Cardioversion for New-Onset AF: Time to Hit the Pause Button?

John Mandrola


August 13, 2014

It's a funny name, isn't it? Cardioversion. A more honest way of describing the procedure would be to say that we are going to deliver a high-voltage shock to your chest. This will stop the heart's rhythm and a normal one will restart in its place. It's extremely painful so we use a general anesthetic.

The problem with using euphemisms in describing medical matters is the same as in writing—obscuring the truth is hardly desirable. The point about cardioversion is that shocking a person from atrial fibrillation to regular rhythm is no small thing.

All doctors have unforgettable cases.

He was a middle-aged professional back in the 1990s. After a stressful few days, he noted a pounding sensation in his chest. Not an alarmist, he waited for it to subside, but when it persisted the next day, he came to the emergency room. The onset of atrial fibrillation was less than 48 hours, so it was "safe" to shock. He was a busy man. He needed to get back to work. And it worked; his heart restarted in sinus rhythm.
The terrible call from the ER came the next evening. He had stopped speaking and moving one side of his body. And then the dreaded words from radiology: "midline shift." He died at the age I am now. It was a long time ago, but it's right there in my memory.

So now let's talk about the euphemism that is cardioversion, specifically, the belief that it is "safe" to shock patients with short-duration AF who are not taking anticoagulant drugs. The restart of the static atria is the problem. Is there a clot?

A research letter published this week in the Journal of the American Medical Association (and covered here by Steve Stiles) provides compelling new data on this common scenario.[1]

First some background: We can agree that the disease atrial fibrillation is serious. The afflicted often feel awful. AF causes fear, for both patient and caregiver alike. And there is an increased risk of stroke, heart failure, or death in the future. So it is right and just that caregivers deliver care. The challenge is how best to treat a disease that is serious but not immediately life-threatening. Can the treatment be worse than the disease?

The study

Researchers from Finland studied 2481 patients who underwent 5116 successful cardioversions from 2003 to 2010. These patients had AF of less than 48 hours and were not on anticoagulation. The research team then retrospectively reviewed their medical records looking for postcardioversion thromboembolic events. To determine the importance of timing of cardioversion from the onset of symptoms, the researchers compared three groups of patients: those cardioverted within 12 hours of onset (n=2440), between 12 and 24 hours (n=1840), and between 24 and 48 hours (n=836).

They reported 38 thromboembolic events in 38 patients (0.7% overall incidence), with eight events (0.3%) in the <12-hour group, 21 events (1.1%) in the 12–24-hour group, and nine events (1.1%) in the 24–48-hour group. A multivariate regression analysis revealed that cardioversion after 12 hours of symptom onset increased the risk of embolic complications more than threefold. The small number of events led to wide confidence intervals, but the p values reached statistical significance.

This report builds on previous work from this group. In 2013, they published portions of this data in the Journal of the American College of Cardiology.[2] In that analysis, age, female gender, and the presence of heart failure or diabetes increased the risk of embolic events after cardioversion of short-duration AF without anticoagulation.


Taken together, these two studies tell us a lot about a frequently encountered problem, one in which the temptation to cardiovert (shock) without anticoagulation is strong. Patients feel bad; they want to get back to normal. And there is that legend about the safety of the 48-hour window.

Yes, of course, these are nonrandomized observational studies. There are caveats. We don't know how many strokes would have occurred in patients without cardioversion. As the authors point out, using patients' symptom recall makes it difficult to time AF onset. (As it is in the real world.) We also don't know if the medical-record review recorded all events. Finally, it's only 38 events in one Nordic country. So, no, this isn't level-of-evidence-A stuff.

Yet, surely there is enough of a signal to warrant a new default, one of caution before action, one of restraint, one of skepticism concerning the safety of that 48-hour window. The key issue is safety. The definition of which reminds me of our guiding rule: Do no harm.

Does a 1.1% risk of an embolic event fulfill that definition?

I would argue, no. No way. Not in this cohort of patients.

What I don't understand about AF treatment: "Why the rush?" What's the hurry to shock or to do the transesophageal echocardiography (TEE) and then shock? We can provide plenty of care without shocking or sticking big black tubes down people's throats. We can deliver anticoagulant drugs; we can control the ventricular rate; we can educate and reassure. We can assist in providing patients a respite from or treatment of the inflammatory milieu that may have caused the AF (caregiving for a parent, stressful work or home situation, or another illness). Remember, AF rarely occurs without a reason.

Then, if AF persists for three to four weeks, we can refer for cardioversion after a period of adequate anticoagulation. It is okay that AF is an inconvenience. Healing disease is supposed to require effort from both patient and caregiver. The time before cardioversion can be used to address important lifestyle factors, such as sleep disorders, alcohol use, blood-pressure control, and stress reduction. Also, many of these (now anticoagulated) patients will spontaneously convert, thereby saving themselves from the euphemism of cardioversion.

This, I believe, is a teachable moment for the electrophysiology community. It is a time when we can educate our colleagues that one of the major risks of the disease AF is the treatment of AF. In the absence of hemodynamic compromise, to shock a heart without anticoagulation, especially after 12 hours or when patients have risk factors for stroke, stretches the definition of safety.

Go slowly, I say. Resist the urge.



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.