Holy Trinity of Neurology, what are the studies for the Holy? This may be the first time you're hearing a neurologist not talk about any of those things, okay? All right, so at the end of this talk, I really just want it to be a bite-sized introduction to when you would use non-invasive monitoring tools, how you would interpret them and have an exploration of the diagnostic accuracy of these. So sometimes in patients with acute brain injury, you know, it's very obvious that they need invasive monitoring. And for these patients shown up here, we've got a patient with a subarachnoid hemorrhage, edema, hydrocephalus, and a patient with primary IVH, edema and hydrocephalus as well, very clear that these patients need invasive monitoring. That's, you know, an EVD, right? So we can monitor what the ICPs are doing. We know that by monitoring and diagnosing elevated ICPs, that is our chance as intensivists to actually move the needle, decrease mortality, decrease morbidity, and it's very important to do. But what if you have a patient like this, a 20-year-old with a left parasagittal hematoma, some perihematoma edema, who has a venous sinus thrombosis and the superior sagittal sinus, and they're on anticoagulation? You know, you see that there's some sulcal effacement here. This is not somebody, you know, you're gonna rush to put in some kind of invasive monitoring because they need to be on anticoagulation. Or what if you're practicing in an area where you don't have readily available invasive ICP monitoring? Thank you, chat GPT for these images, by the way. Or what if your patient has a coagulopathy? Or maybe they just have a metabolic condition that's leading to their cerebral edema and really not a clear indication for invasive monitoring. These patients freak you out. And I think I felt, this is actual video of me as a fellow trying to manage these patients. That's Maya Rudolph, actually, who's brilliant and so funny. So that's when these tools of transcranial Doppler and optic nerve sheath diameter by ultrasound really come in handy. And there are many non-invasive tools for ICP monitoring, but these are ones that I like to use because all I need to do is grab my own and go to the bedside. So this is the way you would do this at the bedside for a point of care transcranial Doppler. So shown here is the phased array probe. You have the probe indicator pointing towards the lateral canthus of the eye. And you're incinating over the temporal window. In most people, the temporal windows are thin enough that you can actually incinate into the brain. So some thick-headed people, their windows aren't thin enough. I won't mention who those might be, the demographics, but for most people, you'll be able to see into where the circle of Willis is. So let's zoom in on that a little bit. So a couple things to be familiar with. So for each heartbeat, it would look like this on transcranial Doppler in a relatively healthy patient. You see this nice systolic upstroke, the systolic downstroke, and then diastole, and then end diastole. This red arrow is pointing to the peak systolic velocity. The yellow star is pointing to the end diastolic velocity. In general, when you're eyeballing this, you want your end diastolic velocity to be somewhere in the range of about a third of your peak systolic velocity. Really, it's probably 20 to 50%. But if you start seeing the end diastolic velocity dropping low, less than a third of your peak systolic velocity, that can be a concerning sign. Most of the machines will give you an output of your mean flow velocity. This is on the high side for an MCA, normal's about 50 centimeters per second. And then the pulsatility index, which a lot of people have heard of, is we think about it as a resistance to flow indicator. So higher numbers, higher resistance to flow, lower numbers, less resistance to flow. So just like waves at the ocean tell you something about weather forecasting, they tell you about climate forecasting, the waves of the brain also can do the same thing, which is just incredible. So learning patterns can be really helpful to tell you if there might be an impending ICP issue. So I'm looking out at the crowd here. This is a really robust, vigorous-looking crowd. I think that most of us, our brain waves are probably looking like this here, okay? So our cerebral perfusion pressure is much higher than our ICP. Our ICPs are less than 20. And we see these nice, healthy waveforms here where that end-diastolic velocity is about a third or more of the peak systolic velocity. But what happens when you get high resistance to flow and your ICP starts increasing? You see a pattern change where you start to see these really sharp waveforms and your end-diastolic velocity is dropping. And then very dangerously, when your ICP is equal to your diastolic pressure, your diastolic flows drop out completely. And this is a really, really dangerous finding, right? The brain is so metabolically active, it wants blood flow throughout systole and diastole. That period in diastole for every heartbeat where it's not getting blood flow, that's time when there's ischemia. So that's a very dangerous pattern there. And then as you go on and the ICPs are increasing more and more, eventually you get into no flow. So this, if you see this on your TCD point of care or if you have techs at your shop that can do these TCDs, this is a big warning danger sign. So how does this perform as a diagnostic tool? How does it really equate to ICPs? Well, it's a very storied history. And in the interest of time, we'll kind of keep it brief. But at one time we thought, well, we can directly convert pulsatility index to ICPs. But it turns out that that kind of breaks down and that's not always a perfect correlate to what your ICPs are. There have been some groups that have described a mathematical model shown here that they call ICP-TCD, which is a formula of your mean arterial pressure minus your estimated cerebral perfusion pressure, which ends up being your mean arterial pressure times your end diastolic velocity on your TCD over your mean flow velocity plus 14. And that has a decent diagnostic accuracy in terms of predicting ICPs. So this was shown in the IMPRESS-IT two prospective multicenter international study in 2022 by Rizula et al. But they really looked at, they compared this TCD ICP number with invasive monitoring at various thresholds in patients with all kinds of brain injuries. So it was TBI, ICH, subarach, acute ischemic stroke. And they looked at the negative predictive value of this TCD ICP number and found that the range for an ICP of greater than 20 was 91%, greater than 22 was 96%, greater than 25, 99%. Positive predictive value was 20 or 30s, range from 20 to 30% across these various thresholds. So quite good in terms of negative predictive value. But if you're getting a high number, may not be accurate, right? So although we don't know the perfect formula to equate TCDs to ICPs, we're starting to learn a lot. And the main take home, I think, is that if you get a baseline TCD on your patient that you're worried may develop elevated ICPs that doesn't have an indication for invasive ICP monitoring and it looks like this, and you trend over time and there's a change to this, that is worrisome, right? So all about trends. So what about optic nerve sheath diameter? So it's another quick tool that is nice to have. So this is, you just use your linear high-frequency probe to actually incinate the eyeball. And when I'm looking at optic nerve sheath diameter, I typically do two measurements in transverse shown here, and then two measurements in sagittal shown here. You wanna be sure to set your presets to an ocular preset or superficial preset with a low mechanical and thermal energy so you don't injure the eye. So when you actually put the probe over the eye, it's really beautiful. And if you have students, med students especially, this is a great way to engage them at the bedside. But you see these structures in the near field, and then in the far field, you'll see the optic nerve and sheath optic nerve and sheath down here, this hypoechoic linear structure. We'll zoom in on it a little bit to talk about some nuances, but you wanna be measuring three meters back from the back of the eye, and then you wanna measure completely across to get the optic nerve sheath. And then I do wanna zoom in on that a little bit more. So just so you can see a little bit of the detail. So the optic nerve itself is hypoechoic like this. This relatively hypoechoic structure right here is actually CSF or the subarachnoid space. And then this other hypoechoic line just adjacent to that is actually the dura, okay? So this whole thing here is the sheath. And then you can see the schematic over here that portrays it as well. The only reason I mentioned that is there is some debate about whether or not we should be measuring from the start of the dura here to over here to versus maybe where the dura is out here. So this is the external versus internal dura. It ends up being tiny little minuscule difference, but there are people working on seeing the best spot of measuring. I think the important thing is stick with what you know. I think a lot of us clinically will use that internal. So we'll measure right at the point where the hypoechoic and the hypoechoic area meet all the way to the other side. In terms of how this is performing diagnostically, so this was a nice meta-analysis by Roba et al that looked at seven different studies looking at optic nerve sheath diameter compared to invasive ICP monitoring. Those different studies are shown here in triangles. And the various studies performed really well in terms of sensitivity and a relatively low false positive rate. So when we're looking at these curves, we think about if we drew imaginary line here, this would be a test that is essentially almost like a random number generator, not very good. And the further we get to this point, this would be an ideal test that would be really excellent. So in terms of how it's performing, it seems to be relatively good in terms of ruling in elevated ICPs with a relatively low false positive rate. The problem with these studies is there's no agreement in terms of what's the cutoff of abnormal. So a lot of us say, well, five millimeters or more are abnormal, but we know that there's a gray area somewhere between that five and six millimeter range could be normal in some people. For most people, less than five millimeters is gonna be normal. So we're still learning a lot and this tool is not perfect either, but it is a tool that we have and we can use. In defining normals, less than five millimeters is likely normal. Somewhere greater than five to six have been associated with elevated ICPs greater than 20. But what's so important, just like TCDs, it's all about trends. So if you have a patient that you're worried is gonna develop high ICPs and cannot get invasive monitoring, get these baselines, look at their optic nerve sheaths, start practicing that. And if there's a change and there's a change when you're looking at the patient clinically, that would be potentially concerning. So it's all about trends. So a potential framework. So working under the assumption that you're getting these baselines on patients you're worried about. Say there's a clinical change and you're concerned for high ICPs and invasive ICP monitoring is not indicated. I think at that point, my practice is I'm gonna go to the bedside and get a point of care Doppler reading of the MCA. If there's no concern for high ICPs using ICP-TCD or just the waveform, then I'm gonna continue close monitoring. I'm gonna be considering other tools and tests. This is not discounting any other tools that we have available to us. If there is concerning waveform change, knowing that there's a lot of false positives with the TCD, ICP and waveforms, then I'm going to check optic nerve sheath and potentially other tools as well. If my optic nerve sheath is less than five millimeters and it was before, when the patient came in, they maybe were around the same range, then I'm gonna think about other monitoring and continue to look at my patient. If the optic nerve sheath diameter has changed and the patient is now more than five to six, I'm very strongly considering invasive therapies, obviously using other multimodal tools and even treating empirically. Dr. Glacomoff-Locken said I could use his actual face in these presentations. Long story, I can tell you about it afterwards. But at any rate, so these are non-continuous tools, which is a drawback and there's certainly a learning curve that I've glossed over here. But they're cost effective in their point of care and they're super fun to go to the bedside and get your hands all gooey and stuff with gel. And this TCD, ICP waveform, I think they give us, it gives us clues and I think that it matters. And optic nerve sheath, I think pretty good diagnostic accuracy for high ICPs, but obviously has its limitations. But trends really matter. So get baselines on your patients. And then CHAT-GPT, I asked it what a brain superhero would look like and it gave this. I love asking it to give me various images and so this is what it did. So I was like, hell yeah, good for you, CHAT-GPT, I like it. So, all right, thank you.

non-invasive monitoring

intracranial pressure

transcranial Doppler

optic nerve sheath diameter

neurology