Okay, so we have a very nice long title, but I simplified it as I'm gonna try to do with a lot of things to what is neuroprognostication. I would like disclosure, so please see me afterwards if you'd like to give me some. You can read the learning objectives. And so we're gonna get right into it, which is what is neuroprognostication? And it's really simple. It's somebody asking you to predict the future. So nothing's easier than that, right? And so the patients that most often they ask us to do this on are comatose. And that could mean a lot of things. Essentially, it's saying it's an unresponsive patient. So if you go to the patient and you provide a stimuli, there's no meaningful response. And this can happen for a variety of reasons. It could be something metabolic, it could be something structural, could be cardiac arrest, could be drug-related, which makes it much more difficult to predict the future, right? And so this becomes really, really important because it guides where you're going with your treatment. So I'd love to say that we all approach this the exact same way, but as we're gonna see when we probably review those previous questions, we all approach this wildly differently. Reading a study from 2022 that's cited on the slide here, we find that physicians, you guys are actually very pessimistic. But what's more surprising is nursing, you are worse. And I can tell you as somebody that's practiced for over 20 years, I thought every ICH patient was gonna die. It's only been within the last couple of years that I kind of don't think that, and I might convince you of that by the end of this. We also found that people that were new to practice were actually more pessimistic than experienced providers, which you might think it would be the opposite of that, right, that people that have been doing this forever would be jaded, but no, it's people that have just learned all of the facts, I think everyone's gonna die. In terms of acute care, also the most pessimistic. So people that see patients on a longitudinal scale are more likely to believe that there's a chance for meaningful survivorship. What also is interesting is that you can actually de-bias somebody by providing evidence-based practice, which hopefully this will do a little bit of. And you can see the consequences, which is that you withdraw life support and the patient dies, or you continue to utilize all of your resources for an infinite period of time, which is great for team morale, right? So the things that you're gonna learn about, and a lot of this you're gonna hear has been echoed throughout the day, is most guidelines for most of the pathologies we'll review are saying to wait at least 72 hours and see what you have. Also, there's the balance between the heuristic and the analytical approach. And so what this means is heurism is when you start to develop these mental shortcuts. And so you do something because it worked before, because you've seen it before. And so that's kind of what you believe is gonna happen, which also goes with this self-fulfilling prophecy that you've heard us talk about in the past. So I will say maybe 15 years ago, all ICHs that I saw, they all died. So what do I think? All ICHs die. And so that's the advice that I'm gonna give the family. So we withdraw and what happens? Well, the patient dies, so I was right, right? But that might not necessarily need to be the case. So one of the important foundations is getting rid of anything that might influence what you're seeing when you're looking at the patient. This is kind of easy stuff. You guys all know this. If somebody's gorked on propofol, can you give a good exam and say how the patient's gonna do? No, so you gotta remove all of these other factors, whether it's medications, whether it's hemodynamic instability. All of these are things that you can control, everything that you can control. Can you always control all these factors? No, and so that's when you might have to think about what ancillary testing there is. And then in terms of the impact of these confounders on your prognosis, I have found that critical illness in general tends to be a barometer for what's going on in the brain. So if you're unstable in any way, the brain's the first thing that's gonna show it. So somebody that's septic, they might not present with a fever. They might present altered. And so whenever somebody has any type of brain injury, you have to look at all of these other things that are gonna influence it. So often that's the first thing that you're gonna see. So we talked a little bit about ancillary testing, and you're gonna find in more recent guidelines that they say that there are some reliable predictors versus unreliable predictors, but nothing is a slam dunk really, as much as seeing the clinical exam and seeing the trajectory. So as was mentioned in previous talks, it's gonna be watching the trends for these patients. There are some things that are more likely to align with poor prognosis. So somebody that has diffuse edema and lack of gray-white differentiation or huge DWI changes are more likely to have a poor outcome than somebody that doesn't have a clear large structural damage. With EEG, if somebody doesn't have any type of reactivity to external stimuli, they're less likely to have a good outcome. Somebody that's just an isoelectric suppression. But again, these are things that you're gonna have to follow over time because it can change as you remove those confounders. And then something else you'll hear us talk quite a bit about, which is not widely used, but probably could and should be, is SSEP revoked potentials. And what that is, is you provide an external stimuli, usually you use the wrist, and you provide a current, and you wait to see if it creates a reaction along your primary somatosensory cortex. But as much as we maybe could use it in the ICU, it's not commonly adopted. So cardiac arrest, when people talk about neuroprognostication, is one of the first things that come to mind. And there've been quite a few papers that have come out on this within the last few years. So 80% of patients that have out of the hospital cardiac arrest, they come in comatose, 80%. So most of the patients that you see come in like this. Three days later at 72 hours, only 50% of those are still comatose. So even without doing anything that has anything to do with neuro, already you're taking patients from coma to awake. And that's without additional special management. What we used to know even five years ago, isn't true anymore. And you'll see even in some of the more recent guidelines, it's already changed. So we used to think we needed to keep all patients hypothermic. Well, now we know we just don't want them to be febrile. We used to think that how long it took to get return of spontaneous circulation was associated with outcomes. It's not, depends on the quality of your CPR. And so these are things that have started to influence. This is gonna be a common theme you're gonna see with every pathology. So I'm taking flying lessons and they always say no matter what's going on, the first thing that you do is fly the airplane. It's kind of the same with patient care. The first thing that you do is take care of the patient. You don't worry about all these other things. So everything is supportive care. And you're gonna do this in all of your patients, regardless of the outside noise and other conversations. So we're gonna start by visiting Europe and we're gonna call it ancient Europe because these guidelines still recommend that you keep patients at a temperature of 32 to 36 degrees, which we've now debunked and is absent in the 2023 guidelines. But a lot of the other themes are gonna be common, which is that at 72 hours, they're more likely to have a poor prognosis. And these are things that are gonna make sense if they don't have a pupillary response, if they have absent N20 evoked potential. So if you try to apply that shock and it never gets there, if they have no electrical activity on their EEG. So these are all things that kind of make sense, right? You've given time for these to come back and they don't, and they're more likely to have a poor outcome. At that time, they were still looking at NSE and they were still recommending it as something that might correlate with a poor outcome. And that's something else that's gone away with later guidelines. So we've had two more sets of guidelines come out in the last year. The Canadians, which is always fun. And it's actually not hugely different. They do mention waiting the 72 hours and that you're gonna look at the trajectory of the exam. So you wanna see somebody with an improving exam. If they're comatose on arrival, they're comatose 72 hours later, and you've managed all of the variables, they're more likely to have a poor outcome. And then they talk about the isoelectric EEG, the loss of gray-white differentiation, the absent N20s. They no longer recommend NSE. They no longer recommend targeted temperature management for hypothermia. So NCS, and you're gonna see, we're gonna talk about NCS a lot because they're putting out a whole series of neuroprognostication guidelines. And so this one that came out last year, not only kind of reiterates the things that we all have common sense and we all know are bad, but they actually look at the other side too, which is what are good outcomes. And so somebody that has an improving exam, if you actually see the structural lesion or the DWI changes aren't that huge, or the CT doesn't look that bad, that maybe there's some reactivity on the EEG, even in the presence of a poor exam, they want you to continue to care. And that's gonna be something that we're gonna continue to talk about with all of these pathologies, and that these are patients that need more time. And that just reiterates that in a nice table format. So ischemic stroke. And again, you're gonna hear some common themes here. What's important to know is stroke is becoming more and more common, and it has probably nothing to do with our lifestyles, but maybe a little bit. And so in the last 20 years, they've said stroke is gonna increase by 50%. So that's 12 million new strokes a year. Worldwide, one in every four people will have a stroke in their lifetime. So if you look around the room, that's a little bit scary, right? So we need to know how to manage it. And my goal is to tell you that the management is not to withdraw on every patient within the first 24 hours. I've got one, okay, thank you. All right. And so we also have heard a little about the changes in management. And I was telling people before we started talking that just in my lifetime, being at the bedside, we went from turning patients Q2, easiest patients in the world to take care of, to TPA, to thrombectomy, to patients that are in and out of the ICU within 24 hours. But there are a lot of different things that play into this. Their age, their comorbidities, where is the stroke happening? Something that's affecting the brainstem is a lot different than something affecting the right MCA or the left MCA. Matters if you're left-handed or right-handed, right? All of these things matter. And what your quality of life was before, what the things that are important to you. And we've seen that since the changes with how aggressive we are with endovascular therapy have occurred, we've really been able to decrease mortality. But that isn't necessarily saying anything about morbidity and disability. And that's an important thing to factor in as we're having these conversations. And so I do mention that predicting these outcomes are very difficult, and we'll show that in just a little bit. You've heard us talk about the modified Rankin scale. How many of you are super familiar with it? So like half the people in the room are familiar. So we have a tendency when we do any type of research to dichotomize these scales to either good outcome or bad outcome. And you can see we've even changed that. And so it's pretty universal that we've said zero to two is a pretty good outcome. So if you could go to the grocery store before on your own, you can probably still go to the grocery store on your own. Three is where it gets a little bit dicey that you need a little bit more help, but you're still able to ambulate. But actually I think that should say with assistance. And then once you get to four, that's somebody that's not able to have any ambulation. Six is pretty obviously probably bad, but the like terrible neuro joke is like six is dead, five is wish you were, because you're not really doing anything and you're completely dependent upon all care. And so the reason that where you split this scale matters is they've actually studied within the last year or two what your quality of life score looks like. And so that's really what is important. Not whether you can walk and skip or whatever, it's do you feel like you still have a good quality of life? And they actually found the difference between two to three is negligible. So most people that are a modified ranking of three at discharge are pretty happy with how their life is. And so all of those other studies that say that a three is a bad outcome, probably isn't true when you actually talk to patients. So I talked about the ability of people to predict the future. This is another study from 2022. And they looked at, it was prospectively done and they looked at 299 patients and this was neuro interventionalists. And I'm not trying to pick on any of the neuro interventionalists in the room, but they had to predict what a good outcome was. And so they divided the scale, not into two, they divided it into three. And they said, good outcome is a modified ranking of zero to two. And then you had three to four is like not a great outcome and five to six is terrible, right? They were only able to predict correctly less than half the time. So 44% of the time they were correct. Well, now that you remember one out of four of you guys are gonna have a stroke, how do you like those odds of somebody predicting your future and whether or not they're gonna intervene? This is recent. This is within the last year that this has come out. So again, this is scary. And this is why we need to change the narrative that's currently going on. This has also been mentioned, but what we didn't really talk about was that with this more aggressive therapy, there are better functional outcomes. So select two moved patients to modified rank in zero to two and zero to three. And the number to treat to do that was only seven and then five to get to that zero to three, which again is considered good quality of life by the patients themselves. They also took those wish you were dead patients and divided it in half. So if your modified ranking was five, decreased by half, angel aspect did the same thing. So we're seeing patients that have large established strokes have better functional outcomes if we intervene, which is the complete opposite of what that satin study was saying we were going to do by the people that are performing the procedures. So luckily we have all of these scales that can help us predict the outcome. None of them keep in mind any of the other things that we just talked about that have changed in the last two or three years. So none of us use these. Anybody that works neuro, have you ever used any of these scales to predict the future? No, I hadn't even heard of half of these, honestly. I was shocked that we had this money, but they're completely useless with how quickly the technology is changing and how much we're actually moving the scale the opposite direction. So you have to keep in mind with all of these prognostication tools to put it into context and that maybe for something that's rapidly moving, that it's not the best to use something like this. And so hopefully that will help with the one that had a little more ambiguity with the answers. So hemorrhagic stroke. So as again, we've alluded to, these patients typically don't do that well. When we're gonna talk about ICH here, we're gonna kind of include everything that's non-traumatic. So this could be anything from a subdural, I'm gonna throw aneurysmal subarachnoid hemorrhage in here, intraparenchymal hemorrhage, IVH, and we're just gonna do a catch-all because all of these patients historically have high mortality. What's good is in general since the 70s, we've actually seen worldwide a slight decrease in mortality, not really a significant decrease in disability. It's not true in the United States though. We've actually seen an increase in the incidence of ICH and really haven't moved the mark much on mortality or disability. So guess what the most common cause of death in these patients is? It's us. We remove all support and the patients die. In fact, about a quarter of these patients, that's exactly what happens. And we're able to justify it as we always do. We say, oh, it's a big hemorrhage, they're probably not gonna do well. And so there starts that self-fulfilling prophecy. And a lot of times you'll see people hide behind the ICH score, which is the most commonly used score when you're talking about ICHs. I think it's actually a joint commission requirement that you have to document this when these patients come in. And it was never meant to be a prognostication tool. It was meant to be a severity metric. And so you can see it's pretty simple scale. We can do this. It's big bleed, it's a little bleed. They're old, they're not old. It's in the brainstem, it's not. And because of that, it is associated with 30-day mortality in the one-year modified Rankin scale. But the people that actually created this scale went back and they did a study and prospectively observed you said you could not give these patients a DNR. They said, if you're gonna be part of this, you can't make these patients DNR. And they actually found five days later, even if you had a high CH score, you actually had decreased mortality without a significant increase in disability. So that right there is telling you that most of what we're seeing in terms of mortality is that self-fulfilling prophecy. And so we see guidelines starting to reflect this change. And so again, two sets of guidelines. We have AHA, ASA, and then NCS. And all of them are telling you the same thing. Don't have these DNR discussions for at least two days is what AHA say. I think NCS actually goes back to that 72 hours. So it's kind of a common number to keep in your head. And that there's no scale that really has anything to do with how the patient's gonna do. And when we talk about TBI in just a few minutes, I want you to think about that and then go back and think about this and how maybe we're really skewing the mortality in ICH patients. Now, aneurysmal subarachnoid hemorrhage isn't quite as clear cut because the sequelae that follows is quite different. But I'll tell you, the guidelines say the same thing. Don't make these patients DNR within the first 72 hours. Give them a chance. A lot of patients that come in GCS3, giant bleeds, they have a reasonable functional outcome when you follow through with care for them. Are there patients that won't have a great outcome? Yeah. So the question is, do you give them a chance? We're telling you to please give them a chance. Now, if they come in brain dead, it's a little bit different, but that has to be a sure thing too. And we're gonna talk about why maybe that's not even as clear as we thought. So TBI, it sounds like you guys heard a lot about TBI earlier. If you go back like almost 20 years ago to some early statistics for TBI, and we're talking like MRC crash in 2008, this looked at 10,000 TBI patients that presented with a GCS of less than 15. And it's not necessarily accounting for polytrauma, which we know is present in a lot of these patients. In two weeks, 20% of these patients were dead. In six months, 25% of these patients were dead. And at six months, 33% were either dead or severely disabled. Well, I'm gonna tell you all of that is different too. So TBI, again, we're gonna kind of cover everybody and just include everybody. So this is everybody from the contusion to the traumatic subarachnoid or subdural, epidurals, axonal injuries. We're just gonna take everybody and loop them in. We know that TBI accounts for 30% of all injury deaths. And again, the most important predictor of outcome in these patients, it's us, withdrawal of life support. So you're hearing a theme here, right? That our biggest enemy in any neurodisease is ourselves. And so keep that in mind when you're seeing these patients. So in guidelines that are forthcoming, that I got a little sneak peek at, we do have a couple models that are relatively reliable prediction models. And that's gonna be crash and impact, looking at 14 day death and severe disability at six months. This includes things like GCS, pupils, if they have an extracranial injury, looking at CT findings and what their comorbid conditions are, if they're hypotensive or hypoxic. Impacted a little bit better than crash at looking at how they were gonna do in six months and whether it was a favorable versus unfavorable outcome. So that's something that is forthcoming to consider when you're doing these evaluations. But really what was exciting, the more that you read about this, and you heard a little bit about rescue ICP in the 400 something patients that it looked at, was they did a secondary analysis. So if we go back and remember what we learned about rescue ICP, was that they took patients that had refractory ICP crisis. And they said, okay, we're gonna take the skull of this one, we're gonna do maximal medical therapy, whether it's a pentobarb coma and just throw everything we have at this other guy. And they found that there was decreased mortality in the patients that got their skull off, even though there was increased incidents of severe disability in those patients. But then they came back and they did a secondary analysis and they looked a year later, and they looked at two years later, and they found that these patients actually continued to improve. So a year later, the patients that got decompression, 30% of those patients got better. That's huge, 30% of those patients got better. And even the patients that didn't have their skull off and had medical management, 14% of those patients got better. And so they continued to improve as long as we continued to treat them. And you hear us talk about these Glasgow outcome scales. So what that's looking at is one is dead, eight is full of recovery and kind of everything in between. And that's gonna be important when we talk about track TBI. So this was another more recent study from 2021. It took 484 patients with moderate to severe TBI. And they looked at two weeks, three months, six months, and 12 months. And again, we dichotomize the scale because it's much easier for us to run statistics on it. And favorable was a Glasgow outcome of four to eight and unfavorable was one to three. And so they took the patients that had severe TBI, not two weeks, 12% had a favorable outcome. At three months, 45% of those patients with severe TBI had a favorable outcome. And at a year, 52% had a favorable outcome. So these are patients that were considered to have severe TBI. Half of them were better a year later. And actually 12 and a half percent of those had a complete full recovery. Then you took the ones with moderate and unsurprisingly, they also got better at a higher percentage. So at two weeks, 41% of those patients had improved, 69% at three months, 75%, so three out of four, a year later, were considered to have a favorable outcome. During that year, nobody had a decline. Nobody went from favorable to unfavorable, which is kind of important that these patients weren't getting worse. What is super interesting and super exciting are patients that were vegetative. So patients that you would look at, and probably most people in this room would assume they weren't gonna do well, 78% of those regained consciousness, with 25% being completely oriented a year later. So these are the patients that come in a GCS of three that have ICP crisis and are completely unresponsive to anything that you do. A year later, we're starting to do better. And they were able to talk to you and make sense. So what are these guidelines gonna tell you to do? Is anybody surprised if I tell you they were going to have you wait 72 hours? Would you be surprised if I told you they're actually gonna recommend that you wait two weeks? So this is gonna be something a little bit different because we did see that these patients were continuing to improve past that 72-hour mark. And so if you think about the pathology of ICH, whether it's traumatic or whether it's something caused by cerebrovascular disease, can we move the scale on that if we were a little bit more patient? It's possible, but nobody's done it because we withdraw on all these patients. So the key points here are gonna be the most important thing that you can do for any of these patients is give them time. That you wanna look for true structural irreversible damage, that you wanna fly the airplane, manage the patient, get rid of all of the confounders and try to make them as stable as possible. And that communication is gonna be super key. Because luckily, the only two things sure in life are death and taxes, right? Nope, that's not true. So brain death even is maybe not as clear. And you'll see when we talk about these guidelines, people say, well, if they come in brain dead, well, that's easy, right? Well, even that's not that easy. We have changes there too. So one, how each hospital is actually doing this is highly variable. Well, that's a little bit surprising because we have guidelines that tell us what to do. And they've actually studied that almost 80% of these physicians that are diagnosing brain death have received all of the adequate training to pronounce brain death. But only one out of four of these same providers are doing it by the guidelines, which doesn't really make a lot of sense to me. In terms of looking at how apnea testing is performed, a lot of times if the patient breathes, what do we do? Well, we get a confirmatory test because we do not believe that. If the patient breathes and they're not dead, that makes sense to me, but we don't follow that. We see portions are omitted or it's performed incorrectly. And then it hasn't been unseen. And you can think about your own experience that somebody has been pronounced brain death. And then maybe there have been signs of functioning hours later that has been concerning to the team. And so this is way more complex than people think. And as technology advances, we don't really have an answer for that either. So these patients with ECMO, it's not clear cut how we're supposed to pronounce brain death in these patients. So I think we rely a lot upon ancillary testing, but we don't really know the answer to that. And then is brain death really irreversible? Well, we're seeing cases in animal models that reanimation is actually a thing. And we're seeing hours later that there's at least cellular function reoccurring. And then there's a whole ethical, we'll just say discussion about what normothermic regional perfusion means in terms of the dead donor rule and how dead is okay. And what role should we have in that? I'm not here to get involved in the ethical discussions, but just bring to mind that some of these things that we think are quite clear, maybe aren't. And so as we're looking at how to handle that, we maybe need to look at redefining what brain death is, and then whether we should include just looking at hypothalamic function and whether we even need to continue to include death by neurologic criteria. And so lastly, what we say really does matter. And I think it's important as we do this to very clearly divulge that, I guess the only thing sure is taxes, because we don't have a crystal ball and we can't 100% say how these patients are going to do unless we withdraw care. And then I can 100% tell you how they're gonna do, but otherwise there's a large degree of uncertainty that exists. And anytime you're gonna do something more aggressive, that's probably a good point to stop and talk. And that anybody in the care team can and should be involved in these conversations. And then you'll get a little bit more information about how to do this in the chapter, particularly if you buy the book. All right, so I think we'll soon be taking questions and potentially lively debate.

neuroprognostication

brain injuries

comatose patients

traumatic brain injuries

life support

brain death diagnosis

healthcare team communication