Current Drug Usage in Atrial Fibrillation

Abstract & Commentary

By John P. DiMarco, MD, PhD, Professor of Medicine, Division of Cardiology, University of Virginia, Charlottesville.

Source: Piccini J, et al. Pharmacotherapy in Medicare beneficiaries with atrial fibrillation. Heart Rhythm 2012;9:1403-1408.

This paper reviews drug therapy in patients with atrial fibrillation in the Medicare population. The author extracted a 5% sample of Medicare beneficiaries who were enrolled in Medicare Part D. They identified patients with prevalent atrial fibrillation as of January 1, 2006, and as of January 1, 2007. Prevalent atrial fibrillation was defined as the presence of an atrial fibrillation diagnosis on an inpatient claim or more than two outpatient claims during the previous calendar year. Part D pharmacy prescriptions that were filled during the first 4 months of each of the 2 years were then analyzed. Comorbid conditions were determined from the Medicare claims files. The rates of drug usage were then adjusted for age and gender so that the values could be compared to other databases.

There were 27,174 Medicare patients with prevalent atrial fibrillation enrolled in Medicare Part D in 2006. Enrollment nearly doubled to 45,711 in 2007. Among patients with prevalent atrial fibrillation in the 2007 cohort, 74% were on rate control medications, with beta-blockers (52.8%) being the agents more commonly used. Digoxin was used by 30% and calcium channel blockers (dltiazem or verapamil) by 19%.

Membrane active antiarrhythmic agents were used in 19.1% of the 2007 cohort. Class Ia agents were used in less than 1%, Class Ic agents in 3.9%, and Class III agents in 14.9%. Of the Class III agents, amiodarone was prescribed more frequently (9.4%) than dofetilide (0.5%) and sotalol (5.1%). Warfarin anticoagulation was used in 59% of the patients and antiplatelet agents in 9.1%. There were minor differences in drug therapy depending on the presence of heart failure. Class Ia drug therapy was less common and amiodarone use was more common if heart failure was present. Oral anticoagulation was comparable between groups with and without heart failure. Interestingly, anticoagulation therapy decreased with increasing CHADS2 scores. Warfarin was used in 62% of the patients with a CHADS2 score of 0 and 1, but only 55%, 52%, and 49% of patients with CHADS2 scores of 4, 5, and 6.

The authors conclude that medication use in atrial fibrillation varies according to underlying risks and comorbid diseases. In older patients, rate control strategies are more common than rhythm control strategies. There is an inverse relationship between CHADS2 risk score for stroke and the use of oral anticoagulants.


This survey covers Medicare patients treated in 2006 and 2007. Since these data were collected, dronedarone, a non-iodinated compound similar to amiodarone, was released with some enthusiasm. Unfortunately, subsequent data have raised questions about both the safety and efficacy of dronedarone, and it’s likely that the medication profile reported here is still largely accurate. The profile shows how disappointing antiarrhythmic therapy for atrial fibrillation remains. Most Medicare patients are treated with a rate-control strategy and when a rhythm-control strategy is followed, amiodarone, an agent with significant toxicity, is the drug most commonly used.

There have been no recent major advances in drug therapy for atrial fibrillation. It remains to be shown that catheter ablation can become effective enough long-term and available enough to the millions of patients with atrial fibrillation to change these depressing statistics.