Applying the medical literature to the clinical practice of medicine, via education, can often be like a big game of “Telephone.” You know the game, the one in which the first person whispers a phrase to the next person, who, in turn, whispers it to the next, so on and so forth. Invariably, by the time the final person recites the message, the meaning is so altered that the original intent is lost. As a game, it’s pretty funny. In medicine, the results are tragic. And yet, this is exactly what can happen when an attending teaches another attending or resident about a concept or article. First, a little bit of bias enters into the interpretation. Then that person relays it to the next resident and students just a little differently, resulting in a viral distribution of misinformation and misunderstandings. Ultimately, we end up with “evidence-based” treatment standards that are nothing more than medical folklore that have drifted way off course from the original articles being cited.
Cardioversion for atrial fibrillation is one of those concepts. Follow the literature. Is there really any research that supports our concerns and angst over cardioversion?
“We have to anticoagulate them!” you say. “If they’ve had symptoms for more than 48 hours, we can’t cardiovert them! They’ll have an embolic stroke!” Oh, and, “The sky is falling!” We need to get a hold of ourselves and get back to the original literature.
It is truly – well, shocking –,to me that electrical cardioversion isn’t used more often in the United States. It is safe, cost-effective, avoids unnecessary admissions, and when successful, restores better function and symptomatic relief to the patient.
It’s time to change the message of our telephone game! Given the option, restoring sinus rhythm is superior to rate control of atrial fibrillation.
The controversy that exists is over safety of cardioversion versus rate control of atrial fibrillation. We have all been convinced that the risks of converting a-fib, in particular embolic stroke, are far too great in most circumstances to offer this option in the ED. The “Big Daddy” was the AFFIRM trial, published in the New England Journal of Medicine in 2002. For most, the controversy ended there. However, acceptance of rate control over rhythm control, using this article as your reason, just doesn’t make sense.
AFFIRM studied 4060 patients over the age of 65 with chronic atrial fibrillation and stroke risk factors. They were randomized to rate control or rhythm control. If a patient failed in one arm, they were allowed to then cross over to the other treatment arm. So, if you failed rhythm control, you could then move over to the rate control arm.
A wide variety of drugs were used in both treatment arms including amiodarone and sotalol, primarily for rhythm control and beta-blockers, digoxin and calcium channel blockers for rate control. A therapeutic INR between 2.0 and 3.0 was utilized in all patients. The mean follow up time was 3.5 years. The investigators reported that more patients crossed over from the rhythm control arm to the rate control arm than vise versa. Interestingly, there was no statistical difference between the groups in the composite end points of death, disabling stroke, anoxic encephalopathy, major bleeding, cardiac arrest or the frequency of central nervous system events or ischemic stroke. Although more complications were noted with rhythm control group, those complications were due to the side effects of the medications utilized. Electrical cardioversion was not an issue. Despite the fact that no treatment advantage was proved for rate control over rhythm control, this study solidified our commitment to the former. Also of note is that primary investigators disclosed financial relationships with multiple pharmaceutical companies.
Although the AFFIRM trial must be viewed with some scrutiny, we do know that patients with chronic atrial fibrillation may be difficult to maintain in sinus rhythm even if they are converted. Although the AFFIRM trial did not prove that electrical cardioversion was harmful, the medications needed to maintain sinus rhythm are proarhythmic, associated with potential harm. Thus, chronic atrial fibrillation, unless unstable, is often not amenable to cardioversion of any kind, leaving rate control as the only option.
So, acute atrial fibrillation is most likely where we can change practice, making a positive impact by doing things differently than we’ve been taught for decades. The primary concern with cardioversion is the issue of embolic phenomena, in particular ischemic stroke. Based on the literature, this is largely a hypothetical concern.
Forty-eight hours has been the cutoff for cardioversion without anticoagulation for decades. But, where is the evidence for this time frame? There isn’t any. This time frame was set arbitrarily, adopted in practice and training, and then passed from one generation of physicians to the next. In 2005, Potier published a best evidence report in the Journal of Emergency Medicine in which he assessed the 54 papers most closely related to this topic. However, none of them addressed the specific issue of anticoagulation before cardioversion. There is no evidence to refute or support the practice of anticoagulation before cardioversion if atrial fibrillation has been present for greater than 48 hours. Furthermore, there is no evidence to suggest that patients require anticoagulation once successfully cardioverted.
Testing the safety of electrical cardioversion, observing the 48-hour window, Burton reported that in 388 patients with atrial fibrillation, 332 were successfully cardioverted electrically. 28 complications occurred in 25 patients. 22 were from procedural sedation, and none were of a thromboembolic nature. 25 of the 39 return visits within 7 days were due to a recurrence of a-fib (Burton, Ann of Emerg Med).
Jacoby, in the Journal of Emergency Medicine, 2005, reported that patients who present with atrial fibrillation within 48 hours of onset could be safely and successfully cardioverted. In 29 of 30 events in 24 patients, cardioversion was successful with no embolic events at 19 weeks. Other studies have shown similar results, validating Jacoby’s findings. Prior to Jacoby, Ian Stiell and Diku Mandavia published a large series in 1999, which included all patients with new onset or paroxysmal atrial fibrillation. 72% had a previous history of atrial fibrillation and 18% had symptoms for greater than 48 hours or the onset was unknown. The treatment groups included rate control, chemical cardioversion and electrical cardioversion. 28% – 80 patients – were entered into the electrocardioversion arm, 83% of which failed chemical cardioversion. 89% were successfully cardioverted. 100% converted were discharged home. 10% returned within one week. However, none of them returned for a cardioversion-related complication. Although some patients were lost to follow up, only one thromboembolic event took place in the 289 patient population. This was in a 78 year old whose a-fib spontaneously converted (Michael, Ann of Emerg Med).
In 2009, a systematic Cochrane database review was performed. Three randomized, controlled trials of high quality were included. They concluded that even for sustained a-fib, electrocardioversion improved the quality of life, while no conclusive evidence was found that the risk of stroke or TIA was increased. (Rhythm Control With Electrocardioversion for Atrial Fibrillation and Flutter. Latha G. Stead, Lekshmi Vaidyanathan. Annals of Emergency Medicine - November 2009).
Finally, in May of this year, Ian Stiell published the largest study to date on the subject, based on the Ottawa Aggressive Protocol. 1057 patients presented with recent onset of atrial fibrillation or flutter, 660 of which were enrolled. The primary reasons for not enrolling the remainder were due to an unknown duration of symptoms or symptoms greater than 48 hours and due to spontaneous conversion. Procainamide was administered first, with a conversion rate of 58.3%. 243 patients subsequently were electrically cardioverted, 91.7% successfully. 96.8% were discharged home from the ED, and 93.3% were discharged in sinus rhythm. No deaths and zero strokes occurred (Stiell, CJEM).
In order to accept new concepts, we have to be willing to let the old ones go, especially dogma that lacks the evidence to support it. The evidence strongly suggests that cardioversion is safe and effective for atrial fibrillation, especially acute or paroxysmal a-fib. Patients feel better in sinus rhythm, their functional capacity improves and no clear and convincing evidence has refuted its safety. Although some investigators have shown this to be unnecessary, some others have abided by the arbitrary 48-hour rule. Thus, the first step for those on the conservative side is to cardiovert patients who present within this window. Some investigators have taken the initial step of attempting chemical cardioversion. Knowing that the conversion rate in all of these studies is better with electrical cardioversion, the most efficient physician should consider skipping the drugs and moving right to the electricity.
Kevin Klauer, DO, is the editor-in-chief of Emergency Physicians Monthly.
Don't miss A Brief History of A-fib and Cardioversion by David Newman, MD