Thank you, Rhonda, for that really kind introduction. So what I'd help to do, over the next 20 minutes or so, is sort of discuss some of the most, some really important articles that were published this past year in 2023 related to neurocritical care medicine and discuss some of the highlights of their data, but more importantly, discuss some of the limitations associated with these studies so that you all can take home and use some of the data to figure out how to best implement these results into your own practice. I don't have any financial disclosures, unfortunately, and the outlines of the study, we're going to talk about two trials involving ischemic stroke, namely BEST-2 and Optimal BP, that looked at best blood pressure practices for patients who've had mechanical thrombectomy after an acute ischemic stroke. We'll then segue over to the INTERRACT-3 trial and interest-rebel hemorrhage. We'll examine bundled care approach in patients who come in with an interest-rebel hemorrhage. Look at the early drain trial in subarachnoid hemorrhage, which examined lumbar drain placement in addition to standardized care in patients with aneurysmal subarachnoid bleeds. And then finally, look at the RESCUE acute subdural hemorrhage study, which looked at two different surgical approaches to patients who presented with subdural hematomas. So a lot of data to cover, and so let's get started and get the round running. So the two articles that we'll start off with are the BEST-2 trial and the Optimal BP trial, and these were both published around the same time in JAMA, and they both complemented each other with relatively similar evidence. Right now, the American Heart Association recommends that we maintain a systolic blood pressure of less than 180 in patients who've had mechanical thrombectomy after a large vessel occlusion. And the idea for this blood pressure target is to sort of optimize trying to reperfuse the ischemic per number, which is what the goal is in all ischemic stroke trials, versus trying to prevent pretty significant reperfusion injury, which can happen when you open up a blood vessel in an area of the brain that's now been infarcted. So despite these recommendations, there's very minimal data to support this particular blood pressure goal, and there's controversy in whether tighter blood pressure control actually can improve outcomes in patients who've had mechanical thrombectomy, i.e., preventing patients from having reperfusion injury and potentially symptomatic ICH, which can be fatal. So the major knowledge gap that these trials are trying to address was, does lower systolic blood pressure targets after mechanical thrombectomy actually improve outcomes in patients with ischemic stroke? So both studies were multi-center randomized control trials to compare liberal versus conservative blood pressure control after mechanical thrombectomy. It's a little bit busy slide, but I wanted to really highlight some of the differences between both trials. BEST-2 was a three-center US comprehensive stroke study involving over 120 patients, and OPTIMAL-BP was largely done in South Korea with over 300 patients. Both studies looked at different blood pressure parameters in BEST-2. They looked at systolics of 180 versus 160 versus 140, whereas in the OPTIMAL-BP study, they compared two different arms that patients were randomized to either a systolic of less than 140 versus a systolic of 140 and 180. Now the BEST-2 trial was set up as a futility trial to see if there was no harm, whereas OPTIMAL-BP was set up as a true intervention study to see if there was benefit. They both looked at modified Rankin scale, which is the most commonly used outcome measure in stroke trials in general, and they were looking at outcomes at three months, comparing modified Rankin anywhere from 0 to 6, 0 being fully independent after an ischemic stroke and 6 being dead. In addition, BEST-2 was looking at infarct volume at 36 hours in CT scan, and this was really trying to determine whether reduction in blood pressure leads to worsening ischemic penumbra, whereas OPTIMAL-BP was taking the other approach and looking at whether higher blood pressure goals would prevent, or lower blood pressure goals, excuse me, would prevent symptomatic intracerebral hemorrhage. So here are the blood pressure goals from the BEST-2 trial, and what we see is that in all three tiers of blood pressure, they were able to achieve target. What's interesting, however, if you look at the actual mean blood pressures in all three categories, they're anywhere from 129 to 122. Now while the authors didn't do any statistical analysis to see if there was a difference between three groups, as you can notice, it really begs the question, and we'll talk about this in a little bit, whether there's a clinical difference between having a systolic of 129 versus 130 versus 122 in ischemic stroke patients. When they looked at infarct volume at 36 hours, lower blood pressure targets were not clearly associated with infarct volumes, and this is based upon the three different tiers of blood pressure. You can see the corresponding infarct volume that was calculated on CT scan, so there may be actually a little bit different if they used MR imaging, but when they looked at the analysis of this, there was really no difference depending on 36-hour infarct volume based on CT scan. More importantly, when they looked at functional outcomes, they found that lower blood pressure targets were actually not associated with any improvement in functional outcomes, and because this was a futility study, they found that there may be a trend towards worse outcomes in patients who have lower blood pressure targets compared to those that are more liberal. In the optimal BP results, something similar, so when they looked at differences at 24 hours in mean systolic blood pressure, they found that there was a statistical significantly different mean blood pressure values, but again, I want to raise the question of whether there's really a clinically significant difference of having an eight or nine millimeter mercury difference in systolic pressure. They found something similar in the diastolic pressures that while they were statistically significantly different, raises the question whether they're clinically significantly different. In this study, they actually found that functional independence was lower in a tighter blood pressure control compared to those patients that were more liberalized. Interestingly, while they did not find a difference in blood pressure for all three targets, what they did find was that patients who were in the more conservative blood pressure arm were more likely to receive antihypertensive medications and were more having episodes of hypotension compared to those in the liberal arm, and what they found in the safety outcome was actually there was no difference in outcomes in some patients developing symptomatic ICH. So the main conclusions of the study is that lower systolic blood pressure targets after mechanical thrombectomy may, and I use the word may because in the BEST-2 trial, there was a question of whether it may be associated with harm, and in the optimal BP study, they actually did find there was no benefit with lower blood pressure targets, may be associated with worse outcomes, and they also found that lower systolic blood pressure targets did not reduce the risk of post-thrombectomy ICH. So limitations, as I've already been discussing, was whether there was a significant enough difference in systolic pressures, even though they were statistically significantly different, to see a difference between groups. And this hardbacks to what we sort of look at in the intensive care world is that, you know, how reliable are systolic blood pressures when we measure them in the intensive care unit and whether we should be using other parameters such as mean arterial pressures, which tend to be more reliable, and provide information such as cerebral perfusion pressure, which is the actual pressure at the circle of Willis in patients that receive, that have intracranial devices placed in. And the other big confounding factor is that patients who were hypotensive actually had, or patients in the more conservative blood pressure arm actually had greater episodes of hypotension compared to those in the liberal arm. And there's emerging data suggesting that absolute blood pressure reduction may not be as important as blood pressure variability as patients are being reduced, meaning having spikes and drops that are occurring consistently with bolus pushes of antihypertensives or with continuous infusions of antihypertensives may be more harmful than good. Moving on to Interact-3. This was a study that also published. This was published in Lancet. And this study was examining bundled care approach in patients with ICH. And the idea was that does goal-directed bundles, are they successful in improving outcomes in patients with intracerebral hemorrhage? And we know as intensivists that goal-directed bundle care has dramatically improved care in the intensive care world, specifically with septic shock. And it's also been studied in ischemic stroke. So you might be wondering, well, what's the big deal about this? In the world of ICH, it's actually quite important because ICH is usually approached across the world, even in the United States and here in North America, as a disease of futility. You know, patients present, we think there's not much we can do for them. So this study was very important in showing that aggressive medical care upfront and early may actually be beneficial. And what they based this trial on was post-hoc analysis of the Interact-2 trial, which looked at two different arms of blood pressure, 140 versus 180. And what they found was that elevations in various physiologic parameters, blood pressure, blood glucose, body temperature, and failure to reverse anticoagulants actually led to worse outcomes. So the idea was, if we fix all of these, can we actually improve outcomes with patients in ICH? So this was a very large study. It was a multicenter randomized controlled trial involving nine low and middle income countries and one high income country with over 7,000 patients. And it was a care bundle that included a systolic blood pressure of 140 within six hours of presentation to the emergency room, a blood glucose target depending upon whether you were diabetic or not, fever control that was aggressive upfront within the first hour of recognition of fever, and reversal of anticoagulation-related ICH, again, within one hour of presentation to the emergency room. And they looked at the modified Rankin scale, again, as a standardized way of seeing if this bundle approach would improve at six months. And what they found was that systolic pressure at 24 hours was, again, statistically significantly different between the bundled care arm versus the non-bundled care arm. But I'll raise the point again. Is a four millimeter difference mercury and systolic pressure clinically significant? That's something up for debate. They found that there was a higher proportion of patients that achieved good blood pressure glucose control early on upon presentation at the hospital. When they looked at the measures at 24 hours, the differences between the two groups were nearly identical. And same thing with fever control. Although fever control was achieved earlier on in the patients that received bundled care, at 24 hours there did not seem to be a difference between patients who received bundled care versus not in the prevalence of fever. They did not find there was a difference in reversal of anticoagulation in patients who presented to the hospital with warfarin-related ICH, although this was a very, very small percentage of patients, less than 1%. And what they found was that the intervention group, surprise, surprise, had a better modified Rankin scale at six months compared to those patients that did not receive the bundled care. And this really harks back to the idea that early aggressive care in these patients has the potential for improving outcomes. What's interesting, however, if you look at the modified Rankin scale, the patients with ICHs that were more destined to have worse outcomes are the ones that actually have the improvement in their functional recovery. And these are the exact same patients. The larger ICHs are the ones that have physiologic streamings, such as fevers, the ones that are going to have glucose spikes, and the ones that are more likely to have elevations in temperature throughout their hospitalization. And it's probably treatment of these patients that drove the overall outcome to show benefit. So conclusion of this study, they found that a bundled directed care approach, including early blood pressure control, treatment of blood glucose, hyperglycemia, pyrexia, and aggressive treatment of abnormal coagulation resulted in improved outcomes. Now some of the limitations, the big question is, can this be generalizable to centers with already established ICH protocols? So especially here in the United States, American Heart Association, we have primary comprehensive stroke centers that are with already pre-signed protocols in place to treat these patients. Would implementation of bundled care protocols such as this really add any benefit? Because majority of these countries were low and middle-income countries with no pre-existing protocols in the first place. The study was designed in a way to sort of emulate how a new protocol would be brought into play. So there was a period of enrollment where patients were enrolled with that standard of care, a training period, and then a period where the patients had intervention based on the bundled care. However, COVID-19 occurred, and it's unclear whether the training that was done early on before the intervention was done, how much that was retained, and whether it was actually followed because of COVID-19. The authors tried to adjust that with various statistical analyses, and they outright state this is a limitation of their study. Another thing was, is there really a difference between the intervention and the non-intervention arm? As we looked at, a lot of the metrics were identical. But what was a important factor in this was early aggressive care. So time to intervention was the most important factor that it probably led to the benefit of a bundled care approach. And we talked about generalizability to established ICH protocols already. This is another article in JAMA. I swear I'm not trying to promote the JAMA publication whatsoever. They refused my last publication, in fact. So this is looking at the use of lumbar drains in patients with aneurysmal subarachnoid hemorrhage. And what is known is that the amount of aneurysmal subarachnoid hemorrhage has strongly been associated with the risk for vasospasm and delayed cerebral ischemia in multiple studies. But there's controversy in whether the additional removal of blood products through a lumbar drain may actually reduce the risk of delayed cerebral ischemia and vasospasm. And the idea behind this is that, you know, the authors say that blood actually will flow to the most dependent parts of the body. So you have the aneurysm that ruptures. Blood is going to probably sit at the fecal sac and the lower parts of the spinal column. So the idea was, if we remove all of that blood through lumbar drain placement compared to just placing an external ventricular drain, would that really improve patients with subarachnoid hemorrhage? And for schematic here, we have two pictures, external ventricular drain, which is a drain that goes to the cranium and sits within the third ventricle to drain blood, and a lumbar drain, which is similar to how lumbar puncture is done, but a catheter is placed in that area to continuously remove blood. And the authors here have a very striking visualization of the difference in blood products between both the external ventricular drain and lumbar drain. On the left-hand side of the screen, you see the sort of palish orange CSF color from CSF drawn in the ventricle. And then from the lumbar drain in the same patient, you have a much more bloodier concentration of CSF that's coming from that port. So this is a multicenter randomized controlled trial with blinded outcomes. 144 patients were randomized to receive standard of care, which is whatever the attending neurosurgeon slash neurointensivist decided, versus standard of care plus the lumbar drain, which required at least five cc's an hour of drainage of CSF over a period of four days, so additional 480 cc's of CSF. And the primary outcome was the rate of unfavorable neurologic outcomes, again, using the modified Rankin scale at six months. And here's the MR shift analysis, and what they found was that in patients who received a lumbar drain in addition to standard of care, they had an improvement in their outcomes with the number needed to treat around eight. They also found that when they looked at the cc scan to look at how much infarct burden was related to delayed cerebral ischemia, they found that there was patients who had lumbar drain actually had less infarct burden on ct scan compared to those that did not. So it sounds like a very promising approach to treating subarachnoid hemorrhage, and they concluded that lumbar drains, in addition to standard of care for aneurysmal subarachnoid hemorrhage, reduced the risk of unfavorable outcomes and resulted in less cerebral infarcts. But there are some, I think, limitations against this study that I want to point out. First of all, the outcomes weren't really adjusted for the amount of subarachnoid blood in the brain. So commonly used in subarachnoid hemorrhage literature is the modified Fisher score or the Fisher score. And that's a way to sort of level the playing field, if you will, to determine which patients and what the risk for vasospasm was. And the authors actually did not do that in this particular study. Interestingly, only 75% of the patients in the lumbar drain arm actually received a lumbar drain. The remaining 25% either didn't have a lumbar drain placed for technical purposes, or it's actually not really clear why it wasn't placed. Now the authors did do a as-treat analysis as opposed to a per-protocol analysis, and they found similar results suggesting that lumbar drain was beneficial. And finally, and the authors I'll come up front and state this, is that premorbid conditions and medical complications in the intensive care unit were actually not accounted for in the outcomes analysis. So we all know that as intensivists, a patient that comes in with end-stage renal disease, congestive heart failure, and maybe poor pulmonary disease up front, that patient's probably going to have a worse outcome than somebody who's young and healthy. And it's unclear what these comorbid conditions were, and it's possible that one arm actually had a healthier subgroup of patients versus another arm. So I think the data is still up in the air as regarding to how this intervention would perform if those things were adjusted. And then finally, we have the decompressive craniectomy versus craniotomy for acute subdural hematoma, the RESQ-ASDH study that was published in the New England Journal of Medicine. Now traumatic subdural hematomas are one of the most common indications to have neurosurgical intervention. And there's two common surgical interventions, a decompressive craniotomy, where the bone is actually removed, the idea being that these patients will develop intracranial hypertension over the coming days, and if we keep the bone off, that prevents that from occurring and leading to worse outcomes, versus a craniotomy, where the bone is taken off and replaced back onto the skull. Decompressive craniotomy may be beneficial, as I mentioned, in preventing intracranial hypertension, but it's unclear which patients are going to benefit and if it's actually effective in doing so. So the question was, does craniotomy versus craniotomy actually improve clinical outcomes overall? So this was a multi-center randomized control trial over 450 patients at 40 medical centers in 11 countries, where they randomized them to craniotomy and craniotomy, and a participant in the study had to be, by the attending neurosurgeon, declared to be a patient that could be a candidate for either surgical procedure. In fact, in the OR, when a patient had a craniotomy done and the neurosurgeon felt that the brain was swelling too much and the bone flap could not be placed back on, that patient was excluded from the study, because that patient definitely needed a craniotomy. And they looked at 12-month extended Glasgow Coma Scale, which is a commonly used outcome measure in patients with traumatic brain injury, along with multiple secondary outcomes that you can see below. And here's a visualization of what craniotomy versus craniotomy looks like. You have, on the far right-hand side of your screen, a traumatic acute subdural hematoma, a patient who undergoes a craniotomy with removal of the bone flap, removal of the blood, and replacement of the bone flap, and a decompressive craniectomy patient where the bone flap is kept off. And what they found in their primary outcome was there's actually no difference across outcome categories between a craniectomy versus a craniotomy. And when they looked at secondary outcomes, they found no differences at six months in quality of life metrics or in mortality. What they did find, however, is that patients in the craniotomy group were more likely to need a decompressive craniectomy in the coming two weeks. And the most likely cause for this was patients developing cerebral edema. However, they found also that the patients who received craniectomies up front, they were more likely to develop wound infections over the coming months. So these were sort of secondary outcome measures that they noticed when they were looking at the different trials. So in conclusion, they actually found that there was no difference in outcomes between decompressive craniectomy versus craniotomy. They did find that there was additional surgical procedures performed in a higher proportion of the craniotomy arm and more wound complications in the decompressive craniectomy arm. Now limitations of this study, and I think a big critique, was time to intervention. So I think this is sort of the dogma of critical care medicine is that early intervention is the key to recovery. And the same thing holds true for the brain. Early intervention is key to recovery, and so time to intervention between the decompressive craniectomy arm versus the craniotomy arm was never outlined and could bias the results. There was no data on the timing and complications related to post-cranioplasty. So after the patient has a decompressive craniectomy, they present back to the hospital anywhere from three to six-month period of time, depending on when the surgeon feels appropriate, to have that bone flap placed back on. And that's another hospitalization, another stay in the intensive care unit. And all the complications, all of the issues related to that were never described in the paper. Presumably, they would have been accounted for in the 12-month outcome measure, but they would have been additional metrics related to whether the patient developed infections or other complications related to that. And again, ICU complications, premorbid conditions were not included in the analysis. So it's possible that patients in one arm were actually healthier or had less complications compared to patients in the other arm. So I know that was a quick overview on five very important studies that were published this past year. I left references here if you're interested in reading them in more detail. And I truly appreciate your time, your attention, and happy to take any questions at the end.

neurocritical care

ischemic stroke

intracerebral hemorrhage

subarachnoid hemorrhage

acute subdural hematoma

clinical trials 2023