Welcome, everyone, to another episode of the Front Row Series. Our topic tonight will be cerebrovascular surgery. Again, the Front Row Series is a new event from the AANS, and it's featured on the AANS the official educational website of the AANS. We're lucky enough to have renowned experts from around the world with many topics. Again, tonight's topic will be in cerebrovascular surgery, and participants can have access to both live as well as archived events with CME. So please join us at any time, and you can watch this also at your leisure. We're extremely lucky today to have two experts that really need no introduction. We have Dr. Gustavo Predilla. Gustavo has been a longtime friend, and in a way, almost like an older brother that led the way for a lot of things. But Gustavo is an associate professor at Emory, and he's the chief of the neurosurgery service at Grady Memorial Hospital, which is a very busy place. But he has spearheaded a trial now, the ENRICH trial, which has the potential to revolutionize the way that we actually treat hemorrhagic stroke. So today, he's going to speak to us a little bit about where we stand, where does all the data lead us, and hopefully, where do we go from here. On the other hand, we have Dr. Jay Malko. Again, Dr. Malko needs no introduction. He's one of the leaders in cerebrovascular surgery at both endovascular and open cerebrovascular surgery around the world, and he has produced perhaps a large part of the literature pertaining to the endovascular treatment of ischemic stroke. So today, he's going to give us his part of his wisdom on where we stand with that and where we're going. So without further ado, I'd like to invite Dr. Padilla to share his screen. Gustavo, we're really looking forward to your presentation. George, thank you so much for the opportunity to spend a few minutes today. Thank you too. Very good. So these are my disclosures. The most important one is that, as George mentioned, I'm the principal investigator for the ENRICH trial, which is sponsored by NICO Corporation, and they make one of the devices that I will be discussing. So intercerebral hemorrhage is an endemic problem for neurosurgeons, neurocritical care physicians, neurologists, and it constitutes a large portion of all strokes, and despite all of the things that we now do well, like neurocritical care, the mortality is still quite high, and the disability of the survivors is significant, and we spend a lot of money caring for these patients. And the main etiologies that cause this, things like hypertension, amyloid angiopathy, coagulopathy, because we now have so many folks who are anti-coagulated, those are contributors that increase and will continue to increase as we go on. When we talk about the pathophysiology of the disease, it sheds light into what are the challenges and what are the potential solutions. So when there's an initial hematoma, that initial injury generates this cascade of events, but if we decide not to intervene, anywhere from four hours to seven days after this hemorrhage happens, all of these factors floating into the parenchyma and the surrogate space cause all of these endothelial and microglial activation, and that leads to cytotoxic edema, and that finally causes the vasogenic edema that is a perihematoma, sort of dark, hypodense halo that we see in the screen. And that's when our patients develop intractable intracranial hypertension, and then we are called urgently to provide a surgical solution. So the rationale for hematoma evacuation has traditionally been focused on these three things, controlling intracranial pressure, preventing these clots from expanding, which they do quite often, and preventing this secondary cascade of inflammation. When we treat this in the traditional way of doing perhaps a decompressive craniectomy, when the patient doesn't have any other options and we're maxed out in care, of course, there are other things that happen. We have patients with syndromes of the trophine, patients who are staying in the ICU for many days and in rehabilitation for many days, and the overall cost of the system can be quite significant. So traditional surgery, of course, was very well evaluated in a couple of famous trials, the STITCH trials conducted by Professor Mandelot, which evaluated conventional craniotomy performed early within 24 hours of the event in some cases. And early surgery was shown to be favorable only in 26% of the patients when compared to conservative treatment, which was beneficial in 24% of the patients. So a trend towards benefit, but not statistically significant. That study had a subgroup of patients with lobar hemorrhages without interventricular hemorrhage that showed an even larger benefit of 49% of them did well versus 37% in the conservative management group. And because this wasn't powered to analyze those two subgroups, they did the STITCH-2 trial, which focused on that specific patient population. And unfortunately, patients that had surgery, 59% of them had an unfavorable outcome versus 62% in the conservative treatment group. So again, another surgical study that was negative. And because of those two studies, most of the neurosurgeons around the world were very discouraged about doing surgery for intracerebral hemorrhage until this group of our colleagues came together, Dan Hanley from Neurocritical Care at Hopkins, Mario Zuccarello, and Yazama Wood, and proposed the MISTI approach using a stereotactically guided catheter for gentle aspiration followed by thrombolysis. And they performed an initial study where they showed that it was safe and that it was effective with a few caveats. Number one, of course, you're putting thermolytics. You cannot do that in an actively bleeding cavity. So you need a stability scan. And number two, sometimes it may take up to 52 hours or longer to get to that very nice result. But nonetheless, that was a very exciting study that led to the design and execution of the MISTI-3 study led by this same group. And unfortunately, the MISTI-3 trial did not show a benefit for the primary outcome measure, which was a modified ranking score at six months and one year. However, they did show that there is now a surgical target. So if you reach an end-of-treatment volume of less than 15 cc, or you evacuate more than 70% of your initial hematoma, the likelihood of achieving an MRS of 0 to 3, a favorable outcome, was significantly higher. So why didn't they get there? Unfortunately, they only reached that end-of-treatment volume in 59% of the patients. So in our minds, the question remained, had they reached perhaps the 65%, 70% of the patients with the right end-of-treatment volume, maybe the primary outcome measure could have been positive. But nonetheless, it taught us a lot of valuable lessons. It was a very well-conducted study. And it showed that evacuation matters and that training in a systematic way can have a pretty significant effect on how these patients will do. So where do we stand was the question after we have STICH evaluating conventional craniotomy for cloud evacuation and MISTI. Well, 30% to 40% of these patients will expand, as I mentioned, and that picture in the top right is common to all of us. We have patients that come in, they look decent, and then about four hours later, they decline and we have critical expansion there. Spot signs, active contrast extravasation is the contraindication for MISTI. So we need an alternative for those kinds of patients. And some patients have clot burdens that are so high that they can't really wait for a stability scan and they can't wait for five or six days in the ICU to relieve their hematoma under pressure. If we think about why these patients die, the survivors that come into the hospital, over 50% of them will die from medical complications, many of those in the ICU. So if we can eliminate or shorten their stay in the ICU, every day that we cut off their ICU stay, we're decreasing the incidence of pneumonias, infections, DVTs, PEs, delirium, et cetera, et cetera. Since those studies were conceived, our understanding of eloquent cortex and its role in the presentation and potential treatment for ICH patients has changed. And these previously thought of non-eloquent areas have been very well defined now with diffusion tensor imaging and awake neurophysiology mapping. So taking this into consideration can potentially lead to less destructive pathways for clot evacuation. So looking at the trials that we have done, we still have some work to do in refining patient selection. What are the limitations and the impact of certain factors for the patients? The timing of the intervention, we have some exciting work by Dr. Moko and Dr. Christopher Klerner at Manzana showing that early evacuation is not only safe, but actually quite effective in improving patient outcomes. So the needle is moving into acting early with the right techniques. And of course, training is critical and incorporating all of the new technologies that has been developed. So ideal surgical procedure is something that we all want to conceive. So for ICH, these are the six things that we would love to have. An immediate impact on mass effect. Doesn't matter how big the clot is, we should be able to relieve that mass effect as soon as possible. We should be able to do so as thoroughly as possible as MISTI 3 showed us, that antitreatment volume is critical. We need to minimize iatrogenic injury. And that was one of the concerns of some of the subgroup analysis with the STICH trials was that perhaps there was some unnecessary collateral damage of the conventional surgical evacuation. Direct visualization for hemostasis, that's something that makes all of us nervous during these cases and having good visibility and good hemostasis is critical. And then of course, if we can cut down on that length of stay and the overall cost of the intervention, that will be fantastic. So in our case, to the study that Dr. Sinone has mentioned, we decided to incorporate this sort of six principles to treat ICH. Number one was using better imaging to understand what's happening to the normal anatomy and design better corridors to access the hematoma, doing so under intraoperative navigation with devices that have a smaller footprint than conventional and craniotomy will have. Using better tools to visualize the hematoma and the bleeding source, better tools for evacuation of those. And then finally, collecting clot, impaired hematoma, edema, and all of the data that is afforded to us by having this privileged access to the hematoma. So as I mentioned, using tractography for planning is something that has gone from the realm of interesting neuroscience to everyday neurosurgical practice. And most people have access to this. Retraction injury is something that used to happen quite commonly in the 80s and 90s. But over the last two decades, this quiet revolution that Dr. Spetzler talked about really took hold. And now it's rare to see this flat blade retractor being used. But the principle of avoiding those and substituting that for radial-based retractors is something that was proposed a long time ago by Pat Kelly, for example, who advocated for using this small cylindrical retractors to resect lesions in the basal ganglia. And that concept remained and evolved. There are several tubular retractor systems available nowadays. The one that we are evaluating in the ENRICH trial is the BrainPad system, which is a 13.5 millimeter port that comes in different lengths from 50 to 75 millimeters. And it's placed under stereotactic guidance. It's designed to be introduced through a solocle opening as opposed to a cortisectomy. And we try to understand what are the loading vectors for the main subcortical structures to be able to avoid transfections and disconnection syndromes during access. We're also leveraging newer imaging techniques using exoscopes, which in addition to giving you a wider volume of view, which is what you see at the bottom, they also allow you to do digital image processing. So you can increase light, you can increase the light absorption and play with different filters to enhance the normal structures and decrease that negative effect of the clot, which absorbs so much light. So this sort of puts all of these concepts together in an example of an anterior basal ganglia hemorrhage that's accessed through a forehead crease incision using this port guided by stereotactic navigation with an exoscope. And once we've reached deletion using your normal tools for microsurgery, suction, bipolar cautery, and a device to remove the hematoma. So this is an example of what that looks like. This is a 74-year-old patient presenting with classic hypertensive hemorrhage in the basal ganglia. This is what we do for planning our trajectory. And we try to understand where these fiber tracks have been displaced in this anterior basal ganglia hemorrhages. Very classically, the cingulate fasciculus is displayed medially and the superior longitudinal fasciculus is displayed laterally, creating a quarter that is anterior to posterior, like you see illustrated in those images. So once we're happy with that, we plan that trajectory and then plan our craniotomy and skin incision and then proceed to execute that plan. So this is the system that we use for visualization. This is a robotically mounted exoscopic device. And this is our exposure. We try to do small craniotomies on the order of four centimeters. The dural openings are about the size of the device, which is about 15 millimeters. That dural opening and we open through a sulcus sharply. In this case, we're using a 75 millimeter port as being introduced under seotactic guidance. And once we are in place, we activate our robotic assisted exoscope that auto-aligns to the bottom of that port and then has auto-focus capabilities. So we're auto-aligning now, and now we turn on our filters that allow us to bring in more light. And then you can see that blood clot immediately starts to erupt from the field. This is our automated resection tool. We're suctioning that and doing as thorough an evacuation as we can. And one of the things, as I mentioned, we want to do is be able to handle hemostasis properly. So in this case, this is suction and bipolar cautery. This, upon inspection of the cavity, was probably the vessel of origin that ruptured and caused the hemorrhage. Here's more of that large clot being evacuated. And these are the last bits of clot coming out. And then this affords you the visibility to inspect the cavity thoroughly. And all of these patients have a CT angiogram to make sure they're in vascular lesions, but not rarely you will see sometimes the micro sort of cryptic AVM, as Dr. Roberta Harris used to call them. And if that were to happen, you have the capacity to deal with that with suction and bipolar cautery. And then perform your meticulous hemostasis, your decannulation, looking at the banks of that corridor, making sure there is no hemorrhage, and then finally retracting the port and inspecting that well-preserved cortical surface. So that's sort of a standard clot evacuation. They are quite satisfactory and the radiographic outcomes are pleasing. And we did an initial multicenter study where we evaluated that technique in very carefully selected patients. And the clot evacuation results were very satisfactory, over 90% clot evacuations. And the modified ranking scores of that multicenter group was very encouraging. A lot of these patients had an MRS less than three. So these results were also evaluated in single center institutions, Dr. Mark Bain at the Cleveland Clinic, Dr. Jeremy Kopens at St. Louis University and others. And following all of those studies, we decided to put this technique to the test in the ENRICH trial, which is a multicenter effort with a good bit of economic analysis. We have 32 sites that are participating in this study. And this is our website if you'd like to learn more about the inclusion and exclusion criteria in the trial design. This is a multicenter randomized adaptive clinical trial using Bayesian modeling that compares standard medical management to medical management using current guidelines. And our goal is to provide clinical evidence of a functional improvement of safety and an economic benefit of this technique. As I mentioned, it's multicenter, it is randomized, and we're doing block randomization by location with a sample size that ranges from 150 to 300 subjects and interim analysis that started at 150 patients and went up at 25 patients after that. And we are at 265 patients to date. And Dr. Sonona's actually enrolled the last patient, 265. So we are close to our next interim analysis will be at 275. And final sample size, which will be at 300, which will be very exciting. But this is only one of many efforts going on. Endoscopic ICH evacuation is an area also great excitement. And Dr. Kellner and Dr. Melko are some of the leaders in the world. This is another quad evacuation. This one by my good friend, Pablo Racines at the Cleveland Clinic using endoscopic assisted evacuation with a side port evacuation device. And we're seeing a lot more variations of these theme using endoscopy with really terrific results. Dr. Kellner and Dr. Melko, as I mentioned, have led this effort. And there are three exciting studies that I'll briefly mention. But from their description of the scuba technique, then one of the most exciting things was the evacuation percentage, which was 88%. In hematomas that were on average sizeable, 42.6 CCs. And they were able to see that actively bleeding vessel in 49% of the cases. So very exciting data. So these three studies that I will briefly mention include the INVEST trial, the NIND study and the Dutch intracerebral hemorrhage study. You can see briefly the comparison there in terms of volumes for inclusion, NIH stroke scales, time of onset and locations. This is a little bit of an expanded view of the INVEST study. These are all images from Dr. Kellner's and Dr. Melko's publications. But the INVEST study is looking at hematomas greater than 30 CCs. And the spot sign on the particular one is excluded. Expanding hematomas are excluded, but thalamic hemorrhages are included for that one. The MIND study, slightly smaller inclusion less or greater than 20 CC hematomas would qualify. Spot signs are excluded. Expansion is excluded. And this one thalamic hemorrhages are excluded. And then finally, the Dutch ICH trial that has an exciting group of multidisciplinary colleagues. Dr. Karen Quinn is one of the main leaders of that study. And they're looking also at endoscopic evacuation. This is a pilot study in what they expect will be 40 patients. So a lot of activity there. The group in Taiwan led by Dr. Abel Huang has also been doing great work. And they are now designing a further study for this device that they call the Absinthe Pen, which has an endoscope, a cautery and a resection tool incorporated all in one. And we're excited to see the results of that. They've presented the preliminary results of the technique with excellent outcomes. And this is my final slide here. So here we have sort of a progression. We have STITCH2, MISTI, MySpace, which was a multicenter study, and we have Apollo and Scuba. And looking at these, the numbers continue to get better in terms of cloud evacuation. And I think that's part of the success that we're seeing in these preliminary studies. The other thing that we're seeing is selectivity. The newer studies are trying to focus on either anterior basal ganglia or lobar, but trying to focus on those because of the delayed recovery that we still see in thalamic hemorrhages by excluding that patient population. So hopefully in the near future, we'll have the results of several of these studies available for review and discussion. And then subsequently, we'll be able to understand which technique might be better suited for a specific hematoma. If we're dealing with a hematoma in a cavity that is not active, that it seems to be fairly easy to aspirate, and that there are no major concern for hemostasis, those may be the best hematomas for endoscopic evacuation. And that may be the best technique for that. If we're dealing with perhaps something where hemostasis is more of a concern or there's an active spot sign, maybe a pore-based approach may be ideal for that. If we're dealing with a patient that shows up in a very delayed presentation, transferred from another hospital four days later with a lot of edema, maybe still decompressive craniotomy in conventional cloud evacuation is the answer for that. So those questions will still remain, but this refinement in studies and techniques, I think will bear fruit in the near future. George, I think I am about 25 minutes into it. So I think I'm gonna stop there. Okay. Well, Gustavo, that was a fantastic review of all the available information. And for everyone watching these webinars, this is really a due-to-force kind of review of everything that we have and where we're headed. We will hold any questions for now. And I will ask Dr. Mocco to share his screen. And we're looking forward to his presentation. Hello. Thank you very much. Always daunting following Gustavo. That was a fantastic review and state of the literature. It's a pretty exciting space and he's been a pioneer in ICH treatment and therapies. Thank you for the invitation. Very excited to be part of this process and this evening's discussion. I was tasked with speaking with an update on endovascular treatment of ischemic stroke. And then I chose to subtitle it, MEVO, medium vessel occlusions, large core, low NIH, et cetera, as a subtitle. Why did I do that? Well, you know, I sort of debated what would be best for tonight's talk. One option, which is honestly probably the more common way to go for these sorts of things is there are about, as many of you may know, a hundred different new stroke technologies and devices and catheters and stents coming out in any given year. And we could spend a long time showing cases and talking about that. But given the educational mission of this effort, of AANS, I can see behind you, George, there's a cap and gown head over AANS Neuro. I thought it was important that we talk about data and we talk about what's the state of the evidence, the state of the guidelines, and whether or not those guidelines are correct and where we should be going as a field. And so that's what we're focusing on. That's what I'm gonna focus on today. Wait, hold on. Okay, oh. So briefly, here are my disclosures. Some of them were actually mentioned in Dr. Padilla's talk that I'm the PI of the INVEST trial. I've also been the PI of the COMPASS trial and the therapy trial, which were both large prospective randomized trials that prove the benefit of one, stroke thrombectomy in general, and two, aspiration thrombectomy, which I think most people would say has become the standard of care first-line treatment. And then the other trials are, oh, positive is also related to this talk, which was a prospective randomized trial of a delayed stroke intervention, which was recently published and showed positive signal. The other trials are for other technologies, coils and liquid embolics and such, and I've invested in or consult for a number of companies, some of which are developing stroke technologies. I don't think I touch on any of them in this talk, but if I do, I'll be sure to mention them if they are. So to start talking about the current state of endovascular therapies and ischemic stroke treatment, I think we need to start with the lay of the land. Where do we currently rest in what we should be doing and how we do it? And that really starts and ends, at least in the United States, with the American Heart and Stroke Association guidelines. These are the documents that the vast majority of the community refer to. That doesn't say that there aren't other fantastic guidelines. The Society of Interventional Surgery has great guidelines on these lines. The AAS has reviewed, as you can see, endorsed by the American Association of Neurologic Surgeons, Congress of Neurologic Surgeons for these AA, American Heart and Stroke Association guidelines as well. There's also a European Society set of guidelines, but these are really the heart of it. And so when you go to the state of endovascular therapy for acute ischemic stroke, it comes down to this formal recommendation in that paper. And what that says is, you can read, patients who receive mechanical thrombectomy with a stent retriever, if they meet the following criteria, one, two, three, four, five, six. Okay, it lists what you have to have in order to get thrombectomy. And I would like to address those specific criteria because it's my supposition that there's a substantial need for meaningful research for us to understand whether we should or shouldn't be treating some of these subgroups in different ways. And also there needs to be some real clarity around the discussion of what did these subgroups mean? And there are large swaths of patients that are currently not receiving these therapies that really should receive them. So that's my hook of controversy, which I'll touch on a little bit. Before I get to that, I'm gonna start with a smaller controversy, which in my opinion, isn't really one. And that is the fact that in this recommendation, they say mechanical thrombectomy with a stent retriever. This was a hotly debated topic that neurosurgical leadership was very involved in over the last number of years, where extensive discussion was had for reasons that are unclear to me. The original writing group, this 2019 guidelines was an update to a set that had come out the prior, I think the prior year in 2018. And that group had relied on this language. As you can see, subsequent to this recommendation, immediately after, they have the following recommendation, which is that aspiration thrombectomy as a first pass for mechanical thrombectomy is recommended as non-inferior to stent retrieval with the same inclusion criteria. Essentially saying that we as physicians, if we have a tool, I'm not gonna tell you which bipolar card we're gonna use. I'm not gonna tell you whether to open the door with a 15 or an 11 blade. The goal is that you affect the procedure safely and quickly. And we now have class one evidence with multiple randomized trials. One of which was one of my disclosures early compass, which was published in Lance in 2019, which definitively states that aspiration thrombectomy is an appropriate way to treat this disease. So it's my opinion that the primary recommendation should have the stent retrieval reference blocked out. And what we should say is thrombectomy should be performed if patients meet these specific criteria. So given that, what do these criteria mean? What's the impact of these criteria and how should we respond to them? So I'm gonna take them one by one and we're gonna focus on the data here, okay? And so the first one I'm gonna talk about is pre-stroke MRS of zero to one. What does that mean? For any of you that aren't up on the modified Rankin score, it's pretty relevant in the neurology literature and the stroke literature. But essentially we're saying someone who has no symptoms that affect their life, okay? Once you're a modified Rankin two, you actually, you now have symptoms that affect your life, but you're still completely independent. Think about that. You're a completely independent individual. And yet, according to these recommendations right now, in terms of class one recommendation, you're excluded. Modified Rankin three is that you can still ambulate and you can still do many things on your own, but you have some level of dependency. You need assistance. Four is you need substantial assistance. Five, you need, and you can't walk. Five is you need constant attention. Six is you are no longer alive. So this is the current guideline recommendation. This is obviously based on the fact that in the initial trials, that was a requirement for inclusion, okay? But does that mean it should really be a guideline? And it's interesting. This is very debated. There are many centers around the country where we'll automatically exclude patients if they're a modified Rankin two or three or four, depending on what that center's thresholds are. There's a great paper that was published by Eva Mistry out of Vanderbilt with a number of collaborators, which you can see here, where they evaluated exactly this across two comprehensive stroke centers. This is a prospective observational study where they evaluate 761 patients, all of whom were either no symptoms, symptoms that don't affect their life but independent, or some dependents but still can emulate. They chopped the groups into two parts, the group that was dependent, who would have been excluded from the trials and is currently excluded from the recommendations, and those patients that had no dependents, and they looked at the good outcome rate. And guess what? Although there was a, when you just looked at the absolute numbers, there was a difference between the two outcome rates, there was obviously substantial selection bias in the patient populations how they present. When you did a multivariate analysis or you did propensity score matching, there was no significant difference in the outcomes between the cohorts, okay? Think about that, it's pretty powerful. The effect of thrombectomy was equal across the cohorts in multivariate and propensity score match analysis, which by the way was the intended analysis. Now there was a higher mortality, I think this is important to emphasize, you know, if one's concern is what their metrics look like and what their outcomes are, you're going to see more deaths in this patient population. Likely has to do with social network support, concomitant diseases and other things, but ultimately, that was pretty consistent and held up in the multivariate analysis. However, a large proportion of these patients returned to their baseline functionality, right? And the other really shocking thing is that on this large dataset across two different large comprehensive very busy stroke centers, a third of the patients, they just looked at anterior circulation, that's current guidelines, a third of the patients had disability or this modified rank in two or three level. This is a huge population, right? People estimate there's somewhere between 250 and 400,000 thrombectomies that would potentially be performed and that's what the old guidelines, but if you just take those numbers and you see 34% of them, we're talking about over 100,000 people who are not being made available to this therapy potentially, right? So in fairness to the guidelines, I will tell you there is a later recommendation buried down in the document that does say this, although its benefits are uncertain, it may be reasonable to treat patients with a pre-stroke modified rank in a greater than three. So we might treat these patients, all right? That's the current state of the recommendations. The reality is we need high quality observational studies. I think it's going to be extremely difficult to randomize this patient population. I think that most physicians feel pretty strongly about how this patient population should or shouldn't be treated, but you could do some really outstanding platform trial design observational studies that could get to the answer and build the evidence to support this patient population. So now I'm going to take everyone to a number two. Number two is a causative occlusion of internal carotid artery or M1 segment. Okay. So what they're saying is you should receive, you definitely should get treatment only if you have a clot in your internal carotid artery or your M1 segment. Think about that. It's powerful. What does that mean? What about more distal vessels? Well, the moment the trials that originally proved benefit came out in 2015, people started wondering this, okay? There's a great paper published by Jeff Saver and a bunch of collaborators that labeled themselves the distal thrombectomy summit group. And they concluded that somewhere between about a quarter and a third, maybe almost a half of all ischemic strokes actually represent distal occlusions and that IV tPA still fails to cure a lot of these patients. And so this is a patient population that needs treatment. And right away, you can see just within, you know, by the time it got published, it was two years later, but essentially immediately once we had evidence for ICA and M1s, people started publishing their results saying, hey, what if there's an M2, right? If I'm missing 50% of my left MCA territory, am I okay leaving the clot there? And people started treating these. What I want to emphasize is these patients have significant disability. So we're presenting NIH stroke scales in the mid-teens. These are very disabling strokes with high, high mortality rates. And this patient population is currently excluded from the, you should do this recommendation, okay? The best data currently that states surrounding M2s, and in fact, I'm going to submit to you that M2s do not belong in the conversation in which we should debate, okay? And that's because of this data. This is a collaborative data set of meta-analysis, patient-level data across multiple definitive trials that showed the benefit of thrombectomy versus mechanical therapy. And they evaluated in many detailed ways, the senior author was Mayan Goyle out of Calgary, the benefit, and you can see there was a substantial benefit to endovascular therapy versus medical therapy in that patient population. You can see the odds of a good outcome improved by over twofold, which by the way, is the same rate of the improvement of odds of good outcome for the M1 occlusions, okay? So this is a consistent effect size, a consistent benefit of intervention, and it reached significance in terms of value. Now, because as a post-hoc analysis, you can't technically say it's a significant study, but the statistical power is quite impressive, right? And in fact, they then did subgroup analyses looking at what if it was proximal M2 versus distal M2, what if it was a dominant M2, they looked at all kinds of breakdowns, and these are the two that were published in that initial paper. They provide supplementary material with the other ones, excuse me, which also had the same direction of benefit, but the number of patients is smaller, so the confidence interval wider, but you can see again, persistent, very powerful effect that has a likelihood of greater than 95% of being real for the treatment of M2 occlusions. So I'm going to start with the supposition that if you're at a thrombectomy center, M2 occlusions should be treated. That is the current state of the evidence from high quality prospective randomized trials. But then what about other diseases? What about what other locations, other places to treat? Another friend of all of us on this talk, Pascal Jabbour and Stav and the rest of their group in Jefferson and some others, Iowa, they collaborated and published out of a cohort of 453 patients that almost 20%, almost one in five patients had distal occlusions. Now the majority of those were M2s, but as you can start to see, we're seeing M3s, M4s, A2s, PCAs, we're seeing treatments of much more distal occlusions. The technology has improved to the point where these make sense. Looking at these distal occlusions are meaningful, right? We're still patients presenting with significant disabilities and we're still getting excellent technical outcomes. Even more importantly, we're getting two thirds of the patients being independent on their 90 day follow-up, right? So despite having severe strokes on presentation, they're getting back to independence. And perhaps the best paper on this specific topic most recently was published in 2018 out of Emory, Dr. Perdia's institution, Jonathan Grossberg, one of the neurosurgeons there and Raul Nogueira and one of the neurologists. And this was a retroactive review, but they identified 69 patients with distal thrombectomies. We're not talking about M2s now, right? So before we're talking about 10% of a series being distal. Now we've moved to 2018 and we're talking about all of these patients are much more distal vessels, ACAs, PCAs, M3s and beyond. And guess what? The patients still had severe deficits, right? If you've got occlusion to your angular branch that's going to your language area and your motor area, you're going to have a significant stroke, right? 45 of the patients presented primarily with only a distal occlusion. The other 23 were patients that got distal thrombectomies, but they also had a more proximal disease. This is the location of the various ones. I won't go into the detail. It's important to note, very high recanalization rate, extremely low complication rate, no episodes of vessel perforation or extravasations. And about one third of the patients were achieving independence, which remember was about the same number as Dr. Mistry's study out of Andy, right? Same kind of thing. Very, very interesting, except for this group, because they don't have baseline disability, have comparable mortalities than all the other trials with a 20% 90 day mortality, right? And so, because I'm a neurosurgeon, I can't get through a talk without showing at least one case. This is a 83 year old woman, normal stroke risk factors, presents with a severe stroke. And she only has a distal, very distal, at the end of M3, M4 kind of range occlusion. You can see it here. We have a little blue arrow pointing at it. You can see we're extremely far back, but that little tiny artery is going to her motor cortex and her language function cortex, and so she has a severe stroke. And this is going to be devastating to this patient and their life, right? So here's a little video. We're able to put a little wire out there. We're able to bring a catheter, a 35 aspiration catheters, a zoom 35, which is also one of my disclosures earlier, a relationship with the company. But we're able to go out to this distal occlusion and you can see the before and the after the branches open. There are two subsequent branches that feed critical tissue. That tissue is now saved and that patient had a tremendous outcome, complete resolution of the pleasure, substantial resolution of the aphasia with only minimal deficits and I stroke scale three, 17 to three. If that's me, if that's my mother, if that's my wife, I want that treatment. And I think we need to start thinking about these. Now the recommendations, well, as I said, this, these lesions do not fall into the should be done, but they do have a recommendation that says it may be a reasonable for carefully selected patients when it's in the more distal arteries, right? The reality is this is a patient population where we probably should do randomized trials, not the M2s, but M3s, ACAs, PCAs. There is a, there is a potential for harm. And I think designing high quality randomized control trials to evaluate this population is probably indicated briefly. One of the other recommendations that you have to be greater than or equal to 18 years of age. Again, this is just a carry over because that's what was done in the prior patient populations of the trials. But I think this is silly. It makes no sense that we would prevent pediatric patients if someone comes in and they're 15 or 14 or 12 or five and they have a clot and you can safely take it out, you should do it. The largest series out there that I'm aware of is this excellent review that was published out of Australia. I was able to be a co-author with my partners on this where we pulled, well, I don't have the details, but we pulled a large number of pediatric patients. I can tell you my personal experience of treating a five-year-old with a basilar occlusion that sat for over 12 hours at other institution before they came over. An MRI was done which showed substantial DWI changes of the brainstem. There's a whole nother talk we could give about the validity of early stage DWI and assessing final impart volume, which I'll just say is skeptical, I'm skeptical of. But this patient had substantial DWI changes in their brainstem, but he was a five-year-old kid. And so I talked it over with the stroke team. We said, let's give it a shot. We went in, we took the clot out. I can't remember now. We published a paper so you can check on me, but it was two or three days later, he went home completely intact, right? DWI changes reversed. His blood flow went back and he returned to normal. And I think we really have to be very considerate of providing this treatment to children. There's no way we're ever going to complete a meaningful randomized trial in that patient population. We need high quality observational studies to show us one way or the other. The next group is low, the current recommendations say they have to have a severe stroke, NIHF six or higher. Again, this is a carryover from the trial. The reality is there's a large number of patients who present with minor strokes, right? A third to half of all the patients with ischemic stroke. And yet, even when they present with small strokes, almost a quarter of them still have dependent outcomes. That's because early neurologic deterioration happens. Early neurologic deterioration happens somewhere between a quarter to about 40% of patients, right? If that happens, your likelihood of a poor outcome goes as high as 60%, goes up dramatically, right? If you have a large vessel occlusion or non-invasive imaging with minor stroke symptoms, the likelihood that you have an early neurologic deterioration over doubles. That's substantial, right? And in fact, if you take patients who present to hospitals with minor strokes, and then they have a neurologic deterioration, if you image them, now the rate of LVL goes up to almost half. So the other interesting thing is this, that deterioration typically happens within the first day, right? Five to 16 hours. But the vast majority, over two thirds of the patients with minor stroke symptoms don't ever get vessel wall imaging, or vessel imaging. So we don't even know if they have large vessel occlusions, to be honest. We estimate that it's somewhere in the ballpark of 70,000 a year based on some population studies. The references for all these are at the bottom of the slides. But if you have a minor stroke and you have a present LVL on imaging, you have seven times higher likelihood of having a poor outcome than if someone presents with a minor stroke without a large vessel occlusion. It's pretty impressive difference, right? So we go back to that Hermes trial, patient level, large prospective randomized data. Unfortunately, the majority of the trials, three of the five excluded patients with minor stroke symptoms. So our data set is a bit limited, but here's their outcomes. I want to emphasize that although the confidence intervals clearly include one, you can see that the point estimate for the effect size for patients with minor strokes less than 10 still was very much in the direction of benefit for treatment, approaching two. So it seems very likely that this patient population would benefit. Now, my own personal disclaimer, I do not treat minor stroke symptom LVL, right? This is very debated. A number of my colleagues disagree. I know if you're in Europe, particularly in Germany, they will take this patient right away. I was on a call, I was on a panel in Australia recently, and they were all laughing at me for this. They were like, it's crazy. If there's a large vessel occlusion, you go fix it. This is an ideal situation. I will tell you that when you take a patient who's a good functional status, NIH strokes have two or three, very minor stroke, you do an intervention on them. And then afterwards, they're at NIH strokes, you have 12 or 14, that's a pretty tough one to swallow. In my opinion, right now, I put those patients in the ICU. I watch them for any change, and then we intervene on them, and I'll talk about that in a second. I'm not going to go into the details. This is a whole bunch of different trials that looked at minor stroke with large vessel occlusion. And the end result is it's very mixed results, right? Here's a paper, no improvement in outcome, potentially higher complication rate, right? It's all over the place. A great study by Siraj et al, where they looked at the same sort of thing. Now, this is interesting, right? These are minor strokes with occlusions. If they had M1 or ICAT occlusions, there was a strong signal to benefit for thrombectomy. If they had minor strokes, but the occlusions were more distal, that benefit was less impressive. It makes sense, right? And just one more further one, again, not definitive, primacing signal of benefit. And then most recently, I was able to be part of a group led by Adam Arthur out of UT Memphis, where we looked at over 250 patients with minor strokes, and we did not see a benefit in outcome. So this is a very debatable topic. This is why I like to observe this patient population right now. What's interesting is clinical fluctuation is one of the strongest predictors. So if you have one of those patients present with a large vessel occlusion, they do look really bad, then they look better. Or they come in looking good, then they look bad, then they look better, right? So those patients have a much higher likelihood, 17-fold, but with huge confidence intervals ranging from 2 to 160, but a 17-fold increased likelihood of having a bad outcome, right? The other thing we know is if you treat, if you do what I'm suggesting, which is rescue therapy, you don't treat them right away, you watch them, then if they worsen, then you treat them, we have much worse outcomes, right? Because many of us are aware of that patient. You put in the ICU, you say, absolutely, if there's any changes, let me know. And then that morning, the residents call you and on 5.30 a.m. rounds, they discover the patient's plegic and no one sends an alert out, right? So we all fight against those kinds of things from happening, but it does happen. And so rescue therapy has worse outcomes than upfront treatment for this disease, right? So we know early neurologic deteriorations associated with poor outcome. We know therapy protects against that. We know that doing it earlier is better, but we also know it carries risk, right? Is there a better way of identifying the patients? Actually, I didn't have this in my disclosures, but I'm running a trial where we're trying to answer exactly this. We're following minor stroke, low NIH patients and seeing, can we identify the patients? Can we find the 40% that appear to worsen, identify them through some other means and therefore bring those patients to treatment without exposing the other patients who aren't going to worsen to the risk of treatment. That in my thinking would be the ideal, but that may not be the case. It might be that there's a strong enough signal, like there is in thrombectomy in general, that we should treat everyone. And we're going to save that 40%, and we're going to harm a small percentage of the other 60%, but the rest of them will be okay. And so it's worth doing. We don't know the answer. There are ongoing randomized trials that have recently started along these lines. One of them is actually out of, one of the PIs is out of Dr. Padilla's institution, Raul Noguera-Endelo with Puja Country. So there are options out there to evaluate this population. As I mentioned, one of the questions is how safe is it? Can we apply it to the patients that wouldn't otherwise worsen? And the data is beginning to suggest, this table highlights from a recent publication of small numbers, but that the complication rates are comparable to established rates, and it's not an unduly devastating intervention. The current recommendation, again, may be reasonable for patients with low NIH stroke scales. This is a population we absolutely need randomized trials. If there's one population I feel the strongest about, it's this one and the next one. We need randomized trials to answer these questions. And so then I'm going to focus on the last one, really. I'm not going to spend time on item six, because since this time, we've had subsequent trials, including positive, the one that I mentioned before that I'm PI on, as well as Dawn and Diffuse, which have demonstrated that treatment after six hours is worthwhile. The last group is what about the patient that comes in with a large vessel occlusion who already has a large established stroke, right? What if their aspects is greater? Currently, you want to aspects is greater than or equal to six. What other aspects is lower? Aspects is a scale that's right off of the plain CT or an MRI to establish how much tissue is dead. So if there's a lot of dead tissue, the argument has been, well, you can't really help that patient. So there's no point in revascularizing the vessels. I'll tell you the data raises question with that supposition. This is a fantastic paper by Vincent Costellat out of Montpellier in France, where they looked at low aspects patients, large established core patients, and yet found a substantially strong signal towards benefit, right? And you can see, this was a comparative study. This isn't randomized. Okay. It's important to note that, but you can see the interventional group, they were 75%. They were successful in recanalizing and they had a much higher rate of independence, right? 30% of the patients received independence versus 2% in the control group. This is significant, right? There's also another paper out of France. As I mentioned to you before, in Europe, they're really pushing some of these boundaries. This is by Bertrand Lapergue and Michel Piotin out of Foundation Rothschild in Paris, but they coordinate the endovascular treatment and ischemic stroke research investigators. They published this paper. Now this is a little funny, right? Because if you'll see what they're looking at, they're looking at this same patient population, but they all got treated. It's just, which ones did we get open and which ones didn't we? Obviously, if you do the treatment, you want the one that's better, but the signal was extremely powerful. And then they followed that up with a subsequent paper. Again, if we get the blood vessel open, the patients do much, much better. This is similar to the old mercy data that came out in the early 2000s. It's not definitive, but it definitely shows there is a treatment effect. Now the question is, are we harming patients when we do that? And that's the question we don't know. Most recently, in fact, very recently, Greg Albers and collaborators published a paper looking exactly at this from the SELECT study. Again, these are from large sort of observational registries. And they looked at 105 patients that they treated that they managed with large core. Those that they did throw them back in before and those they didn't, and you can see over threefold odds of good outcome. If you did the intervention, that confidence interval did exclude one, means over 95% likely this is not statistical randomness, but this is an actual effect. That said, they concluded, which I think is right, we do need randomized trials for this patient population. We don't really know. There's too much selection bias in these observational studies. Another great paper, again, by Saraj, we mentioned earlier, and Georgios from Greek, a great stroke neurologist out of Greece, where they did the same sort of thing and pooled, they published their own stuff and did a systematic review meta-analysis. And again, you can see a almost fourfold, or actually the odds ratio is greater than four of a likelihood of good outcome if patients got thrombectomy, right? Very powerful. Again, they still, they concluded we need randomized trials and the current recommendations it may be reasonable for patients with large cores to do intervention. If there's a population to galvanize behind, this is exactly the population we need to figure out and randomize. There are some trials starting and ongoing, I'd certainly encourage everyone to be a part of it. And with that, I'll say thank you very much. And thanks for the invitation. Wow. Another spectacular presentation. I'm not sure I've seen many presentations where it's such a good amount of information is just squeezed into one talk. It really makes your head spin, but it's also a testament of how much this field has exploded. And we're talking about odds ratios that are four, five, six, seven, 12, that it's really a revolution in medicine. And once in a while that such a thing is seen. Well, a question for Dr. Mokos, we have all this amazing data. What do you think is going to be the next major frontier in treating ischemic stroke? What is going to be the thing that is going to change gears again in how we treat this disease? Sure. Well, okay. I'm going to, I'm going to give you a two part answer. Part number one is the thing that's immediately on the threshold of happening is I think that we're going to demonstrate the value of treating these medium vessel occlusions, these much more distal things, M3s, M4s, ACAs, PCAs. Technology's gotten so good. The likelihood of getting a good outcome and recantalizing those are so high. I think that's going to happen and that's going to make, that's going to dramatically affect what we do and how many patients get treated and how often I get called and woken up at night to go in. But I do think that's, that is the immediate threshold. And in some, in many places it's already happening. In terms of sort of paradigm changing, mind blowing kind of directions, I think, I think there are, there are huge changes going on with the technologies we're now bringing. We're now bringing catheters that we used to be afraid to put in the neck all the way up into the MCA. I mean, we, the technology has just changed. It's so much more navigable. We can bring giant 88 thousands of an inch, you know, huge, huge cannons up intracranially. So I think that's going to make a big change. And the other thing, and this is a little broader, but you know, if you think about whatever, 35 years ago, cardiac intervention was, you know, you go and you use a balloon, you open up the blockage. That was really it. Right? Then they got stents, a little bit of this, a little bit of that. Now it's a whole field, right? Structural heart. You replace valves. You ablate circuitry. You stop heart seizures endovascularly, right? The reality is we're going to be doing that for the brain not long from now. And I think that's going to influence this whole space. One of my disclosures, I'm involved with a company called Syncron, but we've developed a transvascular brain computer interface. You go through the veins, you drop an electrode in the sagittal sinus, you can read the motor and premotor cortex. We've put it in four patients in Australia. They can use their mind to control their computer, their laptop at home, just a Bluetooth. That's the direction the field's going. I think if there happens to be any residents looking at this or anything, I think it's pretty darn exciting. It's going to be a whole nother field. Structural heart. Pretty soon there's going to be structural brain, right? There are people working on endovascular treatments for shunts, for hydrocephalus. I don't know if any of you guys know Adam Malek or Carl Hauman, but they've pioneered some pretty amazing technology in that direction. It won't be long until there's some pretty cool things going on in those ways. Do you think the evolution of networks utilizing AI technology, early identification and routing, things of that nature, how far do you think we are from instituting in, let's say, in a nation like the United States where things are progressing rapidly? I mean, at some level, we're already doing it. So there's a whole nother talk, but FDNY, in New York City right now, all EMSs are trained. I think the same thing is going on in Atlanta, Gustavo can comment, but all EMS is trained on a stroke severity scale assessment. So if they have a suspected stroke, they apply the scale. Now they're still limiting it to the first five hours of presentation, which is silly, but patients are getting evaluated. If they have a positive stroke scale, they're getting routed directly to places that have registered to be able to treat these patients. And that has to be validated by a outside certifying organization. So that's just what scales. And then you have technology. One, again, I'm going to talk about the one I'm most familiar with, which I have a conflict with as a consultant, but this AI, I don't know how many of you have used it. It's been a complete game changer for us because when I deal with the radiologic side of my job, as you guys know, it's often a hassle. You got to open your laptop, you got to VPN in, you got to log into the thing, and it's made by radiologic engineers. And I want something that's made by the same guy that makes Uber Eats or Grubhub or something that's super easy. And that's the way I feel about Viz. They have AI algorithms to identify the occlusions. They alert the treating physician. So it's not just letting the radiologist know to read it quicker, but it's telling me right away before the radiologist has even looked at it that they think there's a clot there. And I can pull it up on my phone and look at it in two seconds. Face recognition, I can zoom in, zoom out, window it. And there are other companies doing similar things. I don't want to just harp on that one. It's just the one I know. But you talk about data analytics and AI being applied. It's being applied right now. We can do more. I mean, at some point, we're going to have self-driving ambulances that the AI tells to drive to the nearest drug center. We're not quite there yet, but I think it would have been hard for me to imagine that we're already to the point where we are even seven years ago. So the rate of change is impressive. We're very fortunate to have that. A question for both of you, but I guess we'll start with Gustavo. Gustavo, do you think that we've designed the most impactful trial for hemorrhagic stroke yet? No, I don't think so. What are the components of trials that would be, in your at least opinion, ideal to evaluate this disease? One of the challenges with these trials is that in order to maintain trial efficiency, you have to compromise on factors that somewhat can limit its applicability to real life. And that's one of the challenges. We try to put, for example, ICH in boxes and say, this is thalamic, this is basal ganglia. Many of these are multi-compartmental. They have different degrees of IVH. In trying to box all of these into one roof, it's just hard. That's the same thing that happens in traumatic brain injury when you try to do trials and you say, well, this is severe traumatic brain injury and the spectrum of radiographic and clinical findings that are put into one box is hard. But I think trial efficiency is getting better. To Dr. Malko's point, using things like this AI and maximizing our ability to screen and enroll patients, leveraging those technologies, is making our ability to do trials much greater. So I think we will struggle still with defining the inclusion and exclusion criteria. But when it comes down to screening and enrollment, we're getting a lot better. And if that's the case, the speed with which we can complete trials is going to get faster and faster. In terms of analysis, all of the trials that we used to do using methods like the ABC over two to calculate volumes, for example, are being replaced by volumetric analysis. These sort of platforms throw that in there as well. In terms of trial design, I think Bayesian design, revolutionary clinical trials and adaptive trial design gives us the ability to do smaller sample sizes and detecting effects earlier and preventing exposing patients to interventions that are not going to be beneficial. The ability to do those interim analysis and make predetermined decisions was critical for the success of MRCLEAN and some of the other early thrombectomy trials that led the way to the things that we have today. So I think we'll see a combination of those things. Better biostatistical design, better trial efficiency, and that in conjunction with all of our tools getting better, I think the future is bright. Dr. Mocco? I couldn't agree more. The technology matters, the way things are changing and evolving. I recently got upset at one of Dr. Perdia and both of our friends. I'm running the INVEST trial, which he mentioned, which is a competitive trial with ENRICH. This guy enrolled a bunch in ENRICH. And I said, what, do you like Gustavo more than me? And he's like, no, I got this Viz AI thing, and it alerts me when there's an ICH for his trial. And I was like, dang, that's pretty cool. So these things matter, and it's going to influence how we do trials. I'm a little less in love with Bayesian designs, although they have their place and are extremely useful. But I think that we also need to utilize the technologies we have. We need to push the field and push the bounds, but we need to do it in rigorous scientific ways, which I can only give incredible congrats to Gustavo for what they're doing with ENRICH. We should do this work, but we should do it in a way that we're studying it, not just in the sort of old school nerve surgery anecdotal, well, I know it works, so I'm going to do it kind of way. I think that's meaningful. I think we also have to get our heads around the binary outcome interpretation of results. So you design a trial, and you have to power that specifically for certain statistical outcomes, and it would be false to do anything else. But that doesn't mean that the rest of the data that comes out of the trial is not meaningful and shouldn't be used to guide our decision making. And sometimes I worry that we've become overly dependent on quote unquote guidelines that by their nature have set up rules that are based on only a particular type of evidence, which is level one RCT evidence, and therefore a lot of patients fall into gaps. And I do think that's a problem because it's in those spaces, it's in those gap spaces that we drive the field forward. It's in those spaces that all of us that are on this talk and all the nurse surgeons are looking are figuring out the best ways to provide the best clinical care for their patients. And so I'm a trialist. I love running trials. It's the main thing I do. But it's silly to say that's the only way to generate evidence, and we have to pay attention to all aspects of evidence and let that drive us in the directions we need to go. Well, I wanted to thank you again both for sharing your wisdom tonight. It was fantastic presentations and discussion. I certainly learned a lot tonight, and I'm sure everyone watching this will as well. I know it's at the end of a long day for everyone. So with that, thank you and have a great night. And I look forward to seeing you both in person with the first occasion again once everything is coming down now. Amen to that. All right. Take care. Thank you. All right. Good night. Thank you. Shannon, thank you so much. All right. Have a great night. Thank you.

cerebrovascular surgery

ENRICH trial

intracerebral hemorrhage

endovascular treatment

ischemic stroke

stent retrievers

aspiration thrombectomy

thrombectomy

randomized controlled trials

AI screening

volumetric analysis