The treatment of atrial fibrillation has always been a problem - the antiarrhythmic drugs used to suppress the arrhythmia work only about 50% of the time, and besides that tend to cause a lot of side effects. Recent studies suggest that aggressive measures to restore and maintain a normal rhythm in patients with atrial fibrillation may not be any more effective than just controlling the heart rate and using anticoagulation with coumadin (to reduce the risk of the most dreaded complication from atrial fibrillation, i.e., stroke.) Click here for a recent review of the treatment of atrial fibrillation.
Now a study from Germany, published this week in the Journal of the American College of Cardiology reveals that a much higher proportion of patients with treated atrial fibrillation have recurrent, prolonged episodes than previously thought. In this study, 101 patients with previous episodes of atrial fibrillation, and who also happened to have an indication for receiving a permanent pacemaker, were implanted with pacemakers that were also capable of automatically recording ECGs if arrhythmias occurred. These patients had their antiarrhythmic therapy optimized to suppress atrial fibrillation, and were followed for an average of 19 months. During this time, 50 patients had episodes of atrial fibrillation lasting longer than 48 hours - and of these, a full 38% had no symptoms, and were in a normal rhythm at the time of their follow-up visits, and thus normally would not have been diagnosed as having recurrent atrial fibrillation.
The investigators conclude that the rate of successful suppression of atrial fibrillation is "grossly overestimated."
What this means
Clearly, the incidence of recurrent, prolonged atrial fibrillation in patients on treatment to suppress this arrhythmia is much higher than previously appreciated. Until now, if patients are in a normal rhythm at the time of a doctors visit, and do not complain of palpitations during the interval between visits, they have been assumed to have had successful suppression of atrial fibrillation. Obviously, this assumption is wrong. (The means of detecting these rare but significant episodes of atrial fibrillation has not previously existed - it was only the recent development of chronically implantable devices that record the heart rhythm over extremely long periods of time that made this observation possible.)
The implication of this observation is important. In essence, it means that anticoagulation therapy should not be discontinued in patients judged to have "successful" suppression of atrial fibrillation. Once a patient is placed on coumadin for atrial fibrillation, the best evidence available today suggests they should remain on coumadin forever. As odious as it may be to use coumadin, it is less odious than having a stroke. In the AFFIRM trial, more than half the strokes that occurred in the rhythm-control arm of the study were seen in patients who stopped coumadin therapy. With the results reported this week, it may very well be that these patients were having undetected episodes of recurrent atrial fibrillation. Fortunately, new anticoagulants that appear to be much more convenient to use than coumadin are right around the corner.

