Thank you, Eddie. I just wanted to start by asking the audience how many of you are in one of the multiple disciplines, nursing, pharmacy, research, physicians, in the neurosciences, and then I'll ask about medicine, and then medicine, anesthesia, other areas. Perfect. Okay. Just so we calibrate a little bit so we can, and how many of you work in the emergency department? Okay. Great. So, part of the reason for me asking is that we're going to talk today about status epilepticus, specifically what's new, and this is a disease that occurs as a continuum, and so it's really important for us to think about how each of us interfaces with the condition. These are my disclosures. The industry ones specifically are not related to this talk, although previously I was a consultant for UCB. So learning objectives here are listed. We're going to talk about these specifically in detail, but we're going to talk about different lines of therapy for convulsive status. We're going to demonstrate how practice pattern variation itself can influence outcomes and specifically how we measure them. We're going to talk about approaches to non-convulsive status, including electrographic status epilepticus after cardiac arrest. We're going to talk in detail the new consensus guidelines for the diagnosis and management of conditions called Nourse and fires, and some of those may require a little bit of education about what those are, and we'll do that today, and then we'll demonstrate situations in which diagnostic evidence may infer an electrographic diagnosis of status that is scalp-negative, electrometabolic, empiric, or possible-slash-uncertain. So this is sort of the borderlands. The International League Against Epilepsy defines status epilepticus as ongoing seizure activity due to either failure of mechanisms responsible for terminating seizures or initiation of mechanisms that provoke ongoing seizures, and specifically that these would cause prolonged seizures after an early time point, T1. This time point, in essence, means the time point after which those seizures would continue and beget themselves, and then long-term consequences after a time point, T2. And you would ask, well, why didn't they just tell us what the times are? And that would be really kind of them, but I think the point is that these times are variables and they vary according to the situation. Here is a diagram on the left from Nathan Fountain, who published a really nice review of the pathophysiology of status epilepticus. It's someone outside the scope of our conversation today, which is going to be geared towards management and guideline-driven approaches, but you can review it here, and I'll just take a moment. One key aspect of this is really receptor internalization, so that both seizures and the medications like anesthetics we use to treat seizures cause internalization of certain receptors that make neurons more at risk of depolarizing. And so both the sedation that we use to treat the seizures as well as the seizures themselves can have a feed-forward effect on downstream effects. And finally, you can see here that these GABA receptors, that endocytose, GABA effects also have downstream potentiation of other mechanisms like AMPA, and you'll see other channels involved in MDA, et cetera. So first I'm going to talk about guideline-driven therapy for convulsive status epilepticus. And before I do, it's important to pause and comment on the status of sort of guidelines from different bodies. The Neurocritical Care Society and American Epilepsy Society have together formed a panel to develop a revision to the existing guidelines that each society had individually developed. We have, as a result of that, a number of co-chairs, Corinne Berger, myself, Jadip Kapoor, Jackie Bainbridge, so it was really good multidisciplinary involvement. So that panel has just convened, so much of my talk is going to talk about what's leading up to those revisions in terms of evidence. So here's a diagram. It's complex, but it gives you a sense in blue of the timeline after status epilepticus. I want you to pay particular attention to, in red, where we demarcate for convulsive status that time one, the time where seizures are self-sustained, is five minutes for convulsive status epilepticus. We know this because after that time, it's highly likely that seizures will not stop. So the old definition at which status epilepticus began or was diagnosed at 30 minutes is really out of date, and most of us would now consider status to begin at five minutes. So we shouldn't wait for 30 minutes, principally also because at T2, that time for harm, we believe that time is around 30 minutes. What's really troubling is that the current and conventional therapy would have us start treatment for convulsive status at time zero, you know, if we had a patient in front of us, and at five minutes would pass, we'd be giving benzodiazepines, and they would have us begin second-line therapy with a medication like phosphonatone, valproic acid, levotracetam at 20 minutes. And if the seizures were then, those established seizures did not respond, we would call them refractory. But we would only do so usually around 40 to 60 minutes, a time when there's already potential for harm. So this is what we would call treatment in series rather than, say, treatment in parallel or stacked therapy. And in fact, this is the existing guideline-driven approach. But many of us have collapsed the second-line therapy earlier on into the first-line therapy given the low risk. This drives home that importance of the early treatment of status epilepticus. Basically the earlier the treatment, the better the outcome. So the percent who respond to first-line therapy, if treated within less than half an hour, is almost 80%. But as you go on in time, that number drops further and further down. This is data from the RAMPART study. Lead author Rob Silbergleit shows that both IV lorazepam and IM midazolam both have a similar time in red from opening the box to stopping convulsions. However, if you already have an IV in, the time from actively treating the patient to those seizures convulsing is shorter in white for IV lorazepam than for IM midazolam. And that is because in blue, it takes a long time to put an IV in. And so if you're in the hospital in a critical care environment, this can give you some sense that treatments obviously should be IV. But for pre-hospital response to status epilepticus, both options are reasonable. The shorter time to IM drug administration was offset by the faster onset of action of IV, resulting in similar latency until convulsions were terminated. Importantly, when we look back, this is by Abby Saeth looking at data from the trial called ESET, which I'll discuss later. If you look at patients who qualified for that study by being refractory to benzodiazepines or other examples there, on the top, adults, and on the bottom, children. And then I'll use the arrow here. On the left is dosing of diazepam. In the middle, dosing of midazolam. On the right, dosing of lorazepam. Doses here in blue are those that are considered meeting the guideline-recommended dose. Doses in red are doses that fell short of the recommendation for the guideline. So obviously, you can see here that there's a lot of red. All right. So everyone in the audience should talk to your colleagues in the ED and the ICU and ask them a little bit about why we're all in the red. We're all underdosing our benzodiazepines. And I think we probably know why, is we're afraid of patients getting intubated. However, there's decent data, even placebo-randomized control data in the pre-hospital setting to suggest that patients who get benzodiazepines rather than placebo have a lower rate of getting intubated. So by treating the patient and stopping seizures, it's more likely that we're going to stop patients from seizing and not have them be intubated. So these are guideline-driven or these are data-driven guidelines from 2016. This restates what we talked about before. In adults, IM midazolam, IV lorazepam, IV diazepam, with or without phenytoin, and IV phenobarb are established as efficacious stopping seizures lasting at least five minutes. Intramuscular midazolam has superior effectiveness compared to IV lorazepam. And we discussed the issue of needing established IV access. And these were published in 2016. And I don't expect much revision there. In convulsive status, if patients progress to what we call established status, meaning they're refractory to an adequate dose of benzodiazepines, there are three recent studies in 2019 that were published. And they're different for varying reasons, but some of them have commonalities. This one, the Eclipse study, was done in the UK at 30 centers in children. Interestingly, consent was deferred, and 92% consented thereafter. And it looked at the treatment of levotiracetam 40 mg per kg and phenytoin at 20 mg per kg with an outcome of the time to cessation of all visible signs, visible signs, of clinical convulsive activity. And there was no difference here. And you can see the numbers in the red box. The need for further anticonvulsants was about 16% in each group. This reflects a better response rate to status in children than in adults, especially in the early setting, convulsive status. The concept clinical trial was performed in Australia and New Zealand, 13 centers, similar endpoint. Although differently, they allowed for a crossover, meaning repeating the anti-seizure medication but the other drug at 35 minutes if necessary. And if you look at the two-hour response rate, these were also high, reflecting also that this being a pediatric setting with patients up to 16 years of age. Response rates were in the 70% range, 78, 72, and were not different between phenytoin and levotiracetam. Importantly here, if you look at the patients who needed another alternative study drug in the first two hours, right above that red box, those numbers are pretty high, 37 and 40% of patients went on to crossover. And so it gives you the sense that while not part of the current guidelines, crossover therapy to a second anti-seizure medication, at least as administered in this study, was part of the protocol and was required in many patients. And it's possible that this may defer intubation in some patients. But it's also possible that given that series-based treatment may push you longer and longer past that T2 of injury time, you have to be cautious about this given that they were repeating anti-seizure medications at 35 minutes past the time we think that injury is already occurring. The ESET clinical trial was a study that was conducted in both children and adults. These were patients over two years of age. And this compared three drugs, adding valproic acid or sodium valproate to the mix. And levotiracetam was studied at a higher dose of 60 milligrams per kilogram. And here, not only did patients have to stop visibly convulsing, but they had to have improved responsiveness at 60 minutes. And they couldn't have had the need for additional medications for either seizures that continued or for intubation. And here, you'll note that the response rates in the colorful graph on the bottom left were lower, ranging sub-50%. There was no difference between these medications. And so the conclusions of the study were that there was no significant difference in the likelihood of seizures terminating with these three medications. On the right, you can see additional data showing the time to seizure termination, suggesting maybe sodium valproate had a slightly shorter time to termination, suggesting that phosphenetone in this population had a slightly higher rate of intubation. Although if you look at the, and this was clustered somewhat in children, but if you look at the other two studies I just presented, they didn't confirm many of those findings. So in essence, this study establishes three drugs that could be given for established status epilepticus. One interesting look at the outcomes we did was to try to understand why patients are getting intubated for status epilepticus. It turns out that one of the highest risks for getting intubated is what hospital you show up to. And when patients get intubated, they do stop seizing. But in the ESET clinical trial, that would have been considered failure. And so on the top in blue bars on the left are the top five highest enrolling pediatric sites. On the top right are the top six highest enrolling adult sites. And you can see that for children in these five sites, the intubation rate at one hospital was 4%. But at another, it was 32%. And in adults at one hospital, it was 19%, and another 39%. So that risk of being intubated was 23-fold if you showed up at a center in the highest quartile of risk compared to a center in the lowest quartile of risk. And that was independent from whether, in yellow, seizures persisted at 20 minutes, or whether you were improving in responsiveness, or whether you had a side effect or adverse effect like hypotension. The reason this is important is that if our trials are predicated on visible effects, there are patients who may be continuing to seize. And this shows you data from Shariar Zetabchi, who also looked at follow-up data, looked at tertiary analysis in the ESET clinical trial. If you look at patients who happened to get an EEG later on, this was not part of the study protocol. There were some 39% who had seizures and some who did not. But if you looked at 39% who had seizures, 33% of these were considered to have, a third of these were considered to have had treatment success. If you're not doing EEG and the patient's quiet before you and not convulsing, then you'll probably be convinced that there's no seizures and you're all set. But a third of those patients are actually seizing. So the lack of EEG data in this setting is somewhat troubling, both because patients may get intubated for the assumption that seizures are continuing, or because the assumption that patients are sedated and are not yet waking up. And we may not consider the outcome to be termination of the seizures, but we're actually considering the outcome of what we see visibly at the bedside. As a result, Jim Chamberlain, who is at Children's National, and colleagues, some of whom are in the audience, and myself, have worked to devise an idea for ordinal measures for determining the response to treating convulsive status, where the best outcome would be a patient who stops convulsing and wakes up. But second best might be someone who stops convulsing and doesn't wake up just yet, and so on and so forth. So these have been proposed, but not implemented in any clinical trials. So here are, for established status epilepticus and second-line agents, recommendations that are derived from the clinical trials that were presented, and some other approaches that historically have been implemented as part of, for example, the VA cooperative study, and also newer regimens that have not yet been actively studied. When patients don't respond to secondary anti-seizure, second-line anti-seizure medications, and continue to seize, that risk for continuing injury becomes much more concerning. And third-line anesthetic agents may be considered. There's not a lot of evidence in this area. This is data from Andrea Rossetti, where over two years, looking at three centers, they really had a challenge randomizing these patients, because many of them had already received therapy in one of these categories, and therefore couldn't meet criteria for the study. But they looked at allocation to either propofol or benzodiazepine therapy, or barbiturate therapy. Of interest here, the intubation time in survivors was much longer, 13.5 days, compared to four days for propofol in patients receiving barbiturates. So I think that's important. Additionally, many are using other medications, propofol and one including ketamine. And here's data from Nick Gaspard, who looked at patients who had a likely response to ketamine and were treated with a dose of seven mgs per kg per hour. That's quite higher than for sedation with ketamine. And possible, likely, or no response to ketamine, where the doses were much smaller, 2 to 3 mg per kg per hour. And so I refer you to this paper, although guidelines to date don't reflect these approaches, because there's not a lot of data in this area. I think the guideline panel will take this on, and you should look for recommendations moving forward. Nourse and fires represent new onset refractory status epilepticus in children, and now we're learning adults, febrile infection-related epilepsy. And these can have quite poor outcomes. And importantly, these are usually identified in the intensive care unit setting and often are non-convulsive. There's complex data here showing different types of therapies that have been tried in these patients. But you can note that the response rates are not great. There are some small samples where the ketogenic diet had response rates of 50% to 67%. Plasma exchange, hypothermia, but we also have other clinical trials showing a lack of benefit of hypothermia. You can see at the top, steroids had positive effects in fires in 17%, but some only transient, and in Nourse of 38%. But also, it's challenging because combination therapy is commonly tried, and differentiating the effects of therapy versus other intercurrent seizure medications may be challenging. So as a result, a consensus group was convened to think about these conditions, which are often caused by inflammatory or underlying paraneoplastic conditions. And I was happy to be part of it. It was facilitated by a group listed here, and Ronnie Wickstrom did a really terrific effort. The Delphi process is really a process of iterative work and surveys to achieve consensus. So it represents expert consensus and not empiric data, although they did publish a really nice set of two publications, one about the recommendations, but another which includes really all the data that were reviewed. And I think it's a wonderful publication review. It was published in Clinical Neurophysiology. So points of consensus were sought, and you can look at the international and multidisciplinary group here. And I'm going to go through some of these. So particularly, the way to orient yourself is to the response here. The responses were 1 to 9. Response of 9 was complete agreement, and response of 1 was complete disagreement. And the level of agreement were those who were close to agreement, 7 to 9. And the level of disagreement were the percent that were at the bottom, which were, I think I should have a typo here, and they should say 1 to 3. So what you can see here is there's pretty good agreement for these aspects that NORS and FHIRs are conditions which may happen in all ages in all patients. That FHIRs, the febrile sort of version, should be considered a subcategory. These don't seem to have a geographical trend, no definite seasonal trend, but there's some concerns. That they seem to be autoimmune conditions, and one should consider thinking about primary autoimmune etiologies in these patients. And in some patients who have non-CNS existing malignancies, a perineoplastic etiology should be suspected. And that post-infectious immune activation is likely very important, and that inflammatory phases may precede the development, but also contribute to their persistence. If you're dealing with sort of a nebulous, either perineoplastic or inflammatory autoimmune condition or syndrome, which is driving these. Less agreement about some of these statements, but that these conditions can lead to permanent cognitive stability, and that they can lead to permanent and more severe epilepsy. In green, they discussed what to do, how to work up these cases. Specifically, that it's appropriate to perform similar investigations for these patients as you would with anyone else with status. Specifically, early testing, in addition to that, early testing with autoimmune antibodies, access to autoimmune antibody analysis. We can talk about what those would be. A comprehensive rheumatologic evaluation, an infectious evaluation, evaluation for inborn errors of metabolism in young children. Autoimmune and onco-neural antibody panels. And additionally, that performing an infectious evaluation relevant to where the location of the patient is, you know, histoplasmosis, et cetera, neurosurgosis, so forth. Evaluation for autoimmune panels, and in the CSF, can also be of value, compared to what I spoke to previously, which regarded analysis of blood. And they talk here also about using GADO for MRI, whole body PET, and then, in young children, genetic testing. And here, you know, they speak about the guidelines speak to the use of EEG. If there's MRI, and there's a targetable lesion, and it's still uncertain what the etiology is, that a brain biopsy could be considered. Malignancy screening with CT chest, pelvis, and abdomen. And at the bottom, that this should specifically include testicular and ovarian ultrasound, as NMD encephalitis and other perineoplastic driving malignancies can exist there. In treatment of these conditions, the guidelines recommended similar therapy as in other conditions, but that first line immunologic therapy should be started in the first three days. That steroids, with modest agreement, are the first line immunologic therapy as methylprednisone, 20 to 30 mgs per kg for three to five days. And that less agreement that IVIG can be given, and if so, that it should be given in a certain dose. And that these, a decent amount of agreement that these should be given simultaneously. Pretty high agreement that ketogenic diet should be administered early in the first week. But much more agreement that it should be administered if cases are refractory over a prolonged time. Finally, that hypothermia is not recommended in these populations, and that secondary immunologic therapy should be considered and started within seven days. We can talk about these at the bottom, number 24. Fairly high agreement that treatment with IL-1 receptor antagonists, or IL-6 receptor antagonists, and Akinra tocilizumab should be initiated in cryptogenic patients without clinical features of autoimmune encephalitis. So this is a quote that's not mine, but that there's no specific anti-seizure medication that appears to be more effective than another. And so a lot of this therapy is what we call rational polytherapy, driving different receptors, NMDA, AMPA, GABA receptor-driven therapy. Here is a designated workup. You can review this in the manuscript that's linked to at the bottom. But it's what I just discussed, and the continued workup as I had just discussed. And importantly, there's a patient family support site, the Norse Institute, norseinstitute.org. I would direct you here. And it's really helpful if this group centered at Yale can receive biospecimens and learn more about what drives these conditions. So I'm going to thank you and stop there. I will say that in the slides I left online, there are some additional discussion about conditions like non-convulsive status. And what you'll see there is that it replicates what we've talked about before. But there are a number of patients for whom the scalp EEG is really not showing the full detail. And there have been studies of either invasive metabolic brain markers. We just had a multimodality monitoring plenary. You heard about some of those. But whether it's microdialysis, brain oxygen, or imaging studies, whether it's CT PET imaging, we have shown some cases there where you will find seizures that you did not see on a scalp EEG, no different than a patient who was brought into the epilepsy monitoring unit for a phased workup for surgical epilepsy. And finally, that in many of those patients, even the addition of an intracranial electrode inside the scalp in the same region, say an ICP monitor would be placed, can augment diagnosis of seizures that are missed on the scalp. So we'll stop there and appreciate your attention and interest. Thank you.

status epilepticus

diagnosis

management

early intervention

convulsive status epilepticus

non-convulsive status epilepticus