Cardioversion for Atrial Fibrillation
Abstracts & Commentary
Synopsis: Adequate oral anticoagulation reduces the risk of embolization postcardioversion of atrial fibrillation.
Sources: Gentile F. Mayo Clin Proc. 2002:897-904; Gallagher MM, et al. J Am Coll Cardiol. 2002;40:926-933.
Recent studies of atrial fibrillation patients have shown that a strategy of heart rate control and chronic anticoagulation is equivalent to a cardioversion strategy with efforts to maintain sinus rhythm with regards to longevity. Thus, there is renewed interest in the safety of cardioversion. Gentile and colleagues report on the Mayo Clinic Rochester experience with elective cardioversion for atrial fibrillation in 834 successful cardioversions in 717 patients. Chronic atrial fibrillation (> 48 hours) occurred in 92% of the patients. No embolism occurred beyond 30 days from cardioversion , confirming the theory that embolism is associated with cardioversion. Transthoracic echo was performed before cardioversion in 731 instances and transesophageal echo in 213. Anticoagulants were administered in 90% of the procedures and after hospital discharge in 81%. Embolic events were infrequent (0.9%) and all were cerebral. In patients on warfarin with a therapeutic INR (> 2.0) the rate was zero. There were deaths observed, but only one was due to cerebral infarction. Of the 7 embolic events 5 were on anticoagulants, but only one had a therapeutic INR level as evidenced by an adequate activated partial thromboplastic time on heparin. None of the embolic events occurred in the patients with acute atrial fibrillation, despite the fact that only half got anticoagulants. If the acute patients are excluded from the analysis the embolic rate is 4% with no anticoagulation, 1.7% in those with inadequate oral anticoagulation, and 3% in those on heparin alone. Hypertension and diabetes were also associated with embolization. Gentile et al concluded that adequate oral anticoagulation reduces the risk of embolization post cardioversion of atrial fibrillation.
Gallagher and colleagues reported on the results of 2639 cardioversions in 1950 patients with atrial arrhythmias in13 European hospitals. A 3-week follow-up was completed for > 97% and 4 weeks for > 90%. Oral anticoagulants were administered for at least 3 weeks in rhythm disturbances of > 7 days duration; was not given if < 24 hours; and was variable between 1-7 days. INR targets varied, but > 2.5 was considered adequate. Anticoagulation was more vigorous in atrial fibrillation (AF) than in other rhythms. Pre- and postanticoagulation was accomplished in 92% with AF vs 57% of other rhythms. Only 22% of atrial flutter for > 2 days had an INR > 2.5 at cardioversion vs 38% of atrial fibrillation cases. All 14 embolic events occurred after successful cardioversion (9 AF, 5 other). In 9 of the 14 patients with emboli, long-term anticoagulation was being given, but all had INRs < 2.5 or it was not measured (2). INR was related to the incidence of emboli: zero if > 2.5 and 0.9% if < 2.5. Hypertension and structural heart disease occurred in all of the embolus cases, but this association was not statistically significant. The rate of embolus in atrial flutter was 0.9%. Two serious hemorrhages occurred within 1 month of beginning warfarin and 2 embolic deaths. Gallagher et al concluded that the INR should be > 2.5 at the time of cardioversion if AF is > 2 days in duration and conversion of atrial flutter requires the same approach.
Comment by Michael H. Crawford, MD
Despite the results of AFFIRM and RACE (Clinical Cardiology Alert. 2002;21:33-35), which suggested that a rate control/anticoagulation strategy was at least as good as cardioversion/sinus rhythm maintenance for longevity and functional status, cardioversion is still preferred in certain patients. Thus, the results of these 2 large retrospective observational studies are of some interest. The Mayo Clinic Rochester study is a large single center experience limited to atrial fibrillation patients that provides considerable clinical detail. The multicentered European study is larger and includes other atrial arrhythmias, but provides less clinical details. Both reach the same conclusion that cardioversion is extremely safe with regard to embolic events if adequate oral anticoagulation has been achieved. One study showed no emboli if the INR was > 2.0, the other had no events if it was > 2.5 and both studies required this level of anticoagulation for at least 3 weeks prior and 4 weeks after cardioversion. So which level of INR should we choose? Unfortunately, the precardioversion INR was all that was available. Most centers require an INR > 2.0 weekly for 3-4 weeks before attempting cardioversion. If this policy is rigidly followed many believe cardioversion is safe, but if there is doubt, shooting for 2.5 makes sense as long as levels > 3.0 are avoided, because they increase the risk of major hemorrhage.
These papers make several other points. First, they confirm that atrial arrhythmias that have been present for < 48 hours can be safely converted without anticoagulation. The trick is determining the time of onset of the arrhythmia. If this is in doubt, the standard approach should be taken. Second, atrial flutter requires the standard anticoagulation approach if the goal of zero emboli is adopted. The role of heparin is unclear. Heparin was frequently used in both studies, but there is doubt whether it is an adequate substitute for warfarin. In the patient with atrial arrhythmias for < 48 hours it was often used but its value could not be established in these retrospective studies. In addition, the role of TEE is not clear from these studies. Only the Mayo study includes echo data, but when viewed in the context of anticoagulation adequacy, it did not appear to contribute to decision making. However, a previous multicentered study of TEE- guided vs the standard approach to cardioversion showed no difference in embolic events (0.8% vs 0.5%), but since the adequacy of anticoagulation was not studied, the role of TEE is now in doubt. Given the choice of a 1% chance of embolus with TEE vs zero with adequate anticoagulation for 3 weeks, which would you choose? The TEE approach makes the most sense for younger patients without overt heart disease who have unknown or relatively short duration atrial arrhythmias (2-7 days). The Mayo study did show an association between hypertension and diabetes and risk of stroke with cardioversion. It may be that such patients have endothelial dysfunction or pre-existing cerebral vascular disease that makes them more susceptible to having clinically apparent cerebral embolic events. These clinically higher risk patients perhaps should not undergo the TEE approach.
Dr. Crawford is Professor of Medicine, Mayo Medical School; Consultant in Cardiovascular Diseases, and Director of Research, Mayo Clinic, Scottsdale, AZ.
Recent studies of atrial fibrillation patients have shown that a strategy of heart rate control and chronic anticoagulation is equivalent to a cardioversion strategy with efforts to maintain sinus rhythm with regards to longevity.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.