[Automatically generated] From the JAMA Network, this is JAMA Neurology Author Interviews. Conversations with authors exploring the latest clinical research, reviews, and opinions featured in JAMA Neurology. My name is Dr. Cynthia Armand, Associate Professor of Neurology at Albert Einstein College of Medicine, Montefiore Medical Center and JAMA Neurology Web Editor. Today, our topic is stroke, atrial fibrillation, and anticoagulation. I have with me Dr. Urs Fischer, who is a professor and chairman of the Department of Neurology at the University Hospital in Basel, Switzerland. He's also the current president of the Swiss Neurological Society and former Secretary General of the European Stroke Organization. Dr. Fischer, thank you for joining us today. Dr. Armand, thank you very much for having me. I think this topic comes up over and over again, especially if you are practicing neurology in any hospital, right? Stroke holds come in all the time. And the question always is we look at the size of the infarct, we find atrial fibrillation, we determine the cause, and then the treatment, anticoagulation. I want you to give us a little sense of what the current guidelines are in starting anticoagulation in a patient immediately post stroke. And tell us a little bit about what inspired your current study. Yes, thank you very much. So the current guidelines, when to start anticoagulation in people with ischemic stroke and atrial fibrillation are mainly not evidence-based. These are mainly recommendations by experts. And the main challenges for physicians that we know that if we could start anticoagulation early, we could prevent recant ischemic events, especially stroke. Nevertheless, we are also scared that if we start early, we produce an intracerebral hemorrhage, which might be fatal for our patients. So therefore, there is a lot of anxiety for us as physicians. The current guidelines say that, if you look at the American guidelines, that you eventually should wait 10 to 14 days until you start anti-calculation in these patients. In Europe, we more follow the expert recommendations saying that, after the TIA, you can start immediately. In people with minor stroke, you have to wait 3 days. In moderate strokes, 6 days. And in major strokes, 12 to 14 days. And that mild, moderate to severe stroke, what are the parameters for determining that in Europe? So the experts said that this classification of minor, moderate and major stroke is based on the clinical picture, which means the NIH Stroke Scale Score. However, the problem is with the NIH Stroke Scale Score, you can have a moderate stroke, which means NIH of 8 to, let's say, 16. However, the lesion can be very small because the lesion is in the internal capsula, whereas patients with rather large strokes, especially the right hemisphere, can have a rather low NIH Stroke Scale Score. So that is not really reliable on the size of infarction on imaging. And from this, I'm looking at the title of your work, Early vs. Late Antibiotic Coagulation in Minor, Moderate and Major Ischemic Stroke with Atrial Fibrillation, and you are doing a post-hoc analysis of the ELAN randomized clinical trial. Now, take us into your inspiration for that based off of what you just shared with us. Well, as a physician and clinician, I've always been asked by the juniors, when can I start with the anti-calculation? And because everybody is scared to start early, but everybody would like to start early. And to be honest, I really also felt uncomfortable to always take this decision, because at the end of the day, it is also a legal responsibility you have for your patients. And therefore, after a lot of dilemmas, I decided why not doing a study on that topic? And that was the motivation of doing the ELAN trial. And the ELAND trial, which has been published in the New England Journal of Medicine, was estimating the event rates in people treated early versus late with anticoagulants in people with isochemic stroke and atrial fibrillation. So, this is quite an issue, because I will be honest, I'm scared. Every time this question comes up in the hospital, I am deathly afraid you have not only a moral responsibility, but a legal responsibility, as you pointed out, to do the best for your patients and to make sure you see them through. So, I want to start early. I want to make sure that we have secondary stroke prevention so that our patients do not stroke out again, but we don't want the complications that may come with friable tissue encountering anticoagulation. To be honest, you're not the only one who is scared. I am and I was scared all the time. And that was basically the reason why we launched this trial. We said, come on, we need more evidence for this question. And that was the reason why we basically launched the ELAN trial comparing early versus late anticoagulation people with acute ischemic stroke and atrial fibrillation. Can you dive deeper into the ELAN trial? Tell us a little bit about that. Yes, the ELAN trial was a trial which was neither a superiority trial nor a non-inferiority trial. So we made example size calculation. We realized if you really want to do a proper superiority trial, you need thousands of patients to be randomized. And even as a non-inferiority trial, you would need a lot of patients to be randomized. And I am an academic, and I have to do a trial which is feasible and affordable. So therefore we said simply let's randomize patients and look at the event rates in the early and in the late treatment arms and provide physicians with so-called estimates, percentages where afterwards you as a physician can balance the risks and the benefits. And that's basically what we have done in the ELAN trial, which has been published in the New England Journal of Medicine. Dr. Fischer, can you tell us about the design of the ELAN trial, the methods? So ELAN was an investigator-initiated clinical trial, and it was performed in 103 sites in 15 countries. And people were randomly assigned in a one-to-one ratio in an early versus a late treatment arm. But it's important to note that the classification into minor, moderate and major stroke was based on imaging. And in people with minor stroke, early meant starting within 48 hours versus late was day three or four. In people with moderate stroke, early was also within 48 hours, but late was day six and seven. In people with major strokes, we waited until day six, and then we randomized the people, and if they were in the early treatment arm, we started immediately, whereas they were in the late treatment arm, they started at day 12 to 14. The primary outcome was measured at 30 days and was a composite of reconditioned chemic stroke, systemic embolism, major extra or intracranial bleeding, and vascular death at 30 days. And what were your conclusions on that trial? The conclusion was mainly that starting early was not increasing the risk of symptomatic intracranial hemorrhage, and numerically there were lower ischemic events in the early treatment arm, which was what we expected. But we cannot say that an early start is superior to a later treatment start because we were not testing a statistical hypothesis. Now, taking us into this post-hoc analysis, currently, how did you set up this current study into figuring out what would be best in terms of treating early versus late? So, for us as physicians, one of the major questions remained was, is that risk-benefit ratio also the same in people with major stroke? So, nobody is scared to start anti-calculation early in people with minor stroke. Also, with moderate stroke, we more or less feel comfortable, but we are all scared about the major strokes. And therefore, we said, let's look at all these three subgroups. So, looking, what are the event rates in early versus late anti-calculation people with minor, moderate and major strokes separately? Now, looking at the current setup of this study in the Pulse Tokenalysis, you did a radiographic analysis of the infarct size to determine minor, moderate and major. Can you tell us about that? Yes. So, the ELAN trial, in contrast to the other classification of minor, moderate, major stroke, used an imaging-based classification. So, in the ELAN trial, the investigators had simple cards with infarct size. So, if an infarct was less than 1.5 cm, it was considered as a minor stroke. If it was a huge stroke involving the whole MCA or PCA or ACA territory, it was considered as a major stroke, and everything in between was considered as a moderate stroke. And now, in the ELAND trial, in this post-hoc analysis, we wanted first of all to see whether the investigators evaluated the infarct size correctly in comparison to an imaging core lab, and then we also wanted to assess whether the event rates are different in people with minor, moderate and major stroke. Now, what event rates are we referring to? Because I'm assuming these are the primary outcomes you're looking at. Yes, so overall event rates are rather low, and lower than we would have expected. So for instance, in people with minor stroke, overall outcome events which were a composite of ischemic stroke, systemic embolism, intracranial bleeding, extracranial bleeding and vascular death, was 2.7% in the early and 3% in the late treatment arm. For moderate stroke, it was 2.8% in the early versus 3.6% in the late treatment arm. And in major strokes, it was 3.7% in the early and 7% in the late treatment arm. I mean, we could talk about those results, but one extra question I had was looking at whether or not the investigators took into the size. I think you were trying to make sure the correlation was equal amongst sites. How did you do that? So the correlation, we had an imaging core lab, and they were looking through all the images again, and then we were comparing the infarct size classification between the local sites and the imaging core lab. And what we could see that the correlation was moderate, but it was very good if you were using MRI data only. The inter-rater reliability among core lab ragers, because we had different ragers in the imaging core lab as it is supposed to, they were very strong and confirmed that this classification is useful. Okay, that's pretty good. So back to our results. Statistically speaking, it more or less seems very similar in terms of mild, moderate and severe stroke with early treatment versus late treatment. What do you take from that? Yes, that's true. So we could not find any treatment effect heterogeneity between the three groups. So if you look at the event rates in people with major strokes, where we were really scared that with an early treatment start, we increased the risk of symptomatic intracellular hemorrhage, there was no symptomatic intracranial hemorrhage in people with major stroke in the early treatment arm, and there were two symptomatic intracranial hemorrhages in the late treatment arm. And overall, the event rates in people with major strokes were lower in the early versus the late treatment arm. So for us, that means that in people with major strokes, it is probably safe to start anticoagulation early. Now, you're a clinician scientist and you're out in the field as well. What do you take from this data? What are you going to do tomorrow? And when I say, what are you going to do tomorrow? I mean, what are you going to do in your lab? And what are you going to do at the bedside? Well, that's a very important question. So for me, I feel really reassured after these results, and it takes a lot of anxiety away from me when treating patients. So if I do have a patient with a minor and moderate stroke, I start anti-calculation immediately, which means within 48 hours. If I do have a patient with a major stroke, I know that I can start at day six, and I do not have to wait 12 to 14 days. So this has changed my attitude in clinical practice already now. What can we see from you in future studies? What would you follow up on in order to further elucidate this? Because I think that a good chunk of our listeners that still may be reluctant, it's a risk benefit analysis, and at the bedside, you're speaking with the patient and the family, and there are so many nuances. We are not the only one doing such a trial. There was a very interesting trial called the timing trial, which has been published in Circulation. And they did a non-inferiority trial. Unfortunately, they had to stop early due to the pandemic. But overall, what they could show is they had no bleeding events neither in the early nor in the late treatment arm. They had a different design. They were randomizing within four days and after four days. But nevertheless, no bleeding complications, and they were able to show non-inferiority of an early treatment start. And there is another very important big trial coming later this year. It's the Optimus trial from the UK. So they were randomizing more than 3,500 people, and they were randomizing people in the early treatment arm, which means within four days, and in the late treatment arms, it was day 6 to day 12. And we are extremely keen to see the results of this trial too. If that confirms that an early treatment start is safe, this will most likely also have an impact on the guidelines. And furthermore, we are planning individual participant meta-analysis of all current trials published and presented on that topic. Right. That's amazing. So for many of you who are still reluctant, look out for all this data coming out and really take that into account. If you were to redesign the ELON trial and do a follow up, is there anything that you would change in the setup of the trial? Yeah, thank you very much. That's an excellent question. And yes, if I could redesign the trial, I most likely would also randomize people with major strokes in the early treatment arm within 48 hours. But please keep in mind, at the time when we designed the trial, we were really scared that we are going to harm people, especially with major stroke. So that was the reason why we said in these people, we most likely have to delay until six days, until we can randomize them. But now in retrospect, we probably should do another trial for people with major stroke, comparing early within 48 hours versus day six as late treatment arm. Dr. Fischer, thank you so much for your work, and thank you for speaking with me today. Thanks for having me. I think it's very important to highlight that such an endeavor, such a trial cannot be done by individuals. So I would like to thank all the collaborators, all the participants of the trial, and especially also I would like to thank you for giving me the opportunity to talk about the ELAN trial. This episode was produced by Daniel Musisi at the JAMA Network. To follow this and other JAMA Network podcasts, please visit us online at jamanetworkaudio.com. This is Dr. Cynthia Armand. Thank you for listening.