first temporal convolution plus the insular cortex? So I can show you again just the onset of that discharge. This is first temporal convolution. And this is here, the insulopercular cortex. OK. And this is the pattern. So you can use the poll now. So here we are. So you are quite good. So 67% of you think that, indeed, the SEG onset pattern was not convincing. You're right. We will go back on that. Semiology, of course, did not start before the first SEG change. So there are a few of you that make mistakes. Excellent. Because all of you saw that, indeed, the semiology was discordant with SEG data. 55% of you considered this. And only a very few, 9% each, considered that you can go through with surgery. So very good. Very good. And so this is what you should have considered. This is a left-side implantation. The SEG onset pattern, as you can see here, is really not convincing. OK. And of course, the ictal semiology, especially because of the left face contraction, strongly suggests a right onset. It explains that we missed the side. I mean, what you see here at seizure onset is probably just a reflection of something which is happening on the contralateral hemisphere. So in that situation, we considered that we do not want to go through a second SEG, and we propose first a VNS. And it was a good option because the patient is right now almost seizure-free with VNS. OK. We can move on to the last case. A few questions, Dr. Kahan. If you stay on the previous case. OK. Yes. I see some questions coming up in Q&A also. So my first question, which is more than a question, is a comment also for all the audience on call, is this case truly highlights, as I was discussing with you, Dr. Kahan, highlights the importance of recording all types of seizure semiology before we implant intracranially. And despite our suspicion might be single focus or single hypothesis, but still it becomes extremely relevant if the patient has two semiology is to capture both because sometimes we do get surprised patients may have outside the focus. Want to hear some of your thoughts on how often does it happen and how that error can be avoided during their pre-surgical evaluation? That's not a very frequent situation, fortunately, but you have seen here two cases I show you where we miss the side. So in that case, it was very, very clear cut. The first one, which was insulopercular in my presentation was a bit more tricky, but it can happen. And your recommendation is extremely good, Jay. I mean, please record all seizure types before going with SEG because sometimes you have very, very tricky things. Thanks, Dr. Kahan. Would you have considered, an open-ended question, would you have considered a repeat SEG on her? Yeah, I mean, that's a good point. I was not convinced by the shape of the discharge, even on the right, it seems to turn around the insulopercular cortex, widely extended over the posterior territories. These are coming from the right side, so we can go through. But I mean, in that situation, we are not quite happy to go with a second SEG. So it could be done, but we prefer first to use a VNS and the kid is doing well. So now we do not have any argument for going with a second SEG. But it could have been done, yeah. That's reasonable. We can move on to our last case. Okay, so the last case is Guillaume, 30 years. He has started having seizures at the age of four, night time, more the daytime seizure. These are seizures with an epigastric sensation, a fearful epigastric sensation, followed by a hyperkinetic behavior, normal neurological exam, verbal memory deficit, EEG that shows left temporal spikes more than on the right side, and video EEG monitoring that strongly suggests seizure coming from the left frontotemporal region. MRI is negative and the PET show right thalamic hypometabolism that we did not consider for our case. So these are the seizures that he had. So he stopped here, something in the throat. and then he has the hypertensive behavior and then it progressively stops and he will have a long-lasting aphasia after the serum set so something which points to the left side of course because of EEG data and that could suggest seizures that come in between the temporal lobe anterior part and the frontal orbital region so something which is in between those two regions. So this guy had the first SEG it was 10 years ago a bit more and we demonstrated so it was an SEG centered over the left frontal temporal lobe region with a few additional electrodes in a range here on the right and we demonstrated that the seizure was starting from the left mesiotemporal lobe region and we performed the left anterior temporal disconnections so we do not have any histology and he stayed seizure free for almost five years and then he relapsed, same seizure, same frequency as before. So we did a new evaluation based on what we got on phase one with anterictal and dictal EEG that always pointed to the left frontal temporal lobe region of course the MRI show only the left temporal disconnection that you can see here and the PET of course was not clearly informative because the patient was already disconnected. So this is a SEG centered over the left frontal orbital region the left temporal lobe a few electrodes in the insular cortex and a few additional electrodes on the right side and this is a seizure onset pattern so it starts clinically here which is written clinical onset with the fiora and then the motor manifestation and you see here just before this pattern here that seems to involve first here the prefrontal ventrolateral cortex so more lateral than mesial then what do you propose? Are you convinced that this pattern do you think we have enough or do we need to add electrodes? Can we perform a radial frequency thermocoagulation because we have a very small spot in the brain? Do we perform a third SEG? Do we go with rejective surgery or do we decide not to do anything? Okay so here are your recommendations so 63 percent of you recommend to add electrodes probably in the insular cortex and in the surrounding region 25 recommend radial frequency thermocoagulation only three percent renew surgery none of you recommend surgery that's good and only nine percent recommend no surgery so that's interesting that's interesting because this is not at all what we did and I will explain why so if we go back to that pattern here for sure it's not a convincing pattern but don't forget that this is a patient that have been already operated on so it can change a bit the shape of the discharge and I mean it's totally coherent with the semiology because it is a frontotemporal semiology and only the frontobasal cortex probably remains epileptogenic and you have indeed that fast activity followed by a low voltage fast here that it's to me relatively convincing to propose to propose a resection in that situation a thermocoagulation won't work because you have only a small sampling of the frontobasal cortex and probably you won't be able to interrupt the seizure activity just doing these small coagulations so this is what we did we perform frontobasal resection here and yeah the patient is not seizure free he is angled two way but he's doing quite well I saw him today so that's very funny and he continues to go well and it was an fcd1a that probably explain this largely extended epileptogenic network so if we are now to go more in deep in that case probably you will explore a bit more the insular cortex so that's it and I let you concluding Jay except if there are additional questions yes unfortunately we are running out of time Dr. Kahan so I'm going to hold up on any further questions I'm just going to make some closing remarks and on behalf of AES I would like to thank you for your participation in today's event and a recording of this webinar will be sent to you within 7 to 10 days please be sure to complete the webinars evaluation in order to claim the CME credit and this concludes today's presentation I'll hand over to our operator thank you so much Jay thank you Dr. Kahan thank you everyone this concludes today's stream thank you and have a great day

SEEG

seizure onset zone

epilepsy

pre-surgical evaluation

electrode coverage

case studies

diagnostic techniques

radiofrequency thermocoagulation

patient outcomes