Jeffrey Tabas, MD, on Cardioversion for Patients With Atrial Fibrillation

In this podcast, Jeffrey Tabas, MD, discusses the pros and cons of cardioversion in the emergency department, which he also talked about during his session at the American College of Emergency Physicians 2021. 

For more parts of this series, visit our resource center for the American College of Emergency Physicians 2021 Scientific Assembly.

Additional Resource:

Jeffrey Tabas, MD, is a professor of emergency medicine at the University of California, San Francisco’s School of Medicine, an emergency medicine physician at Zuckerberg’s San Francisco General Emergency Department, the director of faculty development for the Department of Emergency Medicine, and the director of Outcomes and Innovations for the UCSF Office of Continuing Medical Education.



Amanda Balbi: Hello and welcome to a special series of Podcasts360. I’m your moderator, Amanda Balbi. In this 6-part series, we will be speaking with Dr Jeffrey Tabas, who is a professor of emergency medicine at the University of California, San Francisco’s School of Medicine, an emergency medicine physician at Zuckerberg’s San Francisco General Emergency Department, the director of faculty development for the Department of Emergency Medicine, and the director of Outcomes and Innovations for the UCSF Office of Continuing Medical Education.

He recently presented a session on atrial fibrillation at the American College of Emergency Physicians 2021 Scientific Assembly. In part 4 of this podcast series, he talks about cardioversion for patients presenting with atrial fibrillation in the emergency department.

Let’s listen in as he answers our questions.

The third challenge you talked about during your session was about cardioversion in the emergency department with immediate-onset atrial fibrillation. What can you tell us about that?

Jeffrey Tabas: That's typically classified as symptoms that developed within 48 hours. So, if they can tell you, “I have felt like this for less than 48 hours,” there's a strong push to cardiovert those patients. What I do is I explore the benefits and the drawbacks of doing so.

So, again, I give a young case. A 28-year-old athlete comes in, who had 4 hours of palpitations, says my heart started beating fast and irregularly after heavy exercise and having a few beers. Their EKG is in atrial fibrillation. A lot of people would cardiovert that patient back into normal sinus rhythm. 

The benefits of cardioverting that patient is that you can avoid blood thinners, because the risk of atrial fibrillation causing strokes really increases after the first 12 hours but markedly after the first 48 hours. After 48 hours, we don't cardiovert people without giving them blood thinners first, unless it's an emergency.

Tthe other benefit is you get them back into sinus rhythm, so you don't need any of your treatments for atrial fibrillation. You don't need to keep them on chronic rate control. Then, they've shown that you can discharge them out of the emergency department, that you use less resources.

Some of the questions are, “Is it safe? Do we convert these patients and then cause them to stroke?” And there's pretty good evidence that it's safe to do this from the emergency department if their symptoms are less than 48 hours. That's assuming they know when they developed atrial fibrillation. A lot of people come in, and they are unaware that they're in atrial fibrillation. That would not be the patient cardiovert in the emergency department.

There are several studies—they’re observational—of patients that they cardioverted. There's a big study out of Maine that looked at 887 patients that was published this year, and only one had a stroke. There was a big study of 1100 patients in Canada that was published in 2017, and only one of those patients had a stroke. There was a study that was published out of Europe in 2019 that looked at 218 patients. and only one had a stroke. So, it looks like the rate of stroke is probably somewhere between 1 in 500 or 1 in 1000 for doing this, which many people would consider safe. It's not high risk.

Many places either get an ECHO to look at the heart and make sure there's not a big dilated atrium or evidence of a clot there. There's something else that you have to be careful of recommending because it's not studied robustly. But something that I consider is to check a D-dimer. D-dimer shows if there's a clot in the blood system. We use it to assess for pulmonary embolism. There's pretty good evidence that it can assess for aortic dissection, and I think that there's some evidence that it can assess for a left atrial thrombosis, which is what puts you at risk to throw a stroke when you’re cardioverting someone.

So, there have been several studies—not large—that have looked at D-dimer and if it's negative when they look with an ECHO, do those patients have clot in their atrium. There have been maybe several hundred patients that they looked at, and they found that it's fairly sensitive; 95% sensitive, which is about the same sensitivity that D-dimer is for pulmonary embolism or aortic dissection. And the range is 90% to 98% sensitive, so you have to take the lower end as what you're willing to accept; so, roughly 90% sensitive for left atrial thrombosis.

In a patient who is low risk to start because you know they're in less than 48 hours, I think that really puts you in a safe category. The studies that have looked at the safety haven't done that. This is just another tool that you could use if you wanted to lower your risk even more. So that's one thought.

I think it's also important to look at whether converting those patients actually saves them from anticoagulation in the long run. And it's interesting because we have just talked about how current guidelines say, if you have risk factors—meaning hypertension or you're over 65 years, anything other than lone Afib—you should be anticoagulated, even if it's just occasional Afib.

So, your patient who comes in with Afib once, you convert out, it's not clear that you're saving them anticoagulation, because they should be anticoagulated probably long term if they have definitely 2 risk factors, because this is not the only time. At least you should monitor them for ongoing Afib, like for 30 days or with your smartwatch, to see if they go back into atrial fibrillation because they should be anticoagulated.

The thought used to be that you could avoid 3 weeks of anticoagulation before cardioversion and 4 weeks afterward for recent on set, but it's not so clear. Actually, the European guidelines recommend giving everyone a shot of anticoagulation before you cardiovert them, even if they're low risk.

And then, if they have a CHA2DS2-VASc score of greater than 1, so they’re anything other than lone Afib, they recommend cardioverting them for 4 weeks and then reassessing whether they need it for life. So, it's not clear that you avoid that much anticoagulation by cardioverting them in the ED, but how about the fact that you get them back into sinus rhythm?

Well, I actually think that that's not so clear either. You do get them back into sinus rhythm, you definitely do. Actually, the Canadian group has published their results of 1100 patients who had recent-onset Afib, and they were successful in converting at least 80% of their patients. Other studies have shown 90% to 95% get converted into sinus rhythm.

But when you look—people have not done this study—at just not converting them, 60% convert back into sinus rhythm. So now, we're talking about trying to convert 100 people who come in with recent-onset atrial fibrillation. You'll be 90% successful if you cardiovert them, but you'll be 60% successful if you do nothing.

My mentor taught me don't just do something, stand there. A lot of the times, we intervene and patients are not better off. So, some of the literature that supports this…I have not found studies that specifically look at this, but a meta-analysis of 18 randomized controlled trials of amiodarone compared with placebo found that placebo had the 60% conversion rate. That's where I came up with that roughly 60%. There's some other literature that supports it. 

There was a recent study in the New England Journal of Medicine—great study— from 2019 that looked at immediate vs delayed cardioversion for these acute-onset patients. They looked at almost 500 patients in the ED. They found that they converted 78% vs 28% who had delayed cardioversion. Basically, they converted their patients acutely, and the patients who had delayed cardioversion—if you look at them at 4 weeks, it was the same rate they were in sinus rhythm.

So, delaying the cardioversion doesn't hurt. If you look at patients in 1 year who have been cardioverted in the ED vs not, who came in with acute-onset—because these are the patients who flip back out of atrial fibrillation—looks like the same rate might be in atrial fibrillation at 1 year, not that different. So, it's not clear. You're definitely getting way more patients back into sinus rhythm, so you feel good; the patient came in, they needed your help, you gave them your help, and they're now back to normal. It's great, but it's not clear that it really changes things at 4 weeks or definitely at 1 year when maybe 30% are back in chronic Afib whether you converted them in the emergency department or not.

So, should you do it? Well, ultimately there's no right answer. One, it depends how symptomatic they are. For someone who's really symptomatic, you might err toward converting them out, and you might give them more chance to see their doctor without feeling terrible, follow-up, decide whether they want to have ablation therapy to get them back into sinus rhythm, or things like that.

But it's not clear that it keeps them in sinus rhythm better. It's not clear that it avoids anticoagulation, because we know that just intermittent atrial fibrillation still has a high rate of stroke, and if they have a non-low CHA2DS2-VASc score, they probably should be anticoagulated. You should probably use shared decision-making and talk about the benefits of cardioversion vs just controlling their rate and having them follow-up. 

Again, they will love you if you're their doctor and you convert them into sinus rhythm, and they feel great, but it's not clear how much benefit that's having. That's my take-home point for cardioversion in the ED for acute-onset or recent-onset atrial fibrillation.

Amanda Balbi: Thank you so much for speaking with me today. And for our listeners, stay tuned for the subsequent parts of this series.

Submit Feedback