Stroke, Systemic Embolism Risk Higher When Amiodarone Combined with Warfarin Rather Than Apixaban

Download this article's Factoid (PDF & PPT for Gold Subscribers)  

In patients with atrial fibrillation (A-fib) who require amiodarone therapy, anticoagulation with the factor Xa inhibitor apixaban instead of warfarin carries lower risks of stroke, systemic embolism, and death, according to a substudy of the ARISTOTLE trial published in the October 14, 2014, issue of the Journal of the American College of Cardiology.

Originally published in the New England Journal of Medicine in 2011, the ARISTOTLE trial randomized 18,201 patients with A-fib and at least 1 additional risk factor to either apixaban (5 mg twice daily or 2.5 mg twice daily) or dose-adjusted warfarin (target INR range of 2.0-3.0).
For the new study, investigators led by Greg Flaker, MD, of the University of Missouri (Columbia, MO), conducted an exploratory post hoc analysis of outcomes in patients who did (n = 2,051) or did not (n = 15,856) receive amiodarone, an antiarrythmic drug. Amiodarone status was not known in all study participants.
There were significant geographic variations in amiodarone use, with the highest rate seen in Latin America (17.9%) and the lowest in North America (6.6%). Compared with patients who did not receive amiodarone, those who did were younger, more likely to have heart failure or reduced LVEF, and less likely to have diabetes, prior stroke, TIA, or systemic embolism.

Amiodarone and Warfarin a Bad Combination

At a mean follow-up of 22 months, those taking the combination of warfarin and amiodarone had lower time in therapeutic range than patients on warfarin alone (56.5% vs. 63.0%; P < .0001).

Patients randomized to receive apixaban had lower rates of the primary efficacy endpoints of stroke/systemic embolism and all-cause death, and lower rates of the safety endpoint of major bleeding compared with warfarin regardless of their amiodarone status (table 1).

Table 1. Safety and Efficacy Endpoints: Apixaban vs Warfarin


With Amiodarone

HR (95% CI)

Without Amiodarone

HR (95% CI)

P for Interaction

Stroke/Systemic Embolism

0.68 (0.40-1.15)

0.82 (0.68-1.00)


All-cause Death

0.74 (0.55-0.98)

0.93 (0.83-1.05)


Major Bleeding

0.61 (0.39–0.96)

0.72 (0.62–0.84)



In a propensity-matched analysis, the annual rate of stroke or systemic embolism was higher among patients who were taking amiodarone than those who were not (1.58% vs 1.19%; P = .032). Rates of all-cause death, cardiovascular death, and noncardiovascular death were numerically—but not significantly—higher among amiodarone users.

“The reason for the suggestion of a higher rate of stroke and systemic embolism in patients who received amiodarone versus patients who did not receive amiodarone is unclear,” the study authors write. “There is no a priori reason for class III antiarrhythmic medications to be thrombogenic.” Rather, they suggest, the thromboembolic risk is likely related to characteristics of the A-fib population and not to the specific antiarrhythmic drug.

Limitations Cloud Interpretation

In an editorial accompanying the study, Juan F. Viles-Gonzalez, MD, of the University of Miami Miller School Of Medicine (Miami, FL), and Jonathan L. Halperin, MD, of Mount Sinai School of Medicine (New York, NY), say the findings add to “the uncertainty raised by conflicting reports of mortality (including noncardiovascular death) associated with amiodarone.” However, they say, it is possible that the mortality signal “could dissipate if tested in a randomized format.”

They also point out several limitations of the analysis, including that patients with permanent A-fib are not generally considered candidates for antiarrhythmic therapy, yet most patients in ARISTOTLE, including those treated with amiodarone, were classified as having persistent or permanent A-fib. Additionally, “the duration of amiodarone was not specified and neither was information about its efficacy in maintaining sinus rhythm, which the editorial notes could have implications for clinical outcomes and whether these might differ in patients with recent-onset vs long-standing A-fib,” they write, adding that the analyses also were not adjusted for LVEF or for hepatic or renal disease, which could have contributed to thromboembolism and bleeding.

Difference Makes Biological Sense

In a telephone interview with TCTMD, Michael D. Ezekowitz, MD, PhD, of Thomas Jefferson Medical College (Philadelphia, PA), said the finding of less time in therapeutic range in patients co-administered warfarin and amiodarone is not surprising given the known ability of amiodarone to interfere with the CYP2C9 enzyme system.

“This study highlights that the interaction with warfarin is important in that the beneficial effects of warfarin are somewhat attenuated by the co-administration of amiodarone,” he said. While there is potential for amiodarone to interfere with apixaban, too, Dr. Ezekowitz said the analysis is good news in the sense that no such attenuation was seen in the apixaban patients.

However, the long half-life of amiodarone means its effect can be protracted for a month or longer after stopping it, he added. Therefore, the lack of data on amiodarone duration, as pointed out by Drs. Viles-Gonzalez and Halperin, is an important limitation of the study, Dr. Ezekowitz stressed. Another limitation, he added, is that the study classified patients as being in or out of therapeutic range, but did not specify if they were too low or too high.

“If amiodarone is to be used, what this study shows is that apixaban is less likely to have a negative impact … than warfarin,” he said. “So, if you must prescribe amiodarone, apixaban would be the preferred approach.”

1. Flaker G, Lopes RD, Hylek E, et al. Amiodarone, anticoagulation, and clinical events in patients with atrial fibrillation: insights from the ARISTOTLE trial. J Am Coll Cardiol. 2014;64:1541-1550.

2. Viles-Gonzalez JF, Halperin JL. Efficacy and safety of amiodarone in patients with atrial fibrillation in the era of target-specific anticoagulants [editorial]. J Am Coll Cardiol. 2014;64:1551-1553.


  • The study was sponsored by Bristol-Myers Squibb and Pfizer.
  • Dr. Flaker reports having received grants from Boehringer Ingelheim and Sanofi Aventis and consulting fees from Boehringer Ingelheim, Bristol-Myers Squibb, Daiichi Sankyo, Pfizer, and Sanofi Aventis.
  • Dr. Halperin reports serving as a consultant to Bayer, Boehringer Ingelheim, Daiichi Sankyo, Janssen Pharmaceuticals, Johnson & Johnson, and Pfizer.
  • Dr. Viles-Gonzalez reports no relevant conflicts of interest.
  • Dr. Ezekowitz reports consulting for Bristol-Myers Squibb and Pfizer, as well as multiple pharmaceutical companies that manufacture newer anticoagulants.

Related Stories: