false
Catalog
EP on EP Episode 52 - Device Detected AF & Stroke ...
EP on EP Episode 52
EP on EP Episode 52
Back to course
[Please upgrade your browser to play this video content]
Video Transcription
Hi, this is Eric Berstowski and welcome to another episode of EP on EP. What an absolute delight it is for me to have my guest today, Dr. Daniel Singer, who's a professor of medicine at Harvard University and really the person who I've been reading your literature for years on the topic we're going to discuss, which is what to do with device detected atrial fibrillation. So Daniel, welcome to the show. Let me ask you the first question. Can you frame the problem for us? The issue of finding AFib, let's start actually with implantable devices and we'll move to maybe the newer ones. Tell us about what we actually know. So let me just frame the issue. You know, from the original anticoagulation trials, there was evidence that people with paroxysmal AF benefited from anticoagulation and the guidelines say if you have paroxysmal AF, you should get anticoagulated. But we've learned over time that people with paroxysmal AF have less stroke risk than those with persistent AF. And now with devices that can pick up subclinical AF, we find that there's an extreme level of burden that is quite low burden, much less than was in the original trials. And so the question there is do these people face increased stroke risk and will they benefit from anticoagulation? And there are several studies and emerging more studies that address this issue. One of the interesting findings from the ASSERT-1 trial, the ASSERT-2 trial, is that if you implant the device, a monitoring device in people who are say 75 or so, 30% of those people over 18 months will show some evidence of AF. Some evidence meaning greater than six minutes. There are fewer and fewer people who have 24 hours. From ASSERT-1, we now know that the stroke risk faced by people with subclinical AF is a function of how much AF you have. And they didn't see statistically significant increased risk until you got to an episode of 24 hours or more. The other studies include the TREND study. And the TREND study showed that if you had more than five and a half hours in any given day of atrial fibrillation, you faced a doubling of your stroke risk. Now the actual stroke risk was not that high. It was in the 1 to 2% per year range, lower than if you had persistent AF. There are additional studies that are coming online. So for example, Rod Passman and Paul Ziegler from Medtronic recently published a study where they linked the enormous Optum insurance databases with CareLink. And they found that your stroke risk was a function of your CHA2DS2-VASc score and your duration of the longest episode that you have in any given day on the Medtronic devices. And the people with the highest stroke risk were at the highest CHA2DS2-VASc score and the 24-hour. So that's what I wanted to get at a little bit as a clinician who sees these patients all the time. You've done a great job reviewing what we know, and it's confusing to a clinician like myself. Because one study says six minutes and no risk, another one says no, 24 hours. And neither of those actually talk to the burden, which you just brought up as important. So this is confusing for the clinician. Based on what you now know, just reviewing all the literature, what advice would you give to us? For example, is there a burden more than just one episode? Does it matter if you had X number of hours over time, or is it more important to have a 24-hour episode? Most of the studies address the longest episode. They don't integrate it over time and say if you have 30 days of AF, it gives you a higher risk than if you have one long episode on one day. The TREND study actually showed that about 11 hours over the course of a month increased your risk. I didn't mean to interrupt you, but that's kind of the, my gestalt, right or wrong, is that each episode of AFib people have does something to the endocardium. I mean, maybe it won't cause a clot necessarily, but over time, are there changes taking place that make that atrium more thrombogenic? Right. I don't know that, that TREND's data did address that a little bit in their secondary analysis of the 30-day burden. There's the KP Rhythm study, which is looking at symptomatic individuals who are wearing a ZEO patch, and there it was percentage of time over the two weeks of wearing the ZEO patch, not the longest episode, but the percentage of time. So you know, AF burden, amount of AF, it can be measured in so many different ways. I would say, though, at this point, what we have good evidence of is that short episodes, and what's a short episode? Less than an hour. Short episodes, in and of themselves, don't pose much stroke risk. But as you get longer, and certainly at 24 hours, you're facing an increased stroke risk. Now, for the individual that you're, the individual patient that you're facing, let's say they have, you know, an hour or a little bit less than an hour. You can reassure them about that episode, but what about it's going to be down the line, you know? That's the worry. And that's the worry. And so the question is, do you just watch very carefully? And then, you know, if you collect more information that the episodes are longer, then you act. If you don't, you know, you do have, you have to worry about the harms of anticoagulation. We have benefited from the novel anticoagulants. There's much less intracranial hemorrhage risk, so the really devastating outcomes have been reduced. But it's not trivial to be on anticoagulants. Let me ask you this, though. I just had a patient I saw in the office this week, and she's a Chad's vascular, I think it was either four or five, and she had about a two-and-a-half-hour episode. She actually has an ICD. So I called her and brought her into the office, and I went through the usual discussion of potential risk. But what I couldn't tell her at this point, I want to get your feeling for what you think we, I know there aren't the data there, but you have such a knowledge in this field, what's an appropriate thing to tell her? I will tell you what I told her was, you could be, I think you're at risk, and I can't actually tell you that studies have been done in folks like yourself, but here are the other studies. And I just talked it out with her, but what, is that reasonable? What should we be doing? Sure. I think it's reasonable. I mean, in the absence of randomized trials, there are, by the way, at least three randomized trials that are ongoing that we should hear from. But in the absence of randomized trials, we're just trying to take observational data, talk about stroke risk, and infer whether anticoagulation is going to benefit. And there are lots of people, and you know, you look and you make a sense. Is this person falling? Do they have other risk factors for major bleeding? If not, and you talk to them, and they really, really don't want to have a stroke, and most people are like that, but some have fears of bleeding that counteracts that. You know, this is shared decision making. In the absence of hard data, two and a half hours is getting long. And then, obviously, you want to watch carefully. So with all these studies that you've been doing, and we're going to get into the wearables in a second. Yeah. Do I use six minutes? Do I use 15 hours? I know, you're asking me the crucial question. This is the crucial question, and as we discussed before, you know, we're going to have tons of people who are going to have six minutes of AF on their watch or on another wearable, and they're going to come to their primary care doctor, and that primary care doctor is going to send them to you, and you're going to have to be Solomon, you know. So I think six minutes in and of, the problem with the cutoffs of greater than six minutes that were originally touted as raising stroke risk is that greater than six minutes included a lot of greater than 24 hours. So it's a mixed bag. You want to see mutually exclusive categories of AF burden, and I think, you know, trends, for example, show that if it's less, in any given day, if it's less than five and a half hours, it's not, your risk isn't any higher. Now there's small numbers of events. These are not powerful, powerful studies. I think very brief episodes, and six minutes would be very brief. I'm saying arbitrarily the one hour cutoff. I think less than one hour, and there isn't anything more down the line. I usually walk. I'm actually a 20 minute guy, and I don't even ask me why, because I did that, I sort of got a tweener for no reason, and it really was just my own observations. I've been telling people for years and years and years, I found that most people who tend to have longer episodes had something at least 20 minutes at one point. Yes. It was just a guess. Right. So a small amount of time we have left. Let me ask you this. The wearables. Yes. They're driving us all crazy. Right. I mean, you know, whether it be a watch or something you put on your phone or whatever it is, any advice to the community out there on what we're supposed to be doing with these people? Well, the first thing is to know whether it's AF or not. Right. So they're not perfectly accurate. That's for sure. Okay. The second is, it's exactly the same question, which is how much of that AF poses, puts you at high risk? And you do have to put together their background risk, you know, their CHADS VASc score, if you will, plus, but if you get somebody who's CHADS VASc two or three, and they have a couple of hours that you then validate somehow with a, with a, you know, a patch or something. Right. I was going to ask you that. My current approach is if they pick it up and I see it and it's really AFib, I then follow up with several weeks of monitoring because just all the things you brought up, right? That's reasonable, right? To see, are these a 40 second episode or are these two hour episodes? Right. And also, well that, if you have a, if you have it on a watch, you'll know how long the episode is. But, you know, if you have an individual patient where you have validated that what they're picking up on their watch actually is what you pick up on a good electrocardiogram, then you can follow that. But for the average person in the street, you don't necessarily know that what they're, what they're picking up on their, on their watch or other device actually is accurately AF. So you want to, you want to be, you want to validate it in some way, in some prolonged monitoring with a high quality electrocardiographic. So this is really become a problem, right? And it's not much data yet. Yeah. But they, you know, the companies would tell you it's a benefit and I think there is a benefit, you know? I agree with them. I think knowing you have AFib, you can at least have the discussion, but, so somewhere in the future when we have actually some data published on this, we're going to get you back here for part two, but Daniel, thank you so much. Okay. My pleasure. Thank you. Great information for the community.
Video Summary
In this episode of EP on EP, Eric Berstowski interviews Dr. Daniel Singer, a professor of medicine at Harvard University, about device-detected atrial fibrillation (AFib). They discuss the evidence regarding stroke risk in patients with subclinical AFib and the role of anticoagulation. They mention studies such as ASSERT-1, ASSERT-2, the TREND study, and the KP Rhythm study, which have provided insights into AFib burden and stroke risk. They also touch on the use of wearables for detecting AFib and the importance of validating such detections with prolonged monitoring. The interview ends with the acknowledgement that more data is needed in this area.
Keywords
device-detected atrial fibrillation
stroke risk
subclinical AFib
anticoagulation
wearables
Heart Rhythm Society
1325 G Street NW, Suite 500
Washington, DC 20005
P: 202-464-3400 F: 202-464-3401
E: questions@heartrhythm365.org
© Heart Rhythm Society
Privacy Policy
|
Cookie Declaration
|
Linking Policy
|
Patient Education Disclaimer
|
State Nonprofit Disclosures
|
FAQ
×
Please select your language
1
English