Acute Symptomatic Seizures: Diagnosis & Management

Podcast Transcript

Introduction: Neuro Pathways, a Cleveland Clinic podcast exploring the latest research discoveries and clinical advances in the fields of neurology, neurosurgery, neuro rehab, and psychiatry.

Glen Stevens, DO, PhD:

The growing use of continuous EEG to monitor patients following acute brain injury has significantly increased the number of acute symptomatic seizure diagnoses and consequently the use of anti-seizure medication, but only a fraction of these patients continue to have seizures after hospital discharge. In this episode of Neuro Pathways, we're discussing acute symptomatic seizures and the short and long-term management of these patients. I'm your host, Glen Stevens, neurologist, neuro- oncologist in Cleveland Clinic's Neurological Institute. I'm very pleased to have Dr. Vineet Punia join me for today's conversation. Dr. Punia is an epileptologist in Cleveland Clinic's Neurological Institute's Charles Shor Epilepsy Center. Vineet, welcome to Neuro Pathways.

Vineet Punia, MD:

Thanks Dr. Stevens. Thanks for inviting me for this podcast.

Glen Stevens, DO, PhD:

So I still cover the inpatient service occasionally on the consult service. And quite often we'll see patients who've had seizures, and I'll usually ask the medical students and the residents, "What's our individual lifetime risk of a seizure?" And I'll see what they have to say, and they'll guess 1%, 2%, those types of things. And I tell them, and you can correct me because you're the expert, but I tell them that an individual's lifetime risk of a seizure is actually around 10%, which I think is always shocking to everybody. And if there's 10 of us standing there, it means one of us is going to have a seizure. And there's lots of reasons why that could happen or those types of things, but is the number accurate? Am I out to lunch? Should I tell them something else?

Vineet Punia, MD:

No. You are exactly correct. The lifetime risk of at least one seizure is around 10% in the general population. The 1% risk that is often quoted is about epilepsy. The risk of developing epilepsy is 1%, and that highlights the common prevalence, so to say, of acute symptomatic seizure, because that gap is filled by acute symptomatic seizures from the 1% to 10%. The 9% extra risk is because of these acute symptomatic seizures.

Glen Stevens, DO, PhD:

Yeah. I can't tell you how many conversations I've had with patients that have had multiple seizures and they go, "Well, I have seizures, but I don't have epilepsy, right?" Then we have to have the discussion in turn. And maybe we should just have that now as a baseline as we go in. Won't you just tell us the difference between somebody that has a seizure and somebody that has epilepsy. What's the difference?

Vineet Punia, MD:

This is always a great question. And as you mentioned, there are always patients who do end up confusing seizures and having multiple seizures with epilepsy and their terminology, and especially epilepsy still unfortunately has some taboo attached to it. So people would rather have seizures or recurrent seizure, seizure disorder rather than being labeled as epilepsy. So the primary difference is

that seizure is basically a single event, single episode of a symptom or a sign that the patient experiences because of abnormal, excessive firing of the electrical activity of the brain. So that's a single seizure. But when the brain gets this ability to produce spontaneous and recurrent seizure over time, that's when somebody is diagnosed with having epilepsy.

So essentially seizure is kind of a symptom of a much broader disease, which is epilepsy. However, just having a single seizure or even multiple seizure at times does not mean that the person has developed epilepsy. And that is in the scenarios of this acute symptomatic seizure, because these seizures occur in acute setting of insult to the brain of some kind and are a symptom of that acute insult. This doesn't necessarily mean that the brain has developed inherent ability to produce spontaneous seizure. So these seizures, acute symptomatic, are just symptoms of the underlying disease or an injury. So that is the primary difference between the acute symptomatic seizure, individual seizure, multiple seizure, and then epilepsy.

Glen Stevens, DO, PhD:

So I know in the intro we talked a little bit about acute brain injury, and you can certainly expound on that, but if we're in the hospital and we just go into the intensive care unit and we have a lot of sick patients in there... Let's say there's 10 patients in there. If we hooked up an EEG to these patients, how many of them would be having some abnormal electrical activity?

Vineet Punia, MD:

So normal electrical activity that can be labeled as seizure, depending on if it's a medical ICU or neuro ICU, the prevalence or frequency of finding these could be anywhere from 10%, 20% or even higher, if it is really like traumatic brain injury unit and things like that. So if it's a medical ICU, the figure ranges around 10 to 15%. If it's patients with neurological in neurological ICU with neurological injuries, it's to anywhere from 15 to 20 to 25%. So that's the common.

Glen Stevens, DO, PhD:

So if we're concerned, how do we monitor these patients? What are we doing these days to assess the seizures?

Vineet Punia, MD:

Right. So nowadays, because of exponential improvement in technology of data storage and long-term monitoring, we have continuous EEG that we perform on these patients, especially on the high risk patient. If you have patient who has, let's say, even stroke hemorrhage and you expect next day for them to improve and start responding, but if they're not, that is a time where you should be concerned that these patients may be having non-convulsive or non-clinical seizure. And those patients, we should perform continuous EEG monitoring to detect these acute symptomatic seizure from the stroke or hemorrhage or TBI.

Glen Stevens, DO, PhD:

Mm-hmm. And maybe you can help me. See, I'm feeling very educated here today from you. But the things that always confuse me are that we put a continuous EEG on a patient and we see a sharp wave or we see a PLED. How do I know when I should start somebody on a medication? I mean, I know it's not a simple here or there, but obviously if they have a obvious electrical seizure, it's easier, but what about these other things, these sharp waves, these PLEDs?

Vineet Punia, MD:

Sure. Very critical and important question. And we do not have clear data from research so far, unfortunately, although we've known about the existence of these PLEDs for more than 50, 60 years now. But what should help us make decision of whether to start medication or not depends on if these are PLEDs or what we call LPDs now, lateralized periodic discharges. Those definitely, those patients in the next 24 hours, we know that around 50% or 60% of them would have seizure if you continue to monitor them. So those are high risk patients. So in those patients, even profile actively treating with a anti-seizure medication maybe of benefit.

Sharp waves don't carry as high risk. But if, let's say, the patient has poor mental status, has a cortical injury of some kind, which we know are at high risk of develop seizure. And in case, if you do not have the ability to continuously monitor them for long period of time, depending on the availability of resources, it would be better to start these patients on medication to reduce the risk. However the risk, again, to emphasize with LPDs or PLEDs, is much higher compared to some other epileptiform discharges like sharp wave or even the generalized periodic discharges or what we call periodic patterns and things like that.

Glen Stevens, DO, PhD:

So, you mentioned something that I'd like to delve into a little bit better. Is there a sweet spot for timing for continuous EEG? Is it two hours, 12 hours, 24 hours, 48 hours, and after that you can feel pretty comfortable one way or the other? Is there a sweet spot?

Vineet Punia, MD:

Yes. There is. And there's a lot of research, thankfully at least, done in this sphere of critical care monitoring. So there is in fact a prognostic model called 2HELPS2BE. And what it does is, depending on the first one hour of EEG, it accounts for some clinical variables. Like if you had a clinical seizure or convulsive seizure, that is a high risk. If they have what we call highly epileptiform discharges, like LPDs, if they have seizures or epileptiform activity which is not 10 second, cannot be considered as a seizure seizure, but are considered like what we call BIRDs, which is brief ictal rhythmic patterns or discharges. If the patient has those kind of prognostic indicator of impending seizure, those patients should be monitored for longer period of time.

But if those things are missing and the score is, let's say, two or three, then a 24 hour, 12 to 24 hour monitoring is sufficient. And if you do not see anything, then those patients are low risk, the risk of seizure in the next 24 hours, less than 5%. And you can safely take those patients off of monitoring. So 12 to 24 hours of monitoring and being guided by this scoring system is really helpful.

Glen Stevens, DO, PhD:

Good. Well, you've educated me. I'm was unaware of the scoring system. So thank you. So patient has a traumatic brain injury, has a seizure in the hospital. How aggressively do we treat it?

Vineet Punia, MD:

So I personally believe we, again, it's not backed by strong and great clinical research, but based on our experience and some at least information we have, these should be aggressively treated to the point that we should try and get control on these non-convulsive acute symptomatic seizure, because at least the data we have suggests that these patients who have these seizures have poor outcome at time of discharge, have increased mortality if the number of hours of seizure is longer. So, and these seizures are often recurrent.

So we should aggressively try in general, as a rule of thumb, try to aggressively treat them. Now we may come to a point where we have tried three, four medications, and if, let's say, especially the patient has a good mental status, they're having what we call epilepsia partialis continua, very brief, small focal seizure, then you have to weigh or balance the cost and benefit of we don't want to necessarily intubate them and things of that nature for just those minor seizure. But otherwise a general rule of thumb is that we should treat them because the increased risk of poor outcome and in fact mortality.

Glen Stevens, DO, PhD:

So one of the things I like about our program here is that when I see these patients in the hospital, I can send them to what I believe is called the PASS Clinic on discharge, the post-acute symptomatic seizure clinic that's run by the epilepsy folks. Won't you tell us a little bit about that? Are you involved with that clinic?

Vineet Punia, MD:

Yes. So after my fellowship, which I did at the Clinic, I started working as a independent physician. I realized it was my area of clinical research, but these patients had very poor follow-up. And in fact up to 90% of them used to remain on seizure medication for longer than a year or even more than that, based on the research that I had performed during my fellowship. So that was a big motivator for us to start this PASS Clinic or Post-Acute Symptomatic Seizure Clinic, which we began in 2017. So it's me along with Stephen Hantus, Jessica Fesler and Zhong Ying, who primarily are the people who read the continuous EEG monitoring.

So we understand these patterns, the impact of these patterns or acute symptomatic seizure. So we made it a point that we'll prefer that these physicians who are well versed with continuous EEG monitoring also follow these patients, outpatient in the PASS Clinic. The primary rubric or framework of the PASS Clinic is that we want to see any patient who did not have history of epilepsy when they were admitted, were not on any anti-seizure medications, underwent continuous EEG monitoring because there was a concern for acute symptomatic seizure, and now are being discharged on anti-seizure medication after the hospitalization.

So when we started the PASS Clinic, we worked with our IT team and EPI team to come up with this best practice advisory alert that pops up automatically for the discharging team if they're discharging a patient who was not on anti-seizure medication, underwent continuous EEG monitoring, and is being discharged on medication. It suggests that this patient is a good candidate to be followed up in PASS Clinic. And also we prefer to do a EEG, like one hour EEG at least at the time when they come. So we look at the EEG, we see how they've been doing, and then we use that clinic to decide how long the patient should be on medications and what are the risks of epilepsy development.

Glen Stevens, DO, PhD:

Yeah. So if I can dig into that just a little bit more. 70% of these patients potentially don't need to just be on anti-seizure medication long term. What are the types of things that you use to help differentiate who should stay on and who can come off?

Vineet Punia, MD:

Great question. So the primary drivers of epilepsy development in general... Because of course, there are various etiologies that cause acute symptomatic seizure. So if the etiology was more of a systemic insult, let's say, electrolyte imbalance, like hypernatremia and hypoglycemia, those patients do not need to remain on medication. Even at the time of discharge, they can be safely taken off.

But if, imagine, let's say, somebody develops a stroke, hemorrhage or TBI, the injury remains there. It becomes gliotic. We don't know, in a sense. In those patients there is a gray zone of when to stop medication. And what we know at least, that if there's a cortical involvement, that increase the risk of epilepsy development or recurrent seizure afterwards. If the severity of the injury, so let's say for stroke, higher the NIH Stroke Scale, the more is a risk. For hemorrhage, the more the volume of the hemorrhage. So if it's more than 10 milliliter, that's a higher risk. So those kind of features of cortical involvement and severity of the injury. And if they had a convulsive acute symptomatic seizure, those patients of course are at higher risk than somebody who had a stroke, but didn't have convulsive acute symptomatic seizure.

So what I use in my clinic, there are a couple of prognostic models that have been proposed and have been validated. So for stroke, there is a SELECT model, S-E-L-E-C-T. Stands for severity, which is based on NIH Stroke Scale. Patient gets a score of zero to two. L stands for large artery atherosclerosis. So if there's evidence of the stroke coming from large atherosclerosis, that's a higher risk. Then there's early seizure or acute convulsive seizures. If they happen in these patients, that increases the risk and the score goes up by three points. Then there is involvement of the MCA. If it's the MCA territory stroke, that increases the risk and then the cortical involvement. So the select scores, I use that score to come up with an individualized kind of risk assessment for the patient.

Similarly, there's a CAVE score for hemorrhaging patients, which stands for cortical involvement ages less than 65 years, that's the high risk, volume being more than 10 milliliter of hemorrhage volume, and then early convulsive seizure. So these are the imaging and the clinical variables that we use.

Now, my field of research or interest, I'm an epileptologist. All I do is EEG, looking at EEG. So when I started in my fellowship, and since then, we've been looking at the role of EEG, the acute EEG, the EEG that we do, the content EEG monitoring we do. Then that helped us data mine the risk as well, because as you see, the CAVE score and SELECT score does not account the EEG findings in there. And epilepsy is primarily a electrophysiological dysfunction basically.

And what we have found out that... We recently have published several papers on this, that even after adjusting for all these clinical variables and neuro imaging variables, the patients who have acute symptomatic seizure or these LPDs, or PLEDs you mentioned, they have 10 times increased risk of having seizure or developing seizures later on. So now I also account for what the EEG show during the admission. Did they have these non-convulsive seizures or LPDs? And I consider these patients are being at higher risk than what SELECT score or, let's say, CAVE score or those things suggest.

And lastly, we also perform, as I mentioned, as part of PASS Clinic, we do EEG for an hour at least before the patient sees us to see if there's still some sharp wave, some epileptiform discharges, because those patients, we know at least from the traditional epilepsy literature, like when somebody has epilepsy and they've been seizure free for two or three years, you want to decide about stopping the medication. EEG showing epileptiform discharges means that the patient is at higher risk of seizure recurrence if you stop the medication. So we do that EEG to also guide us at the time whether to stop medication or not. So if they show epileptiform discharges, then I'm more conservative. I keep them for a longer period of time.

And lastly, I really take into account the patient preference or family's preference as well, because it is still in a gray zone. The science is not settled yet, as in we do not have great data. So I discuss at length the risks and benefits and what their individual risk of developing seizure is. To somebody at 20%, 30% risk may not be too high, or to someone who is very risk averse, 10, 20% risk is too much, right? So I have this nuanced discussion with them as in this is your risk. If the medication is not causing any side effect, if they're tolerating it, okay. They are coming out of rehab, finally getting back to normalcy after the stroke or hemorrhage. And if they don't want to, sort of say, rock the boat, then we continue for a little longer, especially if they don't want to take that extra 10, 20% risk of having a clinical seizure. So all these things go into deciding whether to stop medication or continue for some.

Glen Stevens, DO, PhD:

No. I appreciate it. It's very good information. The thing that's always a difficult discussion that I find with patients is if they want to drive, and if they've been on seizure medication and I've seen them post hospital stay and they haven't had any further events and they don't have any of these significant risk factors and they had some type of an event and they've been on it for six months and they haven't driven for six months, I always get anxious about I'm going to stop their medicine at six months. And now I'm going to say, "Drive." I wonder if I should have said, "Let's stop it at three months." Then you'd have three more months where they're not driving, where you'd feel better about what's going to happen. There's no easy answer to this. And obviously it's variable between the states, but how do you deal with this?

Vineet Punia, MD:

I'll first tell how I deal with it, and then I'll go into, or plug my research or some interesting finding that we just found. So the way I deal with it is kind of what you said. So Ohio, thankfully, in a sense, doesn't have very strict time limit. That's where we practice, right? So they leave it to the discussion of physicians a bit. So if the patient is low risk, rather than continuing for six months and then stopping as you said, because I definitely recommend them not to drive for a month or four to six weeks after stopping medication, because that's the high risk period. Nobody knows. Despite our best effort of doing EEG and looking at these prognostic models, we cannot never be sure. So I recommend them not to drive after stopping the medication for four weeks.

But if they're low risk, I tend to stop medication after three months. I say, "There'll be washout period. And then you don't drive, and come five to six month period" Especially if their job really depends on it and if they've recovered very well, then maybe by end of four, four and a half month, I may allow them to drive, especially given that there are no strict time limits here in Ohio. But each state is different. And then patient's preference, patient's dependence on livelihood for driving and all other subtle things have to be considered as well while making this decision.

And the interesting research finding I was initially alluding to was, so this PASS Clinic that you referred to, five of us came together. Five centers came together thanks to American Epilepsy Society grant to form a network of five centers, which have PASS Clinic. It's called PASSION, which stands for Post-Acute Symptomatic Seizure Investigation and Outcomes Network. PASSION. So we are the leading site and our other collaborators are MGH, Yale, Brown University and UNC Chapel Hill.

Glen Stevens, DO, PhD:

So any other research or anything else, hot button items? I know there's always so many new medications that are out there, but anything else that we haven't covered that you feel is important?

Vineet Punia, MD:

The important part is the initial message that these seizures should be aggressively treated. If you don't have resources for continuous EEG monitoring and you do a spot EEG or a short EEG and they have these LPDs or PLEDs, as you mentioned, maybe these patients also should be treated with medication. And if you're discharging patients on anti-seizure medication, in patients with acute brain injury, all medications have side effect. And if they need not be on them, it is important that the discharging team feel the responsibility to make some sort of follow-up with a physician who will be comfortable assessing their need for long-term seizure medication, and then discontinuing if not needed.

So that's my important message, because otherwise patient remain on medication. Nobody wants to touch them, especially primary care physicians. And I understand the concern that what if the patient has a seizure. But as you alluded to in the beginning, around 20% of them, 15, 20% of them develop epilepsy, and those also depends on what kind of underlying factors they are. So majority would not. So they should not be on medication for a long period of time.

Glen Stevens, DO, PhD:

Yeah. The other commercial, I guess I'll put out there, because it affects the epileptologist, but the others looking after these patients, and that is there's a lot of new drugs that come out, and patients get put on these new medications that don't have a generic equivalent, and we're not going to get into a generic versus brand name discussion here, but the patient goes home and then they go to get their first refill and they go, "Well, I have a $600 bill and I can't afford that. What do I do now?" And they're on this medication. It makes it complicated. Any thoughts about that?

Vineet Punia, MD:

It definitely does. And then sometimes I get a call even when I have not seen the patient yet in my PASS Clinic, because they're struggling with this issue. Generally, my experience has been that our neurology colleagues on inpatient side have been cognizant about this and have tried to in fact switch medications. The good news is one of the medications that used to be a really high burden or costwise, recently became generic. I would not name. The medication recently became generic and that hopefully, and it's becoming available in pharmacies. So that should really help us out.

But that is always a conversation. In fact, I had a conversation today with a patient's daughter. They're struggling with one of the medication. So that's why even if there are certain medications that we really have to start or ended up starting, it is good to get these patients in touch with the specialists, the epilepsy physicians who can make these changes while safely monitoring the patients. So that's why hooking them up with the specialists at the time of discharge, if they are patients on these medication, will really helpful.

Glen Stevens, DO, PhD:

Well, Vineet, this is great. I learned a lot. That's what I like most about doing these, is that I always learn something new from you and the other specialists that come in. So thanks for joining us today, and best of luck and good luck with your abstract.

Vineet Punia, MD:

Thank you. Thanks a lot. Thanks for having me.

Conclusion: This concludes this episode of Neuro Pathways. You can find additional podcast episodes on our website, clevelandclinic.org/neuropodcast, or subscribe to the podcast on iTunes, Google Play, Spotify, or wherever you get your podcasts. And don't forget, you can access real-time updates from experts in Cleveland Clinic's Neurological Institute on our Consult QD website. That's consultqd.clevelandclinic.org/neuro, or follow us on Twitter @CleClinicMD, all one word. And thank you for listening.