Jeff Saver (00:01): Hello, everyone, and welcome to this talk on best practices in the use of endovascular therapies and stroke here. In the course of these 20 minutes, we'll race through the determinants of successful reperfusion, the magnitude of benefit of thrombectomy in the early time window, national quality improvement programs for thrombectomy in the US, imaging selection in the later window, and the recent topics of bridging intravenous thrombolysis versus direct to thrombectomy, and treatment of acute basilar arterial thrombosis, ending with a look at organizing EMS regional systems of care. Jeff Saver (00:49): When we undertake endovascular thrombectomy, we currently have two major classes of devices that interventionalists choose among, the stint retrievers and the aspiration catheters. Each of these has their distinctive advantages and disadvantages. The stint retrievers embrace the entire clot, and thereby give more control over the clot. The aspiration catheters suction the tail of the clot. They are able to start the removal faster, since they require less time to place, but they don't have control over the distal end of the clot. Jeff Saver (01:40): The first 20 years of thrombectomy taught us a lot about what works and what doesn't, and what makes treatments likely to work. Sometimes, we have tiny little clots that we pull out, but big enough to block the three millimeter diameter of the M1 and cause a devastating occlusion, and sometimes, we have giant snakes of clots. Here's a 15-centimeter clot taken out from the origin of the internal carotid up through the M2. You can easily imagine that the lytic drugs are not going to be able to dissolve this large burden of thrombus quickly. Jeff Saver (02:22): And also, clot composition matters. White clots, red clots, there's calcium, there's neurovascular TRAPs that affect responsiveness. And overall, between clot burden, composition, the tensile properties of the clot, the forces we can apply based on the branching of the target and the shape of the clot determines what device to choose in achieving success. In these days, often both aspiration and stint retrievers are used at the same time, to maximize effect. Jeff Saver (03:05): In general, if you're going to select just one, the recommendation is for fibrin-rich white clots, those are the best for aspiration. They're cohesive. They'll hold together. You can pull them out quickly. If you have erythrocyte-rich red clots, they can fragment, and using stint retrievers will be helpful. You can sometimes tell this by looking at hyperdense vessel signs, which will tell you that it's a red clot. Jeff Saver (03:42): So, from the time endovascular therapy first started being used, we saw in thrombolytic therapy, the cardiologists going from streptokinase to PTCA at right the start of the millennium, confidently achieved complete reperfusion in 85 to 100% of their cases when they would go to the cath lab. We have a more complicated organ in the brain, and from the tPA era through the initial device era, until 2012, with the development of the stint retriever and aspiration catheters, we're not getting arteries open very often, but that led to the breakthrough in 2012, the positive trials in 2015, continued improvement in devices, and now, we also are getting 85, 90, 95% of arteries open. Jeff Saver (04:54): It was in 2015 that the annus mirabilis of stroke year occurred, with a positive thrombectomy trials coming in one after another, and establishing this as a proven therapy in the guidelines, and based on these trials that were concentrated in early time period, the highest level recommendation in the national guidelines to use thrombectomy for patients who had internal carotid or M1 occlusion, mildly severe strokes, aspects greater and equal than six, and treatments start within six hours of onset. Jeff Saver (05:35): When you use it that way, which is the most common deployment, this is a very effective therapy. Looking at dichotomized outcomes, being independent or not, disabled or not, one out of four patients treated with thrombectomy achieves better dichotomized outcome. That compared to about one out of 10 treated with lytics, and about one out of 25 treated for STEMI with PCI. So this is really a dramatic therapy, and it's even more dramatic when you take into account improvements over the entire range of disability, not just a single transition point, where it's one in two-and-a-half patients benefit. Jeff Saver (06:21): A key aspect of applying this therapy in the early time windows, when we use broad selection imaging criteria, is that there's a rapid drop-off in benefit with delays in start of therapy, up through six to eight hours. You can see here that the benefit per 100 patients treated starts out at 43 and declines to 18 steadily during this time period. In fact, minutes matter. For tPA, or tenecteplase, every eight-minute delay causes one out of 100 treated patients to have less improved outcome, but for thrombectomy, every four-minute delay causes one fewer of 100 patients to experience reduced disability, so every minute counts. Jeff Saver (07:21): Therefore, the American Heart American Stroke Association launched the target stroke campaign phase three. Phases one and two had focused on lytic therapy, but phase three included, for the first time, a door-to-device placement [inaudible 00:07:39] that if the patient arrives from the field, delivered by EMS directly, the door-to-device on clock time should be less than 90 minutes in half of patients, and if they arrive by transfer from another facility so you have time to get ready, it should be on clock within 60 minutes. Jeff Saver (08:00): The program distributes this set of recommended best practice strategies that have some evidential base in the literature, including prenotification of the neurointerventional team, going directly from brain imaging to angio suite, skipping the ED, having rapid anesthesia protocols, and the others seen here. We've developed this patient education tool to let families and patients be able to be informed to make rapid decision-making, and here you can see out of 100 patients treated, the number who benefit in green, and the number who have worse outcome in the end in red. Jeff Saver (08:49): Now, the important next development in thrombectomy was the extension of the time window beyond six hours, and this occurred as it was recognized that patients differ in their speed of progression of stroke. Some of fast progressors and complete their infarct within a few hours, but some are slow progressors, and continue to still have salvageable tissue up to 24 hours, and some patients will have variation at different time points between fast and slow progression, in the same patient. The speed of infarct progression is determined by a variety of factors, how well the heart's pumping, how the quality of the blood delivered, the tolerance of the tissue to ischemia, and most importantly, the collaterals present in that individual patient. Jeff Saver (09:51): So there's a lot of interindividual variation. We know that two million neurons per minute, on average, are lost in a large vessel ischemic stroke, but in some patients, it's going to be much more, in some patients much less. Nonetheless, even half speed, one million per minute, is truly important, so we still need to race to the bedside in all patients, but we can use, beyond six hours, multimodal imaging to determine patients who still have salvageable penumbra, and are therefore likely to benefit from therapy. And we've seen from the trials that within the first six hours of onset, just about every patient will have, if they have large vessel occlusion, salvageable penumbra, and doing fancy extra imaging may not be necessary. Jeff Saver (10:47): But beyond six hours, more and more patients will have completed their infarct, and the slight extra time it takes in this era of automated analyses to get penumbra imaging helps to select the patients who will be responsive to therapy. Here are the two pivotal late window trials, DAWN and DEFUSE 3, which with imaging selection, show about the same major treatment effect as under six hour therapy for broad patients. Jeff Saver (11:22): Now, very exciting development within the past year is proof of benefit of endovascular thrombectomy when treating basilar artery occlusions, not just anterior circulation occlusions. The first two trials that had been done in this space were not positive. They showed a trend in the favorable direction, but did not reach statistical significance, probably for a variety of reasons that included myodeficit patients, included patients with extensive infarcts by the time they arrived, and because people were afraid to not do thrombectomy, there was a high crossover rate increasing use outside of the trials, causing selection bias. Jeff Saver (12:16): But when these came in neutral, that spurred the community to do trials that improved upon these aspects, and this year, the ATTENTION trial and the BAOCHE trial both have come in strongly positive, showing doubling in the rate of ambulatory outcome, ranking zero to three, in patients with basilar artery occlusion with thrombectomy. Here is the... If you take all the trials, the degree of benefit is about a 10% absolute improvement, and the P value has four zeros before the one, so highly statistically significant. In some of the publications reflecting the excitement about our achieving demonstration of benefit in basilar, shown here, we have finally conquered the basilar. We have resurrected EBT for posterior circulation stroke. Jeff Saver (13:31): Another recent development is analyzing, in trials, the benefit of skipping lytics on the way to the cath lab versus giving them, and each approach has physiologic advantages and disadvantages. Giving the lytic could open the artery before thrombectomy starts, but it could delay the start of thrombectomy. Giving it could change the clot composition in a manner that makes it more responsive to EBT, but it could also dissolve it and cause it to distalize, making it less responsive. And giving thrombolytics could dissolve residual or newly embolized distal thromboli that are present after clearing out the large vessel, but giving the lytics increases the risk of symptomatic hemorrhage and increases care cost. Jeff Saver (14:28): So, we have had six randomized trials come in over the last two years looking at the benefit of skipping versus not skipping, and as you see here, one aspect is that the rates of reperfusion with lytics alone before thrombectomy tended to be low across all trials, and only mildly increased in patients who received lytics, so they had a mild beneficial effect with the opening, but not major. And there was not a major difference in the degree of final successful reperfusion. There was mildly more symptomatic hemorrhage when lytics were used, and overall, as you see here, both approaches ended up fairly close together, independent outcome achieved in 49% of patients with direct versus 51% of patients with bridging. Jeff Saver (15:44): However, there's a difference by region, with more Asian patients having less positive results, or more positive results from skipping, but non-Asian patients having less beneficial results from skipping, so the European guidelines issued within the last few months have recommended, in general, sticking with bridging therapy for most patients. To my mind, the clinical implications are that if you... Certainly, if you're not going to be able to get there, and you're not sure you're going to be able to get there with your catheter, then start lysis. At least you'll have given the patient that chance, and do that if the patients [inaudible 00:16:37] outside hospital, where we get to the cath lab quickly, or if they have a very tortuous or chronically blocked access to the intercranial circulation. Jeff Saver (16:49): And then, for patients in whom you can reach the clot, nuanced, tailored decision-making. There are several features which increase responsiveness to lytics, hyperdense vessel sign from dispersiveness, low clot burden, and treatment within the first 60 minutes of onset, as in mobile stroke units. In those settings, strongly favors lytics. But when there's increased bleeding risk, like when you think you may have to stint a carotid stenosis in the neck or intercranially, then you'll want to use double antiplatelet therapy after the procedure to protect the stint. Having lytics on board, increasing bleeding risk, might be avoided, and other risk factors for increased bleeding with lytics might be reasons to skip lytic therapy in individual patients. Jeff Saver (17:49): So finally, let's talk briefly about organizing and reorganizing our systems of care to recognize the dramatic benefit that thrombectomy provides and delivers many patients as possible to thrombectomy centers. We have a four-tier system. Two of the tiers are at the level of spokes, acute stroke-ready hospitals, that can give lytics by telemedicine and then ship the patient, primary stroke centers that can give lytics and do organized inpatient care, but don't offer thrombectomy, and then we have two levels of thrombectomy centers, thrombectomy stroke centers, that add thrombectomy to the [inaudible 00:18:44] care and comprehensive stroke centers that also add neuro intensive care units and sophisticated neurosurgery, that do thrombectomy. Jeff Saver (18:59): We have a challenge, that for patients in who the occlusion is in a medium or small vessel, they will benefit most or only from lytics, and not from thrombectomy. So we want those patients to go to the nearest hospital offering thrombolysis. But in patients with large vessel occlusions, lytics, as we've seen, work poorly, and thrombectomy works very well, and if it's not too much further away, we'd like those patients to go to comprehensive stroke centers. Therefore, a variety of prehospital stroke severity tools have been developed to help paramedics distinguish patients likely having large vessel, more severe strokes versus milder, and LVOs versus non-LVOs. Here's an example of the simplest, the Los Angeles Motor Scale, three items, and probably the most complex, the RACE scale with seven items, including judgment of aphasia and agnosia. Jeff Saver (20:12): When these scales have been compared, in general, they all do pretty much the same. They all have about 70% accuracy. They certainly help with enriching the number of patients going to the right place, but they don't perfectly satisfy our needs. Given this, there have been modeling studies of when patients should go to comprehensive stroke centers versus when they should go to primary stroke centers, showing that it varies on the performance of the centers, and EMS medical directors can use this type of modeling to set up systems of care appropriate for their area. As an example, here in LA, we have 10 million patients, 10 million persons, and 50 designated stroke centers. Jeff Saver (21:11): In 2018, we instituted a two-tier routing policy, where if you scored four or more on the modified LAMS scale, then you were thought to likely have a large vessel occlusion, and you'd go to the nearest thrombectomy-capable center if it was within 30 minutes of where the patient was. Otherwise, go to a primary stroke center with accelerated interfacility transfer. And we found that by putting this system in place, we dramatically increased the number of patients with access to endovascular thrombectomy, because many hospitals stepped up to become thrombectomy centers when they knew this was coming. We increased the number of ambulance transported patients who got thrombectomy from 5% to 14%, and reduced interfacility transfers from 4% to 1%. Jeff Saver (22:11): Now, we need to do better in the field. One way to do that is to have mobile neurology evaluation of patients in the field. Medicare now pays for this, but just how much better neurologists can do with their exam over the stand device paramedic exams, it's probably mild but not major incremental benefit. Very exciting are the emerging helmet technologies for use by paramedics, that use a variety of energy techniques [inaudible 00:22:48] microwave, brain vibrations, and other to determine if a patient's having stroke versus non-stroke, hemorrhage versus ischemic stroke, and/or large ischemic versus small ischemic stroke, and improve routing. None in practice yet, but some getting close. I suspect in five years, we will be using one or more of these. Jeff Saver (23:12): So, it is a very exciting time to be practicing vascular neurology, and to be able to offer endovascular thrombectomy to patients, and I hope this update of where we are has been helpful to everyone. Thank you.