Despite numerous new drugs developed over the last 30 years, medications alone aren’t enough to control seizures in one-third of patients with epilepsy. Fortunately, explains pediatric neurologist Roxanne Simmons, MD, advanced surgical techniques and implantable devices, as well as recently approved drugs, mean that children with refractory epilepsy can have a better quality of life than ever before. This illuminating presentation includes images to clarify the placement and function of various neuromodulation devices as well as help with answering patients’ questions, such as whether to try CBD or follow a ketogenic diet.
So, thanks for letting me speak today. I thought I would talk about um a topic that is near and dear to my heart as an epileptologist, so treatment of medication resistant epilepsy. And my goal is to not, of course, make all of you become epileptologists or learn how to manage these specific patients, but we do share a lot of many mutual patients, and so I'd like you to be aware of what's going on behind the scenes from Our epilepsy clinic and to be able to recognize when some of these initial patients are receiving certain therapies or treatments, why they're receiving them and how those actually work. So, I hope you find this valuable and interesting. And it's an ever evolving field too, so I know that many of you probably learned a little bit about this in training, but there has been so much that continues to evolve in the field of um uh treatment of epilepsy, uh, including epilepsy surgery, so I'd love for you guys to um be aware of these new and recent updates. So I have no disclosures. And in terms of our objectives for this talk, so first of all, I'd like to just basically identify which patients have medication-resistant epilepsy, and we'll start off by even defining what epilepsy is because that is also an evolving definition. Um, then we'll switch gears and talk about different surgical options and devices for treatment of medication-resistant epilepsy, and then we'll go into a couple of new medications with newer mechanisms of actions and also dietary. Um, therapies for treatment of medication resistant epilepsy, and some of these may sound familiar too if you have mutual patients, but I hope that I can flush these out for you. So, first of all, what is epilepsy? So, you know, you might see, I'm sure you guys see lots of, you know, first-time febrile seizures in young infants or unprovoked seizures in children, and the definition has shifted over the years, but according to the most recent um ILE official report that was produced back in 2014, this is the International League Against Epilepsy, which helps us a lot of our standardized definitions for patients. Um, epilepsy is defined as a disease of the brain by basically two criteria. So a patient has had at least 2 unprovoked seizures that are occurring more than 24 hours apart. So, you know, let's say if a patient presents for a first time unprovoked seizure, we usually say like, well, it's not epilepsy at this point, so let's continue to watch closely and maybe do some other ancillary studies or other workup to really think about the risk of epilepsy. Or if a patient has one unprovoked seizure, plus a greater than 60% risk of having another seizure occurring over the next 10 years, and this is a tricky kind of flexible definition that allows basically a neurologist and epileptologist to make this diagnosis um earlier. So, you know, we might say like, well, what is a, who determines what a 60% risk is? And so we base off of various things, but it could be, you know, the patient has an abnormal EEG background. Um, or they have abnormal imaging, they have something structural in imaging that could explain their seizures, um, or they fall into the bucket of this. Number 3 point here, they have an actual epilepsy syndrome. So, we see, oh, this fits with, um, you know, at, at early infantile onset epilepsy or an early childhood epilepsy. And that way, we're not waiting for subsequent seizures to actually begin our treatment regimen for these patients. And then when someone is diagnosed with epilepsy, think about, well, what, what constitutes medication resistant epilepsy? And this is referred to as different names. Sometimes you might hear it referred to as medication refractory epilepsy, drug resistant epilepsy if you're looking through various studies, but the basic definition is a patient with epilepsy who has failed at least 2 or more anti-seizure medications, and we abbreviate those as ASMs. And this means they failed them, um, not because of specifically side effects, but when they're using adequate doses for the right seizure type and still not achieving seizure freedom. And what I'd love to highlight with this too is that, you know, 1 in 3 people with epilepsy have medication resistant epilepsy. It's a huge percentage of the epilepsy population. And so I think about, you know, all the medicines that have been produced to try to treat seizures, um, we realized that we're, um, you know, still at a loss to treat a lot of these patients. So I love this slide. This is from a study that came out, gosh, I can't remember the name of this. It's probably over 1015 years ago at this point, um, but I think it really illustrates the need for other treatments aside from anti-seizure medications for this unique population of patients who have medication resistant epilepsy. So, despite all of these new anti-seizure medications, still after all of these years, a third of our population remain uncontrolled. So you can see here on the X-axis they have the year and on the Y-axis, the number of anti-seizure medications that were produced. And you can see, you know, back way, way, way back, you know, potassium bromide was one of the first drugs and then the big thing was in, you know, early 1900s, they came out phenobarbital, and that was a medication that was helping with seizures. Um and then they started to pick up around the 1950s and beyond developing a lot of new types of drugs and you can see here that basically since the 1990s and 2000s, there's been an explosion of tons of different new anti-seizure medications that um have somewhat different mechanisms of action. Um, but basically haven't moved the needle in terms of decreasing that 1/3 of patients that have medication resistant epilepsy. So this was like a shock to me when I learned this, so I was like, wow, these new medicines. And still, you know, 1/3 of our patients remain uncontrolled. Um, and one thing that I like to emphasize too is a lot of people have this thought that, oh, you know, some of these older medications like phenytoin and phenobarbital are much stronger, and those will do really well at controlling different types of seizures and even if you survey different neurologists and epileptologists, everyone has their preference what they consider to be a strong medication or not, but in reality, when you look at uh broad population use, basically all these medications, aside from Um, some exceptions for certain patient populations have really the same. Efficacy, which is always really astounding to me. What we really see benefit from these 3rd generation anti-seizure medications, the newer ones, is um better ease of administration and a decrease in side effects. So I think that's important to realize too. So now I'd like to shift gears a little bit now that we talked about, you know, what is epilepsy, what is medication resistant epilepsy, and think about um when we find these patients who aren't responding to medications, what are other treatments we can offer to them, specifically surgical options or devices. And in terms of epilepsy surgery, this is kind of a, a boiled down simplistic view of it, but the way that I take it, that I tend to think about different types of epilepsy surgery, there is resection, which actually means going in and taking out a part of the brain tissue that could be causing seizures. There's what we call intracranial EEG, so for the patients who need more of a surgical workup to think about what parts of the brain can be involved with producing seizures, we can actually monitor intracrannially to capture seizures. Um, there is hemispherotomy, hemispherectomy, we're actually again disconnecting a very large portion or multiple lobes of the brain. There's corpus callisotomy, we're disconnecting the corpus callosum, and then there's laser ablation, which is actually ablating with heat, very small areas of brain tissue. So in terms of resective epilepsy surgery, um, we do something called electrocorticography, which is really interesting, I think. So what we do is, let's say a patient has, you know, a tumor, like this is, I think the example I gave here was a patient who had a tumor that was resected. Um, and so what we do is we go in the OR with our neurosurgery colleagues and they actually remove part of the skull and expose the, uh, cortical surface of the brain tissue with the dura, they peel back the dura there. And what we do is we take electrodes, you know, just kind of how we would do a scalp EEG, but we put those direct electrodes directly onto the cortical tissue itself. And what we do is we listen. So we listen for spikes, you know, same sort of thing that we would do with scalp EEG looks different, of course, because this is a, you know, monitoring directly from the, the cortical surface itself, um, but we use this to help us think about, you know, what are the margins of resection. So, if we have, you know, a tumor that's encompassing, you know, one part of the brain that's close to the surface, um, we can move those little electrodes around to think about, you know, what part of the brain tissue is active and producing spikes and which, which parts are more quiet, that can be helpful for finding. In these tumor specific cases, um, margins of resection, so, you know, it's not always visible when you look at the brain tissue itself, where the tumor, where the tumor margins, and this type of procedure can help us understand that. Um, and there have been some studies looking at the utility of, of ECOG guided resections, and this can help with seizure freedom outcomes, basically because you're not missing tissue that might be involved, that's not, you know, visually looking different on visual inspection. Now, with intracranial EEG, so I always think about this, um, and I'll give you the example of, you know, we monitor patients in what's called the epilepsy monitoring unit, which is they're admitted to the hospital, they have scalp electrodes placed, we might, you know, reduce their seizure medications to provoke typical seizures, capture them and see, you know, well, it looks like it's coming from the left side of the brain. Maybe from the back part, but I really can't be more specific than that. And this can also be used for patients who, you know, obviously have focal onset seizures coming from, we suspect, based on our hypothesis, one region of the brain, but they may not have other concordant imaging findings like they may not, you may not see anything on um a high-quality MRI brain or a PET scan. And so we're not exactly sure where the seizures are coming from. What we often do is what's called intracranially gene monitoring. So that's to really hone down and look at, you know, what, where do we think the seizures are starting from, um, and to really help us inform our resective surgical plan. So, this is not, I always call, tell patients this is like a mini surgery before the actual resective surgery to think about um what are the margins of resection and where the seizures are actually originating from. Um, and what can be helpful with this is going back to scalp EEG, you really just get like a 2D representation. We're looking at things that are close to the cortical surface, but for patients who have seizures that are originating in really deep parts of the brain and Everything arises generally from the cortex, but as you can imagine, there are a lot of enfoldings and gyri, and you think about, you know, inside the insular regions and the temporal lobes and the occipital regions where you have a lot of depth to there too. You lack that 3D representation about where the seizures are starting from where they're spreading to. So, oftentimes we might see just on scalp EEG is basically the ripple effect, the sort of propagation effect of a seizure, and it may not be sufficient for understanding exactly what the origin of the seizure is. And so with this intracranial monitoring, we can get a 3D representation, we can get a better sense of where the seizures are coming from and where they're spreading to. So, I'll give you the example here of um what's called stereotactic EEG. Um, so this is actually a case that we did at Oakland for um a young girl who was having focal onset seizures originating from her right occipital lobe. What we did is based on our hypothesis, we inserted um 5 different stereoelectrodes. This is kind of a, a, a visual interpretation of what it looks like from the skull and then what it looks like on a medial section of her brain. Um, and then once those are placed, we actually confirmed them with CT to make sure the, the electrodes are in their proper place. And this is used again to capture both the 3D representation, the origin and the spreading effect of the seizures. In addition to the stereotactic EEG, we can also do just like I showed you on the past slide for the electrocorticography, we can insert grids and strips, electrodes that are also placed on the cortical surface when a part of the skull is temporarily removed. So with those combination, we can get a better sense of um where the seizures are originating from. And usually these patients are admitted to the ICU after the procedure, and then we kind of treat it just like we would a normal admission to the epilepsy monitoring unit. So we might reduce or hold anti-seizure medications to provoke typical episodes, and we usually like to see, you know, 2 to 3 typical seizures on EEG to see what they look like, um, and then the patients are usually explanted. We then send them home, you know, they recover from their, their small surgery, we think about what we want to do and then we bring them back eventually, ideally for resective surgery if we, if we find them to be candidates based on this procedure. So moving on to different forms of epilepsy surgery, so, um, I'll talk briefly about hemispherotomy, hemispherectomy, and so this is basically a disconnection or removal of an entire hemisphere or multiple lobes. And this is not done commonly, but there are a couple um unique situations in which we do this type of procedure. Um, the biggest one indication or the, the, some of the main indications here I would say is, um, what's called Rasmussen syndrome. This is a chronic focal encephalitis where a patient at a very young age can have progressive atrophy due to this, uh, inflammatory process, and really the only cure, um, to alleviate the seizures and prevent spreading to the contralateral hemisphere is actually removing the hemispheric origin. And so this is an example of a, um, not a case that we did, but a 7-year-old girl who had a hemispherectomy at age 3 years. Um, because of Rasmussen syndrome, and um she had this hemispherectomy and actually, uh, these children, if the surgery is done before age 5 years, do very well. So they're still, they have some hemiplegia, but they can usually ambulate, um, and this patient was actually fully fluent, I think in Turkish and Dutch, um, and did have a little bit of spasticity of her left arm and leg, but otherwise was, you know, leading a very normal productive life. So, Um, the key here is making sure that these patients are diagnosed and treated early, ideally before age 5 years, where they can recover better from this type of procedure. The other example is hemomegaloencephaly, so this is a, um, usually the type of patient that presents early on in infancy or very early childhood who has intractable seizures and has actually overgrowth or a really significant, um, architectural distortion of one of the hemispheres and same sort of thing where if they have this procedure. Um, it can basically cure their epilepsy, and they can be very functional and lead nearly normal lives afterwards. So moving on to different types of procedures, so I wanted to highlight here what's called corpus callisotomy, and so this is technically not a curative procedure like the hemispherectomy hemispherotomy or resective um surgery, but it is considered a palliative surgery. And so, you might be familiar with these if we share some mutual patients who have um really severe epilepsy or specifically atonic or drop seizures, um, and that's what this procedure tends to work best for. So, the atonic or drop seizures, the kids who, you know, have to wear a helmet because they're falling down constantly. throughout the day and you know, injuring their face, um, you know, hurting their teeth, um, doing things like that. And so we think about it typically for atonic, um, seizures, but we can see this for a tonic seizures, you know, myoclonic seizures, spasms. So, basically think about any brief seizures that are causing someone to collapse. Um, and then of course, when we see patients with refractory infantile spasms as like a, you know, 3rd, 4th line agent, we sometimes think about corpus callisotomy too. And in terms of outcomes of this procedure, so about 90% of patients do experience a reduction in drop attacks for those who have what's called a total corpus callisotomy. So, um, basically disconnecting the entire white matter bundle. Uh, patients can also get a partial, we call it a 2/3 corpus callisotomy, the idea being that, um, those patients maybe have, uh, less risk of developing complications after their procedure. Um, and speaking of complications, the most common ones we see are what's called disconnection syndrome. So you guys may have learned about this a long time ago in medical school or training where a patient who has this, you know, white matter bundle disconnected, which helps both of the hemispheres communicate, might have when they're trying to coordinate tasks between their left and right hands, may have their left hand doing something completely different than what they expect it to in the right hand, like not coordinating with those two. Um, and then of course with any type of surgery, if it's open skull surgery, you can have complications such as hydrocephalus or infection, but now they actually have um a newer technique too that just involves, uh, basically a laser to sever that corpus callosum, so there's not actually any, um, you know, open, opening of the skull itself. And then lastly, I'll highlight here laser ablation. So this is also known as um laser interstitial thermal therapy or LIT, and a lot of patients like it because it is technically minimally invasive. There's no craniotomy involved, so no skull opening. Um, it can be really good at treating very small and well-defined regions. So the indications I think about mainly for this are things like mesotemporal sclerosis or cavernous malformations, very, very small tumors, and then in some cases, um, people are looking at the utility of laser ablation for tumors, for tuber sclerosis. And basically what our, what our neurosurgery colleagues do is once we've identified an area, um, what they do is they bring the patient to the MRI scanner and in real time, they insert. They do a very small burr hole, so make a small, small hole in the skull, insert their, looks like a stereo electrode basically, and then with MRI guided help they actually heat up the tip of that, which then burns the tissue that's surrounding that, but it's very, very small. It's kind of like a core of several millimeters. And so you can imagine again it works well for those really small lesions but doesn't do as well for patients who have large areas of dysplasia or need to have, you know, an entire lobe removed. It's not appropriate for those cases. And sometimes we actually do this first, um, and patients you're like, well, you know, the, the seizures are starting in a very eloquent area, worried about taking things out. So, let's start with this almost with the expectation that it's a palliative therapy and then if it doesn't work, it doesn't preclude someone from going to resective surgery afterwards. All right. So now we'll shift gears and focus on some devices that are used for patients with medication refractory epilepsy. And you may have seen some patients with this, you may have heard your patients ask about these things, but I wanted to make sure that you guys are aware of um sort of the classic neuromodulation devices that we use in our patient population, plus a lot of uh wearable seizure detection devices that I always get questions about and I'm sure you guys are getting them about too as well. So, we'll start off what's called VNS or vagal nerve stimulation first. This is one of the oldest devices that's been around for, gosh, I wanna say since the 70s or 80s. Um it's an implanted device that actually stimulates the vagus nerve. Um and here's a schematic diagram to showing what it is. So, the patients have um a surgery that is still technically done by a neurosurgeon, but it's almost like an outpatient procedure. They usually go home the same day. And what they do is they have a small battery pack implanted underneath the left chest with a small incision there, and then they have a wire that is sort of snaked up here and coiled around the vagus nerve. So they've got two little small incisions. Um, again, in terms of neurosurgeries, this is the easiest one because there's not actually any Intracranial intervention involved with this. Um, and then I tend to think of uh VNS as a palliative intervention. So, I think it works, it can be a good option for patients with, you know, Lennox-Gasteau syndrome, for example, patients who don't have uh an area of brain tissue that can be resected, um, and it has, you know, very minimal to little side effects. So, in terms of outcomes of this, the typical numbers that we quote is about 50% of patients experience a 50% or more reduction of seizure burden in one year. And then at 5 years, about most patients experience about a 70% reduction in seizure burden. And this again, uh, as very commonly with other devices, takes time to see effect. So, basically, you're implanting the device and then over the course of several months, you're gradually increasing the settings, which is why this isn't a great option for a quick fix, but can be really helpful in terms of over time, um, reducing a patient's seizure burden. And then like I mentioned, the side effects are usually very minimal, and you can change or adjust the settings to help mitigate some of these, but potential side effects can be dysphonia, so change in voice, dyspnea, so difficulty breathing, coughing, or then some patients complain of a little bit of like oropharyngeal or or throat pain when they actually feel the, the VNS device stimulating. Um, and so you might see these patients come in, they have, you know, the, the VNS device you may not be able to see, but they might have a magnet. So with a lot of our seizure action plans for patients who have this VNS, um, there is a magnet that can be swiped over the battery itself that can trigger temporarily higher settings and attempt to abort a seizure. Great. And then in terms of newer devices, so, sort of newer one that's been around for about 1015 years at this point, it's called RNS or responsive neural stimulation. Um, this is a much more exciting device for a lot of people because unlike the, the VNS, it's not a, the VNS is kind of obviously it's a dumb device. It doesn't really give you any sort of feedback, um. It, you can't really do much in terms of the settings once you've reached your goal dose, but RNS is very versatile, so it is an implanted intracranial device, and then you see this, the schematic diagram here. So patients actually have um a, a battery pack that's placed underneath the skull, so our neurosurgery colleagues will carve out a very small portion of the skull underneath, put the battery in, and then based on our pre-surgical workup or surgical workup doing. Those prior procedures, like I mentioned, the intracranial evaluation, once we've really localized the area of tissue and we say, oh, I think it's coming from you, in this case, this patient has left temporal onset epilepsy, like I'm going to use this RNS device to try to again reduce seizures. Um, it can be very helpful for patients who are not receptive candidates but might have focal onset epilepsy. So the example I'll give is someone who has um Uh, seizure onset in eloquent areas. So if someone is left language dominant, you may not be able to remove their left temporal lobe because that could be the, um, you know, cause them to become aphasic. So in these cases, people might opt to do this RNS where the tissue stays in place, but you're using this device to stimulate the tissue with the goal of, um, aborting and preventing seizures from happening. And so the RNS device, um, again is a smarter device which both detects and responds, so it's a closed feedback loop, um. And then in terms of trials, um, there have been some papers that came out a couple of years ago looking at long-term use of RNS in patients with epilepsy, and the median reduction in seizure frequency up to 9 years was 75%, so it's actually pretty good. And then we had these super responders for about a third of patients responded with a 90% reduction of seizure frequency. So again, we think that in general RNS is considered a palliative device, so it's not always leading to seizure freedom, but you can imagine for patients who, you know, don't want to pursue resective epilepsy surgery or have seizures originating from really eloquent areas of cortex that simply can't be removed due to uh the, the comorbidities that would inflict on a patient. This can be a really good device and it's still a, um, it's still a neurosurgery, but in terms of risks, you know, the, the risks are relatively low, of course, you always have the risks of infection or bleeding when you're doing any intracranial evaluation. Um, but I think it can be very helpful for the right patient population. I think our biggest challenge is, while VNS is very easy to get for all ages, RNS, and I'll talk about DBS in a moment too, are very difficult to get approved in children. So, and that's just basically because FDA hasn't technically approved it for children. Um, and, you know, hopefully there'll be some more studies in the future that can demonstrate its use. We still use this off-label, um, but there are certainly limitations with insurance payments for placing these devices in children. So, another new device that's come out within the past 10 years or so, it's called deep brain stimulation, um, and I should say, I should correct that that it's been used actually for quite some time in patients with movement disorders. You might have heard this being used in patients with Parkinsonism, for example, um, but it's use in epilepsy has emerged more recently. Um, and what it is is kind of similar to the RNS device. Um, there's a battery pack that in this case is actually inserted in the chest, and then there are two electrodes that are actually targeting, um, not the cortical tissue, but actually the, um, the deeper structures, so usually nuclei of the thalamus that are being stimulated, um, and there was a trial done called the Sante trial, which stands for stimulation the anterior nucleus of the thalamus and epilepsy. Um, and then during longer follow-up periods, it was shown that there was, there were lasting reductions in seizure frequency and also improvements in quality of life. Um, again, the biggest challenge that I find with this device is that it simply just won't get approved in children. Um, I've had a couple of patients now that we've had denials for and tried appealing. Um, but I think this can be a good option, and while it's still being studied for its long-term use, kind of like the NS RNS, it can show pretty significant reduction in terms of seizure burden over the course of several years. We don't know again exactly how it works, but the idea is that maybe with pulsating the thalamus, you have some relay systems between the cortex and the thalamus, and that sort of setting that frequency can kind of help reduce the seizure frequency as well. Um, and same thing with VNS too, we don't understand the mechanism of it, but I like the analogy that one of my neurosurgery colleagues gave that it kind of sets a beat for the brain to go to, and so for those patients who are constantly having a high burden of spikes in the background, it allows them something to synchronize to where things don't sort of go out of line to cause seizures. Anyway, that's, that's a hand wavy explanation of how it works. And then in terms of devices, so I have a lot of patients ask about these. So the biggest thing are the seizure detection watches. Um, the one of the first devices that came out was by a company called Empatica, and they developed the Embrace watch. I think Apple also has a, um, an app now too that can be purchased. Um, the Embrace watch is actually an FDA approved device for both, um, children and adults, and, um, it has some proprietary system that uses skin conductance, temperature, motion intensity, and blood changes. Um, and what I find about these is that the devices themselves can be expensive. So the example for the, the Embrace watch, um, it costs about $250 and then there's a monthly subscription fee of about $20 to $25. Um, same thing with the Apple Watch, obviously you have to purchase an Apple Watch itself and then pay for the subscription, which I think is on a similar amount, so. Um, it's certainly not a, not a, um, negligible cost, and from my experience, it can be helpful in the sense that it allows a child to, for example, sleep independently from their parents, not co-sleep with them because they're so anxious or worried about seizures. It can be helpful to detect nocturnal convulsive seizures, so a lot of those have the highest risk of pseudap, um, which is sudden unexplained death and epilepsy, which we worry about the most for patients who have uncontrolled nocturnal convulsions. And these devices can basically detect a convulsion and send a ping to a parent or a caretaker's phone to alert them that the patient's having a seizure, um. What I find too is that a lot of patients tend to have false positiveness too, which can be a little bit annoying if you're waking up constantly. So, you know, the patient, for example, wakes up at night, is wearing the watch and is, you know, scratching themselves or itching, it can go off from that. I had another patient, one of my patients telling me that they were jumping on a trampoline and it went off, so it's no surprise there. So, I, I do suggest these in select populations. So again, patients that have nocturnal convulsive seizures and for patient families that don't mind the cost that feel like this device would really alleviate their anxiety and allow a child to sleep independently, for example, not co-sleep with their parents or allow them to do more things independently, then I'm all for this advice, but knowing that it certainly has limitations and that it's not an inexpensive option. So now we want to talk about different medications um for treatment of epilepsy that have new mechanisms, and then also dietary management for treatment of um medication resistant epilepsy. So I'm sure many of you have heard and gotten questions about the ketogenic diet. I think it's still in that fad phase where a lot of patients are looking for, you know, keto-friendly options. Um, and while the ketogenic diet has been around for ages, it's actually, um, it's use in epilepsy has emerged more over the past couple of decades, but even before that, like way, way, way back then and like the Roman Empire, Um, there's documentation of people with epilepsy noticing that when they fast for periods of time that their seizures become better. So without knowing it, they were probably putting themselves in ketosis, which are reducing their seizures. So the concept of this has been around for I think for for many years, but only recently it's really been utilized in terms of um how it can be an effective treatment for epilepsy. And the ketogenic diet has specific indications too, so medication resistant epilepsy, um, a condition called glut one deficiency where patients actually have a genetic cause that prevents them from transporting glucose into their brain, so they have to use a different energy source in that situation. And then also for um patients with pyrubic dehydrogenase complex deficiency, which can cause epilepsy and their developmental delays as well. Um, and in terms of outcomes, so in randomized controlled trials in children, The mean seizure frequency was about 56% lower. So actually, this is very much on par with other medications. So we look at medication use, um, about half of patients will experience like 50% or more reduction in seizure frequency. This kind of meets our standard for what's an effective treatment. The mechanism is ketosis, but of course, we don't actually know how that works, just basically switching the energy utilization of the brain from glucose to ketones, and that seems to reduce seizures, but we don't fully understand why. I think this can be again a very good option, um, but it certainly has limitations. So the keto diet, I think works best in patients who are either formula or G tube fed. It can be very simple in that you switch over their formula, switch over their G tube feeds. Um, for patients who are otherwise taking everything orally, very active, you know, in school. It can be really tough to adhere to, especially when they're having lunch with their, you know, friends and seeing them eat other things and, um, you know, sometimes the options are always presented to them are, are, are not the best options, like usually a lot of high carbohydrate, high, high glucose options, um. But it can, there are variations on the diet too. So, I think one thing that I learned in training is it's not simply full ketosis, uh, meaning that patients are eating only fats and proteins and very, very little carbohydrates, um, but there's a spectrum to it. So, sometimes we have patients on what's called like a 4:1 ratio. That ratio can be adjusted and the ratio is always presented in 4:1 meaning, um, ratio of, uh, fats and proteins to carbohydrates. Um, but there, it's a spectrum. So, I'd say on the, the lighter end of the spectrum would be something like a modified Atkins diet or a low glycemic index diet. And some kids can do very well and still see benefits on a modified Atkins or a, a, a low glycemic index diet. Um, so, it's not all about just, you know, in the past, it would be about like putting heavy whipping cream and like, you know, only eating bacon grease and things like that. But I find that it's really dependent on your dietitian, so, Um, here at Children's Oakland, I work with Marion Wren, who's a fabulous dietitian and like really works not to um press any of these concepts onto children, but really find ways to work with their family and their food preferences to try to swap things out that could be better, like maybe, you know, switching out for keto-friendly foods or supplementing the MCT oil. Um, she provides really great recipes to our families too. We even help with meal plans, so she's fantastic, and I think a lot of the adherence to the diet. And those more complex cases for patients who are taking everything orally is really dependent on your dietitian's health as well. Um, and there are adverse effects just like anything else with the keto diet. Um, so chronic use can lead to bone density issues, nutrient deficiencies, um, increased risk of, of renal stones as well. And then digestive issues. So a lot of patients, I find that we first start them experience weight loss or constipation or diarrhea. So those are things that we manage, um, and things that you may be asked about too as a pediatrician who is co-managing these patients. Um, and then in adults too, there's description of heart disease, but that's long term and not usually seen in pediatric patients. Um, but these are things we monitor. So for example, we, um, patients who are on the ketogenic diet, we check, you know, regularly every 3 to 6 months for Um, checking for a lot of different, not only electrolytes, but different nutrients, you know, vitamin D, um, and so forth. And then we do monitor regularly, like maybe annually or biannually with um ultrasonography to look for renal stones. We usually get annual DEXA scans to look at bone density issues, and this is really important in patients, for example, who um might be severely disabled and non-ambulatory, so already not weight-bearing, already have very low bone density issues. And on top of that, if they have a convulsion, we have the consequence of things like, you know, femur fractures, things like that. So, we do watch closely in these patients. I'm sure you guys have also gotten a lot of questions about um cannabidiol, which has been around now for probably 1015 years. Um, the brand name that you might be familiar with is Epidiolex, and um this was a new medication that was initially um described in patients with Dravet syndrome and Lennox-Gasteau syndrome, but now has been approved for almost any patient with epilepsy. Um, and then in terms of outcomes, so in the randomized controlled trials for patients with Lennox-Gasteau syndrome who were treated with cannabidiol, the mean reduction in monthly drop seizures was almost 50%, so that's pretty good. Like I said, we aim for like 50% reduction in seizure frequency for any treatment that we're using in patients. The mechanism is entirely unknown. So, there's been some hypothesis that with CBD it can reduce neuronal hyperexcitability, that it antagonizes these uh GPR 55 synapses, um, can inhibit adenosine reuptake, but we don't honestly know. Um, but that being said, we still use it quite commonly. It comes in a, like an oil tincture, so it can be helpful for children who can't take tablets. Um, and we get a lot of questions about using CBD oil. Um, I think it can be good for specific types of seizures and epilepsy. So for a child who has, um, you know, broad spectrum epilepsy or patients who have Lennox-Gasteau syndrome, it can be really helpful, um, but there are of course side effects to this like any other medication. It's still a medication. It's still purified CBD. It's produced by a pharmaceutical company. Um, I find that parents are always asking about natural alternatives like, well, this is a purified substance, it's still a medicine, still has side effects, and, you know, um. Kind of, uh, paradoxically, it actually causes decreased appetite, so we can see patients, you know, have decreased appetite to lose weight on this medicine. And also it can cause some liver inflammation, so we have to monitor things like AST and ALT with some regularity as well. So I have a lot of families that always ask about this thinking that it's a natural product, but I always frame it as, you know, it's another treatment in our toolbox. It's still a medication, it still has side effects, and it doesn't work phenomenally like other medications. It works just about as good as the other ones, and, you know, it can have some interaction. The medications, for example, Onfia is one where it can raise the levels of that, not in a bad way, but we do have to watch for certain things. Um, I have some families too who will buy just artisanal CBD and use that, so I always tell them, let me know if you're using that. There are still potentially side effects. Um, it can still interact with their medications. So I think again it's another tool in a toolbox, but not like a, a phenomenal like cure-all type, um, medication. Some other new fun uh fun medications, so femfluramine, which, um, the sorry, this is the generic name, the brand name is Fenttela, came out about 5-6 years ago at this point, initially developed for Dravet syndrome, now has been expanded to Lennox-Gasteau syndrome, and um we were excited about this medication because it has a totally different mechanism than our other medicines. So we don't actually understand the full mechanism of it, but it apps it. Supposedly acts at the selective serotonin releasing um uh receptors and also stimulates 5H2 receptor subtypes too. Um and interestingly, this one has a fun story behind it. So back in the 70s, you guys might recall there was something called Fen-fem which is used as a diet pill. And then it was eventually pulled from the market because in very high doses in adults in obese females, it was found a cause for some of them mitral valve hypertrophy and because of that it was pulled off the market, you know, wasn't used again, and then about 20 years ago. And it was in, in Europe, it was used for patients with Dervat's syndrome for compassionate use, and many patients have been on that for about several decades now and used of course at much lower doses and monitored closely for any cardiac changes, and thankfully none of these patients have developed, none of these pediatric patients using phenfluramine at low doses, lower doses for epilepsy have developed any cardiac or valvular anomalies. Um, this medication comes in a liquid form. It's given twice a day like most other medications, and, um, the typical side effects that we'll see with this is it can cause decreased appetite, not surprising because it was used as a weight loss drug in the past. Um, so it's something to monitor and do things like, you know, use Periactin, but in those cases, I usually ask my pediatrician colleagues who are co-managing the patients to help monitor the weight and help treat the decreased appetite. And aside from that, it can cause like a little bit more sedation, but it's mainly just the appetite changes that we tend to see with it, um, and then of course patients who are also on other SSRI agents, we think about um monitoring for things like serotonin syndrome if someone's on multiple SSRIs. Um, and in terms of outcomes, so when this was initially tested in patients with Gervat syndrome, they had a really robust response. So nearly 2/3 reduction in convulsive seizure frequency permanent Gervat syndrome patients, which is outstanding because these are the patients who, you know, come in with prolonged febrile seizures, always admitted status epilepticus. So for a really refractory population, they responded phenomenally. Um, the data has not been robust in patients with Lennox-Gasteau syndrome. Um, and then we have used this off-label in other patients that don't have these syndromes but do have medication resistant epilepsy. I would say it tends to work again, just like the other medications too, but I like this one because it does have a different mechanism of action. So going on to one of the newer medicines, it's called synovamate. You might have heard of this. Um, the brand name is Excorpre, has an indication for focal epilepsy. It was first used in adults, but now we are using it off-label in children. Um, and in terms of outcomes, so in randomized controlled trials for uncontrolled focal epilepsy, um, there was a pretty good responder rate of 64% of patients, so about meaning responder rate, half of those patients had, uh, I'm sorry, 64% of those patients had a 50% or more reduction in their baseline seizure frequency. Um, so, it was pretty good. So, kind of better than we think, a little bit better than think about for typical medications when we're using that. Um, the mechanism is a bit different, so it, it works by inhibiting both fast and slow in activation of sodium channels, um, and then it's also a positive allosteric modulate. Habit channel, so kind of a combined two different combined mechanisms here. Um, what's nice is it comes in these blister packs when someone's starting it. And I kind of think of Sonoma is a little bit similar to Lamictal. That's one of those medications that you have to start really slowly, because when they were doing the clinical trials, there were a few patients who developed um severe skin rashes or Stevens-Johnson syndrome, which is the same thing with Lamictal, which is why we go up really, really slowly. So again, not a, not a great medication choice to starting something quickly and increasing it. But if you have a couple of months, this is a good option, and I like these blister pack options where um families can use this to do the up titration. And it's given once at bedtime. It comes in a tablet, but it's really small, so I've had patients who even have trouble taking pills, be able to take this. Um, and again, I think it can be a really good medicine. It has a different mechanism than older medicines, um, and it tends to work a little bit better than other medicines out there. And then lastly, before I wrap up here with my slides, I just wanted to highlight health disparity since um many of you in the community as well as us here at Children Oakland um treat a large population, many um many lower income resource limited patients, um, and this is a study done that was back in just 2023, so rather recent study, um, identified that patients who were younger, of minority race ethnicity, and who had Medicaid insurance were more often um treated medically than surgically. So, again, I just spent this past hour emphasizing to you that within our epilepsy population, a third of our patients are medication refractory. We know that medicines, more medicines won't work to cure them of their epilepsy. And so what we do is we pivot and talk about those things like dietary modifications, devices, surgeries. So, if those have better better chances of making these patients seizure-free, you know, why are we not offering this more to our patients over here as well, just because they have, you know, different Um, access to care and, and different um, uh, family circumstances. Um, and so one of my goals here at being part of the Oakland community is helping to build our epilepsy surgical program at Children's Oakland and so I really want to focus on this in terms of, um, health disparities and be able to provide that to all of our patients that we see here as well. Great, and here are my references. And move on to questions. And I've included here too, you know, we always appreciate um working with our community pediatricians and if you guys have patients who maybe had a first-time seizure or febrile seizure, you'd like to refer to our clinic. I've included our number here for our new patient coordinator Jamata, her phone number, our fax number for referrals, and then a couple of other numbers here too for our central scheduling, um, and then where to call to check the status of your referral. And these will be uploaded too, so you should have this as well.
