All right, good afternoon, everybody. It's great to see so many people out. Let me rephrase that statement. It's great to see so many people, just in general. I hope you've had a great conference, as I have. It's been wonderful to get back with folks, to see a lot of people that I haven't had the opportunity to talk to in a while. So as our esteemed moderator mentioned, my name is Sean Rowe. I'm an associate professor of clinical pharmacy and translational science at UT College of Pharmacy. My practice is in neurocritical care, and the reason that I'm here talking with you today is I'm also the co-chair of the Neurocritical Care Guideline on the use of anti-seizure drugs for seizure prophylaxis. So as the title kind of suggests, treating something that isn't there, which is something that's pretty common in most of our ICUs, that this does occur. So a little bit about me. This is for the recorded session. We'll skip over that. Disclosures. I've got no relevant financial disclosures to go along with this. As I've already mentioned, I am the guideline co-chair for this actual guideline that will be coming out in the next little bit. So I certainly do have some vested interest in the way that that is going to be perceived on that. What am I going to be able to share today? So first things first, the guideline is not 100% complete yet. So unfortunately, if you came in here hoping to see a completed 100% recommendations about what to do in this type of situation, my apologies. What I am going to be able to present to you today is what our guideline process looks like as well as the quantitative evaluation that we've done. We've done a lot of meta-analyses and that's what I'm going to be talking about today. That's what the most of this will be focused on. First things first, though, I want to recognize my co-chair, Jim Frontera, as well as all the committee members that have worked together on this for quite a while now. The picture down in the lower left-hand corner is kind of what we all looked like as we started this process. I've not got a picture of what we look like currently, though. I can tell you that's probably not quite as excited and enthusiastic as this is, but we have worked very well together and it's been a great committee to work with. Just a quick overview of what the NCS guideline process is. We follow the grade process. So if you're not familiar with grade, the little figure at the very bottom of this kind of gives you an idea of what this is like. First off, we develop a set of clinical questions that we're interested in. We do a systematic review based on what those questions are and what the outcomes are that we're interested in, and then we synthesize all the evidence that's involved with it and come up with a certainty of evidence. We truly are looking at how well can we take the data that we have gained from this literature and truly apply it to these questions that we've asked. In addition, we'll increase the grade of that or decrease the grade of that based on the risk of bias that we see in other issues that are going on and finally come up with a final recommendation and certainty of evidence. So that's the grade process. Trust me, if there are grade folks in here, I apologize if I butchered that, but that's the gist of what goes on with this. So this multidisciplinary panel of experts was assembled in July of 2019. As most of us know, what happened not too long after that was COVID, right? So we've got delayed a little bit by all the things that have happened with COVID, but we have been consistently working on this since that time. Our panel consists of experts from neurology, nursing, neurosurgery, and pharmacy, and it's been a great group to bring together to make sure that we're looking at all the things that we want to. The big thing to us was that as we were writing this document, as we're putting this document together is that we wanted to make sure that this was something that could be usable by clinicians at the bedside. We wanted to make sure that the recommendations that we were able to give helped to guide therapy in an appropriate way and in a meaningful way. So the disease states with that in mind that we thought about was that what do we mostly see in our neurocritical care population that there's some questions about the best way to use seizure prophylaxis. The four main disease states that we came up with were traumatic brain injury, aneurysmal subarachnoid hemorrhage, intercerebral hemorrhage, and supratentorial surgery. Most of those are pretty self-explanatory. The supratentorial surgery, though, I'll put the caveat on there, is that we really didn't want to include those patients who had some form of malignancy that required surgery. This was purely looking at someone who required surgery for a reason other than a malignancy. So that's the one that's kind of giving you the caveat there. And across the board, I think that these four disease states were what interested the committee most as we were looking at this. So we've got our four disease states. Then what were the questions that we wanted to come up? And these are formatted in the PICO type format for you so you kind of get an idea of what we were looking at. This was mostly focused on the adult patient without a history of seizures. And we had three comparisons that we were really interested in. Number one, we wanted to see if there was a difference between having an anti-seizure drug on board versus not having an anti-seizure drug on board. Secondarily, we wanted to look at old school versus new school, right? And so, unfortunately, when you're trying to synthesize data, you can't take in a ton of different medications. And so we chose to focus on phenytoin versus levotracetam as our markers for the new school versus old school medications. We could have a discussion about that probably for quite a while about the appropriateness of that, but that's what we chose to look at was levotracetam versus phenytoin. And then finally, what we wanted to compare was going to be duration of therapy. We were looking at long versus short duration of therapy and whether or not that made any difference in the outcomes. So four disease states, three questions, and what about the outcomes? Well, we had five of those. You can start to see and do the calculus of this, hopefully, about how large of a beast this got. It was quickly turning into as we started looking at all the things that we wanted to look at with this. First things first, we wanted to look at early seizures. So in this situation, based on the literature that we had, we were able to define that as less than 14 days. Now granted, I know a lot of the literature when we're looking at traumatic brain injury, the Timken articles initially looked at that, defined that as something less than seven days. And so we really had to look at this across all of our disease states, and there was a lot of back and forth between the committee members to make that definition. Late seizures were everybody else, so greater than 14 days. Adverse events. I'm a pharmacist. One of the things that I commonly look for is the safety of the medications that we're going to be using. So I felt that this was a critically important thing that we were looking at, especially in this situation, right? We're looking at using and treating and preventing something that's not there already. And inherently, all of these medications have at least some risk that's going to be associated with it. So adverse events were important. And then the last two, mortality and functional outcome. These were considered important outcomes to us, and the great process, we rank these. We say, okay, is this something that is absolutely pertinent to this that we have to have and make some comment on? Three out of these fives were critical, so we have to make some sort of comment on those. And the following, the bottom two, were important. Part of the reasoning behind that was the lack of data. Mortality may or may not be affected by this, but we felt that it was an important thing to look at. And functional outcome would be wonderful. We would love to be able to say that there's some change in the functional outcome with what's going on with this. But that leads us down the road of whether or not we have enough data. So the methods for this, we used a medical librarian to be able to search Medline, Embase, Emcare in the Cochrane databases. Pre-standard set, I could put up all of these search terms if you guys want to. No? Okay. Okay. Just checking, just to make sure. There was a lot. A lot of pages of search terms that we used to be able to go through this. That resulted in about 2,600 or 2,700 articles that we screened and trimmed it down to about 171. Of those 171, only a small percentage actually had the exact outcome we were looking at and the exact comparisons that we were looking at in order to be able to include it in a meta-analysis. So what I'm presenting to you guys today is that meta-analysis or those meta-analyses that we looked at. So our meta-analysis was conducted by a statistician from McMaster University. We stratified these based on randomized controlled trials and non-randomized controlled trials, used random effects models as well as mental hand-cell statistics to be able to stratify those. From a evaluation of risk of bias for a randomized controlled trial, we used the ROB2 method as well as the ROMIDS-I for our non-randomized controlled trials. And then finally, we used GradePro to be able, as a committee, to come together and put all of this information into a readable format that gave us a good evaluation of what's going on. Okay. Let's actually get into some of the data. There's a lot of information on the screen. And this, again, is going to be the quantitative evaluation of what's going on. Traumatic brain injury was the first one that we really started looking at. And it included studies from anywhere from 3 up to 10 studies and up to about 7,500 patients. Across the board, regardless of what we were looking at, there was no statistical difference in prevention of early, late seizures, mortality, so on and so forth, or even adverse events. This is something that, as a committee, we were all kind of surprised by, because of all the data that has been put out previous to this, going way back into the 1990s with all of the seminal papers that were done on treatment and prevention of seizures in patients with traumatic brain injury. And what the committee came to conclude from this was that there's probably some more information and not probably, there is more information in those other studies that we couldn't include, that we couldn't include in the total meta-analysis that we need to go back and look at. So that's where we're at now. We feel like some of this may be a type 2 error. And as we are actually looking into what our final recommendations are going to be, we're going to take into consideration the other articles that were not able to be included in the meta-analyses. From an intracerebral hemorrhage standpoint, when we start looking at that, there are a few things. So, first things first, the one thing that we saw statistically significant difference in was anti-seizure drug versus no anti-seizure drug and a functional outcome. So having a modified Rankin scale of 4 to 6 at 90 days, which is pretty bad. That's going to be someone who's no longer functionally independent. That's a difficult situation. And what we found was that those folks who got an anti-seizure drug actually had a worse odds of having that outcome. And this is whenever you pull together four different studies. You can see that the odds ratio was 1.2, confidence interval didn't cross over 1, and it was 1.1 to 1.4 was the way that it worked out. This was the one thing that certainly did stand out to us. All the other things really didn't have a statistically significant difference. One of the questions that we had as a committee and as we were looking at this was, why? Why is that the case? How could this be? Well, one of the things that we had to take into consideration, these were all retrospective studies that were done there, and there's probably some selection bias that was occurring with this. In other words, those patients who were the sickest and those who were least likely to have a good outcome anyway were probably the ones that were developing or getting some form of anti-seizure drug to start out with. Again, as a committee we're coming through and we're going to be looking at the data that was not able to be included in that meta-analysis as well. Supertentorial surgery. Well, when we look at this one, what we find is that first, before I talk about the red box, let me point out this. When we look at anti-seizure drug versus no anti-seizure drug, we didn't see that there was any difference in early seizures. However, whenever you look at levotiracetam versus phenytoin, there was a improvement, pretty big improvement as far as levotiracetam is concerned for early seizures. These four studies that only included 600 patients, one thing that we noticed was that this was very periprocedural, right? This was giving a very short course of using levotiracetam or phenytoin in this situation. So interesting finding that we had from that and more to be coming out on it. And then finally, from a disease state standpoint, we had aneurysmal subarachnoid hemorrhage that we were looking at. A couple of different things that we saw here, interestingly enough, was this finding. Again, if you're looking at no drug versus some drug, there was no difference in early seizure. However, if you're comparing levotiracetam to phenytoin, what we saw was that if you got levotiracetam, folks did worse. There was more chance of having an early seizure. At first, you may look at this and think, well, that's kind of a condemnation of levotiracetam, right? I think a couple of things that we wanted to point out as we were talking about this was that even though there was four studies that evaluate this, only two of the studies actually had any events that occurred in them. And in those two studies, the most commonly used dosing regimen for this was 500 milligrams twice a day. So it really does beg the question and it makes us wonder whether or not there's some confounding due to the dosing that was used. Did people have a non-therapeutic type dose of levotiracetam? And it's certainly a question that the committee has not necessarily addressed specifically as it relates to looking at the literature that goes on with it, because it's a difficult thing to evaluate. But certainly from a good practice statement standpoint, we do feel that if you're going to use a drug, that you need to use it at the appropriate good dose for this. All right. So let's summarize that. When we look at the actual quantitative evidence that we have, it's pretty limited as compared to the body of evidence that is out there. If you have looking at change in outcomes, there's a higher odds of having a modified Regan scale of 4 to 6 in patients with ICH that received an anti-seizure drug. Again, probably some confounding going on there because it was retrospective studies and likely some selection bias that occurs with that. However, that's what our meta-analysis was finding. Whenever you look at comparing phenytoin to levotiracetam in aneurysmal subarachnoid hemorrhage patients, it's difficult to say. But again, the meta-analysis did show us that levotiracetam didn't do quite as well. However, there's quite a bit of confounding due to dosing that was associated with that. So what are our next steps? Well, the committee is currently meeting. We are actually in the process of writing. This initially started out as a single document we wanted to put together. Once we got into this and we saw that it was quite a large undertaking and the product was going to be very large, we wanted to make sure that we could get it out. So what we're actually doing is we're working with the Neurocritical Care Society journal to actually get this published in a series of articles that's going to be coming out throughout this year. So it will be published out this year. We're developing our final recommendations that are consistent with GRADE, and we'll have it out throughout the summer. All right, and that's what I have for you guys. Thank you very much.

Neurocritical Care Guidelines

Anti-seizure drugs

Seizure prophylaxis

Disease states

Meta-analyses