Challenges in Monitoring and Detecting Deterioration in Neuro ICU Patients

In neurocritical care, the initial evaluation is often fast paced, and assessment and management go hand in hand. History, clinical examination, and workup should be obtained while considering therapeutic implications and the need for lifesaving interventions.

In this episode, Aaron Berkowitz, MD, PhD FAAN, speaks with Sarah Wahlster, MD, an author of the article “The Neurocritical Care Examination and Workup,” in the Continuum June 2024 Neurocritical Care issue.

Dr. Berkowitz is a Continuum® Audio interviewer and a professor of clinical neurology at the University of California, San Francisco

Dr. Wahlster is an associate professor of neurology in the departments of neurology, neurological surgery, and anesthesiology and pain medicine at Harborview Medical Center, University of Washington in Seattle, Washington.

Additional Resources

Read the article: The Neurocritical Care Examination and Workup

Subscribe to Continuum: shop.lww.com/Continuum

Earn CME (available only to AAN members): continpub.com/AudioCME

Continuum® Aloud (verbatim audio-book style recordings of articles available only to Continuum® subscribers): continpub.com/Aloud

More about the American Academy of Neurology: aan.com

Social Media

facebook.com/continuumcme

@ContinuumAAN

Host: @AaronLBerkowitz

Guest: @SWahlster

Full Episode Transcript

Sarah Wahlster, MD

Dr Jones: This is Dr Lyell Jones, Editor-in-Chief of Continuum, the premier topic-based neurology clinical review and CME journal from the American Academy of Neurology. Thank you for joining us on Continuum Audio, a companion podcast to the journal. Continuum Audio features conversations with the guest editors and authors of Continuum, who are the leading experts in their fields. Subscribers to the Continuum journal can read the full article or listen to verbatim recordings of the article by clicking on the link in the Show Notes. Subscribers also have access to exclusive audio content not featured on the podcast. As an ad-free journal entirely supported by subscriptions, if you're not already a subscriber, we encourage you to become one. For more information on subscribing, please visit the link in the Show Notes. AAN members: stay tuned after the episode to hear how you can get CME for listening. 

Dr Berkowitz: This is Dr Aaron Berkowitz, and today I'm interviewing Dr Sarah Wahlster about her article on examination and workup of the neurocritical care patient, which is part of the June 2024 Continuum issue on neurocritical care. Welcome to the podcast, Dr Wahlster. Can you please introduce yourself to the audience?

Dr Wahlster: Thank you very much, Aaron. I'm Sarah Wahlster. I'm a neurologist and neurontensivist at Harborview Medical Center at the University of Washington.

Dr Berkowitz: Well, Sarah and I know each other for many, many years. Sarah was my senior resident at Mass General and Brigham and Women's Hospital. Actually, Sarah was at my interview dinner for that program, and I remember meeting her and thinking, “If such brilliant, kind, talented people are in this program, I should try to see if I can find my way here so I can learn from them.” So, I learned a lot from Sarah as a resident, I learned a lot from this article, and excited for all of us to learn from Sarah, today, talking about this important topic. So, to start off, let's take a common scenario that we see often. We're called to the emergency room because a patient is found down, unresponsive, and neurology is called to see the patient. So, what's running through your mind? And then, walk us through your approach as you're getting to the bedside and as you're at the bedside.

Dr Wahlster: Yeah, absolutely. This was a fun topic to write about because I think this initial kind of mystery of a patient and the initial approach is something that is one of the puzzles in neurology. And I think, especially if you're thinking about an emergency, the tricky part is that the evaluation and management go hand in hand. The thinking I've adapted as a neurointensivist is really thinking about “column A” (what is likely?) and “column B” (what are must-not-miss things?). It's actually something I learned from Steve Greenberg, who was a mutual mentor of us - but he always talked me through that. There's always things at the back of your head that you just want to rule out. I do think you evaluate the patient having in mind, “What are time-sensitive, critical interventions that this patient might need?” And so, I think that is usually my approach. Those things are usually anything with elevated intracranial pressure: Is the patient at risk of herniating imminently and would need a neurosurgical intervention, such as an EVD or decompression? Is there a neurovascular emergency, such as an acute ischemic stroke, a large-vessel occlusion, a subarachnoid hemorrhage that needs emergent intervention? And then other things you think about are seizures, convulsive/nonconvulsive status, CNS infection, spinal cord compression. But I think, just thinking about these pathologies somewhere and then really approaching the patient by just, very quickly, trying to gather as much possible information through a combination of exam and history.

Dr Berkowitz: Great. So, you're thinking about all these not-to-miss diagnoses that would be life-threatening for the patient and you're getting to the bedside. So, how do you approach the exam? Often, this is a different scenario than usual, where the patient's not going to be able to give us a history or maybe necessarily even participate in the exam, and yet, as you said, the stakes are high to determine if there are neurologic conditions playing into this patient's status. So, how do you approach a patient at the bedside?

Dr Wahlster: So, I think first step in an ICU setting (especially if the patient has a breathing tube) is you think about any confounders (especially sedation or metabolic confounders) - you want to remove as soon as possible, if able. I think as you do the exam, you try to kind of incorporate snippets of the history and really try to see - you know, localize the problem. And also kind of see, you know, what is the time course of the deterioration, what is the time course of the presentation. And that is something I actually learned from you. I know you've always had this framework of “what is it, where is it?” But I think in terms of just a clinical exam, I would look at localizing signs. I think, in the absence of being able to do the full head-to-toe neuro exam and interact with the patient, you really try to look at the brainstem findings. I always look at the eyes right away and look at, I think, just things like, you know, the gaze (how is it aligned? is there deviation? is there a skew? what do the pupils look like? [pupillary reactivity]). I think that's usually often a first step - that I just look at the patient's eyes. I think other objective findings, such as brainstem reflexes and motor responses, are also helpful. And then you just look whether there's any kind of focality in terms of - you know, is there any difference in size? But I think those are kind of the imminent things I look at quickly.

Dr Berkowitz: Fantastic. Most of the time, this evaluation is happening kind of en route to the CT scanner or maybe a CT has already happened. So, let's say you're seeing a patient who's found down, the CT has either happened or you asked for it to happen somewhat quickly after you've done your exam, and let's say it's not particularly revealing early on. What are the sort things on your exam that would then push you to think about an MRI, a lumbar puncture, an EEG? You and I both spend time in large community hospitals, right, where “found down” is one of the most common chief concerns. In many cases, there isn't something to see on the CT or something obvious in the initial labs, and the question always comes up, “Who gets an MRI? Who gets an LP? Who gets an EEG?” - and I'm not sure I have a great framework for this. Obviously, you see focality on your exam, you know you need to look further. But, any factors in the history or exam that, even with a normal CT, raise your suspicion that you need to go further?

Dr Wahlster: It's always a challenge, especially at a community hospital, because some of these patients come in at 1 AM where the EEG is not imminently available. But I think - let's say the CT scan is absolutely normal and doesn't give me a cause, but as an acute concerning deterioration, I think both EEG and LP would cross my mind. MRI I kind of see a little bit as a second-day test. I think there's very rare situation where an acute MRI would inform my imminent management. It's very informative, right, because you can see very small-vessel strokes. We had this patient that actually had this really bad vasculitis and we were able to see the small strokes everywhere on the MRI the day later, or sometimes helps you visualize acute brainstem pathology. But I think, even that - if you rule out a large-vessel occlusion on your CTA, there's brainstem pathology that is not imminently visible on the CT - it's nothing you need to go after. So, I do think the CT is a critical part of that initial eval, and whereas I always admire the neurological subspecialties, such as movements, where you just – like, your exam is everything. I think, to determine these acute time-sensitive interventions, the CT is key. And also, seeing a normal CT makes me a little less worried. You always look at these “four H” (they're big hypodensity, hyperdensity, any shift; is there hydrocephalus or herniation). I think if I don't have an explanation, my mind would imminently jump to seizure or CNS infection, or sometimes both. And I think then I would really kind of - to guide those decisions and whether I want to call in the EEG tech at 2 AM - I would, you know, again, look at the history and exam, see if there's any gaze deviation, tongue biting, incontinence - anything leading up towards seizure. I think, though, even if I didn't have any of those, those would strengthen my suspicion. If I really, absolutely don't have an explanation and the patient off sedation is just absolutely altered, I would still advocate for an EEG and maybe, in the meantime, do a small treatment trial. And I think with CNS infection - obviously, there are patients that are high risk for it - I would try to go back and get history about prodromes and, you know, look at things like the white count, fevers, and all of that. But again, I think if there's such a profound alteration in neurologic exam, there's nothing in the CT, and there's no other explanation, I would tend to do these things up front because, again, you don't want to miss them.

Dr Berkowitz: Yeah, perfect. So many pearls in there, but one I just want to highlight because I'm not sure I've heard the mnemonic - can you tell us the four Hs again of sort of neurologic emergencies on CT?

Dr Wahlster: Yeah. So, it's funny; for ages - I'm actually not sure where that's coming from, and I learned it from one of my fellows, one of our neurocritical care fellows - he's a fantastic teacher and he would teach our EM and anesthesia residents about it and his approach to CT. But yeah, the four H - he was always kind of like, “Look at the CT. Do you see any acute hypodensities, any hyperdensities?” And hypodensities would be involving infarct or edema; hyperdensities would be, most likely, hemorrhage (sometimes calcification or other things). Then, “Do you see hydrocephalus?” (because that needs an intervention). And, “Look at the midline structures and the ventricles.” And then, “Do you see any signs of herniation?” And he would go through the different types of herniation. But I thought that's a very good framework for looking at the “noncon” and just identifying critical pathology that needs some intervention.

Dr Berkowitz: Yeah – so, hypodensity, hyperdensity, herniation, hydrocephalus. That's a good one – the four Hs; fantastic. Okay. So, a point that comes up a few times in your article - which I thought was very helpful to walk through and I'd love to pick your brain about a little bit – is, which patients need to be intubated for a neurologic indication? So, often we do consultations in medical, surgical ICUs; patients are intubated for medical respiratory reasons, but sometimes patients are intubated for neurologic reasons. So, can you walk us through your thinking on how to decide who needs to be intubated for the concern of depressed level of consciousness?

Dr. Wahlster: It's an excellent question, and I think I would bet there's a lot of variation in practice and difference in opinion. There was actually the 2020 ESICM guidelines kind of commented on it, and those are great guidelines in terms of just intubation, mechanical ventilation of patients, and just acknowledging how there is a lack of really strong evidence. I would say the typical mantra (“GCS 8, intubate”) has been proposed in the trauma literature. And at some point, I actually dug into this to look behind the evidence, and there's actually not as much evidence as it's been put forth in guidelines and that kind of surprised me - that was just recently. I was like, “Actually, let me look this up.” I would say I didn't find a ton of strong evidence for it. I would say, as neurologist – you know, I'm amazed because GCS, I think is a - in some ways, a good tool to track things because it's so widely used across the board. But I would say, as neurologists, we all know that it sometimes doesn't account for some sort of nuances; you know, if a patient is aphasic, if a patient has an eyelid-opening apraxia - it can always be a little confounded. I'm amazed that GCS is still so widely used, to be frank. But I would say there is some literature - some school of thought - that maybe just blindly going by that mantra could be harmful or could not be ideal. I would say – I mean, I look at the two kind of functional things: oxygenation and ventilation. I think, in a neuro patient, you always think about airway protection or the decreased level of consciousness being a major issue (What is truly airway protection? Probably a mix of things). Then there's the issue of respiratory centers and respiratory drive - I think those are two issues you think about. But ultimately, if it leads to insufficient oxygenation - hypoxia early on is bad and that's been shown in several neurologic acute brain injuries. I think you also want to think about ventilation, especially if the mental status is poor to the point that the PCO2 elevates, that could also augment an ICP or exacerbate an ICP crisis. Or sometimes, I think there's just dysregulation of ventilation and there's hyperventilation to the point that the PCO2 is so low that I worry about cerebral vasoconstriction. So, I worry about these markers. I think, the oxygenation, I usually just kind of initially track on the sats. Sometimes, if the patient is profoundly altered, I do look at an arterial blood gas. And then there are things like breathing sounds (stridor, stertor [the work of breathing]). And I think something that also makes me have a lower threshold to intubate is if I'm worried and I want to scan, and I'm worried that the patient can't tolerate it - I want an imminent scan to just see why the patient is altered, or seizing, or presenting a certain way.

Dr Berkowitz: All great pearls for how to think through this. Yeah - it's hard to think of hard and fast rules, and you can get to eight on the GCS in many different ways, as you said, some of which may not involve the respiratory mechanics at all. So, that's a helpful way of thinking about it that involves both the mental state, kind of the tracheal apparatus and how it's being managed by the neurologic system, and also the oxygen and carbon dioxide (sort of, respiratory parameters) – so, linking all those together; that's very helpful. And, related question – so, that’s sort of for that patient with central nervous system pathology, who we're thinking about whether they need to be intubated for a primary neurologic indication. What about from the acute neuromuscular perspective (so, patients with Guillain-Barré syndrome or myasthenic crisis); how do you think about when to intubate those patients?

Dr Wahlster: Yeah, absolutely - I think that's a really important one. And I think especially in a patient that is rapidly progressing, you always kind of think about that, and you want them in a supervised setting, either the ER or the ICU. I mean, there's some scores - I think there's the EGRIS score; there's some kind of models that predict it. I would say, the factors within that model, and based on my experience, often the pace of progression of reflex motor syndrome. I often see things like, kind of, changes in voice. You know, myasthenia, you look at things like head extension, flexion - those are the kind of factors. I would say there's this “20/30/40 rule” about various measures of, like, NIF and vital capacities, which is great. I would say in practice, I sometimes see that sometimes the participation in how the NIF is obtained is a little bit funky, so I wouldn't always blindly go by these numbers but sometimes it's helpful to track them. If you get a reliable kind of sixty and suddenly it drops to twenty, that makes me very concerned. But I would say, in general, it's really a little bit the work of breathing - looking at how the patient looks like. There's also (at some point) ABG abnormalities, but we always say, once those happen, you’re kind of later in the game, so you should really - I think anyone that is in respiratory distress, you should think about it and have a low threshold to do it, and, at a minimum, monitor very closely.

Dr Berkowitz: Yeah, we have those numbers, but so often, our patients who are weak, from a neuromuscular perspective, often have facial and other bulbar weakness and can't make a seal on the device that is used to check these numbers, and it can look very concerning when the patient may not, or can be a little bit difficult to interpret. So, I appreciate you giving us sort of the protocol and then the pearls of the caveats of how to interpret them and going sort of back to basics. So, just looking at the patient at the bedside and how hard they are working to breathe, or how difficult it is for them to clear their secretions from bulbar weakness. Moving on to another topic, you have a really wonderful section in your article on detecting clinical deterioration in patients in the neuro ICU. Many patients in the neuro ICU - for example, due to head trauma or large ischemic stroke or intracerebral hemorrhage, subarachnoid hemorrhage, or status epilepticus - they can't communicate with us to tell us something is getting worse, and they can't (in many cases) participate in the examination. They may be intubated, as you said, sedated or maybe even not sedated, and there's not necessarily much to follow on the exam to begin with if the GCS is very low. So, I'd love to hear your thoughts and your pearls, as someone who rounds in the neuro-ICU almost every day. What are you looking for at the bedside to try to detect sort of covert deterioration, if you will, in patients who already have major neurologic deficits, major neurologic injury or disease that we're aware of? I’m trying to see if there is some type of difference at the bedside that would lead you to be concerned for some underlying change and go back to the scanner or repeat EEG, LP, et cetera.

Dr Wahlster: Yeah. I think that's an excellent question because that's a lot of what we do in the neuro ICU, right? And when you read your Clans, your residency, like, “Ah, QNR neuro checks, [IG1]  ” right? We often do that in many patients. But I think in the right patient, it can really be life or death a matter, and it is the exam that really then drives a whole cascade of changes in management and detects the need for lifesaving procedure. I would say it depends very much on the process and what you anticipate, right? If you have, for example, someone with a large ischemic stroke, large MCA stroke, especially, right, then there's sometimes conversations about doing a surgical procedure before they herniate. But let's say, kind of watch them and are worried that they will, you do worry about uncal herniation, and you pay attention to the pupil, because often, if the inferior division is infarcted, you know, you can see that kind of temporal tickling the uncus already. And so, I think those are patients that I torture with those NPi checks and checking the pupil very vigilantly. I would say, if it's a cerebellar stroke, for example, right, then you think about, you know, hydrocephalus. And often patients with cerebellar stroke - you know, the beauty of it is that if you detect it early, those patients can do so well, but they can die, and will die if they develop hydrocephalus start swelling. But I think, often something I always like to teach trainees is looking at the eye movements in upgaze and downgaze because, often, as the aqueduct, the third ventricle gets compressed and there's pressure on the colliculi – you kind of see vertical gaze get worse. But I would say I think it's always good to know what the process is and then what deterioration would look like. For example, in subarachnoid hemorrhage, where you talk about vasospasm - it's funny - I think a really good, experienced nurse is actually the best tool in this, but they will sometimes come to you and say, “I see this flavor,” and it's actually a constellation of symptoms, especially in the anterior ACA (ACom) aneurysms. You sometimes see patients suddenly, like, making funky jokes or saying really weird things. And then you see that in combination with, sometimes, a sodium drop, a little bit of subfebrile temperature; blood pressure shoot up sometimes, and that is a way the brain is sometimes regulating. But it's often a constellation of things, and I think it depends a little on the process that you're worried about.

Dr Berkowitz: Yeah, that's very helpful. You just gave us some pearls for detecting deterioration related to vasospasm and subarachnoid hemorrhage; some pearls for detecting malignant edema in an MCA stroke or fourth ventricular compression in a large cerebellar stroke. Patients I find often very challenging to get a sense of what's going on and often get scanned over and over and back on EEG, not necessarily find something: patients with large intracerebral hemorrhage (particularly, in my experience, if the thalamus is involved) just can fluctuate a lot, and it's not clear to me actually what the fluctuation is. But you're looking for whether they're developing hydrocephalus from third ventricular compression with a thalamic hemorrhage (probably shouldn't be seizing from the thalamus, but if it's a large hemorrhage and cortical networks are disrupted and it's beyond sort of the subcortical gray matter, or has the hemorrhage expanded or ruptured it into the ventricular system?) And yet, you scan these patients over and over, sometimes, and just see it's the same thalamic hemorrhage and there's some, probably, just fluctuation level of arousal from the thalamic lesion. How do you, as someone who sees a lot of these patients, decide which patients with intracerebral hemorrhage - what are you looking for as far as deterioration? How do you decide who to keep scanning when you’re seeing the same fluctuations? I find it so challenging - I'm curious to hear your perspective.

Dr Wahlster: Yeah, no - that is a very tricky one. I mean, unfortunately, in patients with deeper hemorrhages or deeper lesions - you know, thalamic or then affecting brainstem - I think those are the ones that ultimately don't have good, consistent airway protection and do end up needing a trach, just because there's so much fluctuation. But I agree - it's so tricky, and I don't think I can give a perfect answer. I would say, a little bit I lean on the imaging. And for example - let's say there's a thalamic hemorrhage. We recently actually had a patient - I was on service last week - we had a thalamic hemorrhage with a fair amount of edema on it that was also kind of pressing on the aqueduct and didn't have a lot of IVH, right? But it was, like, from the outside pushing on it and where we ended up getting more scans. And I have to say, that patient actually just did fine and actually got the drain out and didn't need a shunt or anything, and actually never drained. We put an EVD and actually drained very little. So, I think we're still bad at gauging those. But I think, in general, my index of suspicion or threshold to scan would be lower if there was something, like, you know, a lot of IVH associated, if, you know, just kind of push on the aqueduct. It's very hard to say, I think. Sometimes, as you get to know your patients, you can get a little bit of a flavor of what is within normal fluctuation. I think it's probably true for every patient, right? - that there's always some fluctuation within the realm of like, “that's what he does,” and then there's something more profound. Yeah, sorry - I wish I could give a better answer, but I would say it's very tricky and requires experience and, ideally, you really taking the time to examine the patient yourself (ideally, several times). Sometimes, we see the patient - we get really worried. Or the typical thing we see the ICU is that the neurosurgeons walk around at 5 AM and say, like, “She's altered, she's different, she's changed.” And then the nurse will tell you at 8 AM, like, “No, they woke up and they ate their breakfast.” So, I think really working with your nurse and examining the patient yourself and just getting a flavor for what the realm of fluctuation is.

Dr Berkowitz: Yeah - that's helpful to hear how challenging it is, even for a neurocritical care expert. I'm often taking care of these patients when they come out of the ICU and I'm thinking, “Am I scanning these patients too much?” Because I just don't sort of see the initial stage, and then, you know, you realize, “If I'm concerned and this is not fitting, then I should get a CT scan,” and sometimes you can't sort it out of the bedside. So, far from apologizing for your answer, it's reassuring, right, that sometimes you really can't tell at the bedside, as much as we value our exam. And the stakes are quite high if this patient’s developed intraventricular hemorrhage or hydrocephalus, and these would change the management. Sometimes you have these patients the first few days in the ICU (for us, when they come out of the ICU) are getting scanned more often than you would like to. But then you get a sense of, “Oh, yeah - these times of day, they're hard to arouse,” or, “They're hard to arouse, but they are arousable this way,” and then, “When they are aroused, this is what they can do, and that's kind of what we saw yesterday.” And yet, as you said, if anyone on the team (the resident, the nurse, the student, our neurosurgery colleague) says, “I don't think this is how they were yesterday,” then, very low threshold to just go back and get a CT and make sure we're not missing something.

Dr. Wahlster: Exactly. Yeah. I would say the other thing is also certain time intervals, right? If I'm seeing a patient that may be in vasospasm kind of around the days seven to ten, for the first fourteen day, I would be a little bit more nervous. Or with swelling - acute ischemic stroke says that could peak swelling, when knowing which [IG2]  , I would just be more anxious or have a lower threshold to scan. Yeah.

Dr Berkowitz: Yeah - very helpful. Well, thank you so much for joining me today on Continuum Audio.

Dr Wahlster: Thank you very much, Aaron.

Dr Berkowitz: Again, today we've been interviewing Dr Sarah Wahlster, whose article, “Examination and Workup of the Neurocritical Care Patient” appears in the most recent issue of Continuum, on neurocritical care. Be sure to check out Continuum Audio episodes from this and other issues. And thank you so much to our listeners for joining us today.

Dr Monteith: This is Dr Teshamae Monteith, Associate Editor of Continuum Audio. If you've enjoyed this episode, you'll love the journal, which is full of in-depth and clinically relevant information important for neurology practice. And right now, during our Spring Special, all subscriptions are 15% off. Go to Continpub.com/Spring2024 or use the link in the episode notes to learn more and take advantage of this great discount. This offer ends June 30, 2024. AAN members: go to the link in the episode notes and complete the evaluation to get CME. Thank you for listening to Continuum Audio.