Author + information
- Received March 13, 2014
- Revision received June 17, 2014
- Accepted July 8, 2014
- Published online October 14, 2014.
- Greg Flaker, MD∗∗ (, )
- Renato D. Lopes, MD, PhD†,
- Elaine Hylek, MD, MPH‡,
- Daniel M. Wojdyla, MS†,
- Laine Thomas, PhD†,
- Sana M. Al-Khatib, MD, MHS†,
- Renee M. Sullivan, MD∗,
- Stefan H. Hohnloser, MD§,
- David Garcia, MD‖,
- Michael Hanna, MD¶,
- John Amerena, MBBS#,
- Veli-Pekka Harjola, MD, PhD∗∗,
- Paul Dorian, MD††,
- Alvaro Avezum, MD, PhD‡‡,
- Matyas Keltai, MD, DSc§§,
- Lars Wallentin, MD, PhD‖‖,
- Christopher B. Granger, MD†,
- ARISTOTLE Committees and Investigators
- ∗University of Missouri, Columbia, Missouri
- †Duke Clinical Research Institute, Duke University Medical Center, Durham, North Carolina
- ‡Boston University Medical Center, Boston, Massachusetts
- §J.W. Goethe-University, Frankfurt, Germany
- ‖Division of Hematology, University of Washington, Seattle, Washington
- ¶Bristol-Myers Squibb, Princeton, New Jersey
- #Geelong Cardiology Research Center, Deakin University, Victoria, Australia
- ∗∗Department of Medicine, Helsinki University Central Hospital, Helsinki, Finland
- ††St. Michael’s Hospital, Toronto, Ontario, Canada
- ‡‡Danta Pazzanese Institute of Cardiology, São Paulo, Brazil
- §§Hungarian Institute of Cardiology, Semmelweis University, Budapest, Hungary
- ‖‖Cardiology and Uppsala Clinical Research Center, Uppsala University, Uppsala, Sweden
- ↵∗Reprint requests and correspondence:
Dr. Greg Flaker, University of Missouri, 1 Hospital Drive, CE306, Columbia, Missouri 65212.
Background Amiodarone is an effective medication in preventing atrial fibrillation (AF), but it interferes with the metabolism of warfarin.
Objectives This study sought to examine the association of major thrombotic clinical events and bleeding with the use of amiodarone in the ARISTOTLE (Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation) trial.
Methods Baseline characteristics of patients who received amiodarone at randomization were compared with those who did not receive amiodarone. The interaction between randomized treatment and amiodarone was tested using a Cox model, with main effects for randomized treatment and amiodarone and their interaction. Matching on the basis of a propensity score was used to compare patients who received and who did not receive amiodarone at the time of randomization.
Results In ARISTOTLE, 2,051 (11.4%) patients received amiodarone at randomization. Patients on warfarin and amiodarone had time in the therapeutic range that was lower than patients not on amiodarone (56.5% vs. 63.0%; p < 0.0001). More amiodarone-treated patients had a stroke or a systemic embolism (1.58%/year vs. 1.19%/year; adjusted hazard ratio [HR]: 1.47, 95% confidence interval [CI]: 1.03 to 2.10; p = 0.0322). Overall mortality and major bleeding rates were elevated, but were not significantly different in amiodarone-treated patients and patients not on amiodarone. When comparing apixaban with warfarin, patients who received amiodarone had a stroke or a systemic embolism rate of 1.24%/year versus 1.85%/year (HR: 0.68, 95% CI: 0.40 to 1.15), death of 4.15%/year versus 5.65%/year (HR: 0.74, 95% CI: 0.55 to 0.98), and major bleeding of 1.86%/year versus 3.06%/year (HR: 0.61, 95% CI: 0.39 to 0.96). In patients who did not receive amiodarone, the stroke or systemic embolism rate was 1.29%/year versus 1.57%/year (HR: 0.82, 95% CI: 0.68 to 1.00), death was 3.43%/year versus 3.68%/year (HR: 0.93, 95% CI: 0.83 to 1.05), and major bleeding was 2.18%/year versus 3.03%/year (HR: 0.72, 95% CI: 0.62 to 0.84). The interaction p values for amiodarone use by apixaban treatment effects were not significant.
Conclusions Amiodarone use was associated with significantly increased stroke and systemic embolism risk and a lower time in the therapeutic range when used with warfarin. Apixaban consistently reduced the rate of stroke and systemic embolism, death, and major bleeding compared with warfarin in amiodarone-treated patients and patients who were not on amiodarone.
Amiodarone is the most effective antiarrhythmic drug for the prevention of atrial fibrillation (AF) (1–3), and it is recommended for patients with frequent, symptomatic AF recurrences, especially in the presence of structural heart disease (4). However, several potential drawbacks may limit the use of amiodarone in patients with AF.
Warfarin is prescribed frequently to prevent stroke in patients with AF. Because amiodarone interferes with warfarin metabolism through the CYP 2C9 pathway, the international normalized ratio (INR) values are more difficult to maintain in a therapeutic range of 2 to 3 (5–8). This leads to a suboptimal time in the therapeutic range (TTR), which is associated with increased clinical events (9,10). Noncardiovascular side effects are also concerns with amiodarone. In addition to well-known hepatic, thyroid, neurologic, and ophthalmologic side effects, amiodarone occasionally causes pulmonary toxicity, which may be life-threatening (11,12), and it has also been associated with an increased risk of cancer (13–15). An analysis from the AFFIRM (Atrial Fibrillation Follow-Up Investigation of Rhythm Management) study noted excess noncardiovascular deaths with amiodarone (16).
We studied the effects of amiodarone on outcomes in the ARISTOTLE (Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation) trial. The first set of objectives compared the effects of randomized apixaban or warfarin treatment on stroke and systemic embolism, death, and bleeding events in relation to amiodarone use at the time of randomization. In the second set of objectives, we compared the rates of cardiovascular death and noncardiovascular death in relation to amiodarone use.
The design and results of the ARISTOTLE trial have been reported (17,18). Patients eligible for this study had AF documented on 2 occasions, at least 2 weeks apart within the 12 months before enrollment. AF was documented by electrocardiogram or rhythm strip, Holter monitor, or intracardiac recording, and lasted >1 minute. In addition, at least 1 of the following risk factors for stroke was required: age ≥75 years; previous stroke, transient ischemic attack (TIA), or systemic embolism; symptomatic heart failure within the previous 3 months or left ventricular ejection fraction (LVEF) ≤40%; and diabetes or hypertension requiring pharmacological therapy. Key exclusion criteria were AF due to a reversible cause, moderate or severe mitral stenosis, conditions other than AF that required anticoagulation (e.g., a prosthetic heart valve), stroke within the previous 7 days, need for aspirin >165 mg/day or for both aspirin and clopidogrel, and renal insufficiency with a creatinine level >2.5 mg/dl or a creatinine clearance of <25 ml/min.
Patients were randomized to receive either warfarin or apixaban. Warfarin was adjusted to achieve a target INR of 2.0 to 3.0. Apixaban was administered in doses of 5 mg twice daily or 2.5 mg twice daily in patients with 2 of the following criteria: age ≥80 years, body weight ≤60 kg, or serum creatinine ≥1.5 mg/dl.
The primary efficacy outcome of the study was stroke, which as defined as the abrupt onset of a nontraumatic, focal neurological deficit lasting at least 24 h, or systemic embolism, which was defined as symptoms consistent with acute loss of blood to a noncerebral artery confirmed by autopsy, angiography, vascular imaging, or some other objective testing. Secondary endpoints included myocardial infarction (MI) and death. Death was classified as cardiovascular (stroke, systemic embolism, MI, sudden death, heart failure, or indeterminate) or noncardiovascular (bleeding, malignancy, infection, trauma, pulmonary). The primary safety outcome was International Society of Thrombosis and Hemostasis major bleeding, which was defined as 1) bleeding that resulted in a decrease in hemoglobin of ≥2 g/dl over a 24-h period; 2) bleeding leading to a transfusion of ≥2 units of packed red blood cells; 3) bleeding occurring in a critical site; or 4) bleeding leading to death. A clinical events committee blinded to treatment assignment adjudicated all primary and secondary outcomes.
The baseline characteristics of those patients who received amiodarone at the time of randomization were compared with those who did not receive amiodarone. Continuous characteristics are reported as medians and 25th and 75th percentiles and compared using the Wilcoxon rank-sum test. Categorical variables are reported as frequencies and percentages and compared using chi-square tests. Events are summarized as rates per 100 patient-years of follow-up and the number of events. Kaplan-Meier curves were used to graphically describe the incidence of the primary efficacy and safety events by randomized treatment and amiodarone use at randomization. Hazard ratios (HRs) that compared randomized treatments (apixaban vs. warfarin) in patients on or off amiodarone were derived from Cox proportional hazards models. The interaction between randomized treatment and amiodarone was tested using a Cox model, with main effects for randomized treatment and amiodarone and their interaction. Propensity score–based matching was used to compare patients on and off amiodarone at the time of randomization. Variables included in the propensity score were selected from 35 baseline patient characteristics believed to be important in the selection of antiarrhythmic drug therapy. Candidate variables for inclusion in the propensity analysis were 1) demographic characteristics, including age, sex, and country of enrollment; 2) medical conditions, including a history of MI, coronary artery disease, pulmonary disease, intracranial bleeding, congestive heart failure, cardiomyopathy, valvular heart disease, congenital heart disease, unstable angina, diabetes mellitus, hepatic or renal disease, ventricular fibrillation and/or ventricular tachycardia cardiac arrest, stroke and/or TIA, peripheral artery disease, noncerebral thromboembolic event, and either carotid stent or carotid endarterectomy; 3) characteristics of AF or flutter, including duration of most recent episode; 4) findings on physical examination, including systolic blood pressure and body mass index; and 5) medications at the time of randomization, including beta-blockers and diuretics. Each patient on amiodarone was matched with 3 patients not on amiodarone enrolled in the same country. Countries where amiodarone was rarely or never used (≤3 patients) were excluded from this analysis (Finland, Malaysia, Norway, Puerto Rico, Singapore, and Turkey). Overall, 844 patients from these countries were excluded from this propensity analysis. Finally, to ensure that treatment effects with amiodarone did not vary between randomized therapies, the tests of interactions between amiodarone and randomized therapy were repeated with propensity-matched samples.
All statistical analyses were performed with SAS version 9.2 (SAS Institute, Inc., Cary, North Carolina) and R version 3.0.1 (Vienna, Austria).
There were 18,201 patients enrolled in ARISTOTLE. Amiodarone status was known at randomization in 17,907 patients; 2,051 (11%) patients received amiodarone at randomization and 15,856 patients did not receive amiodarone. The mean follow-up was 21.7 months for amiodarone patients and 21.8 months for patients not on amiodarone.
There were marked geographic differences in the use of amiodarone. Amidarone use was 17.9% in Latin America, 12% in Europe, 9.5% in the Asian Pacific countries, and 6.6% in North America.
In general, patients who received amiodarone were younger (median age 68 years vs. 70 years; p < 0.001), were more likely to have heart failure or a reduced LVEF (44.4% vs. 34.3%; p < 0.0001), were less likely to have diabetes (21.8% vs. 25.6%; p = 0.0003), and were less likely to have had a previous stroke, TIA, or systemic embolism (15.7% vs. 19.9%; p < 0.0001). The CHADS2 score was 2.0 ± 1.0 for patients on amiodarone and 2.1 ± 1.1 for patients not on amiodarone (p = 0.0003). The CHA2DS2-VASc score was 3.2 ± 1.5 for patients on amiodarone and 3.4 ± 1.5 for patients not on amiodarone. More patients on amiodarone had paroxysmal AF compared with those not on amiodarone (34.2% vs. 12.8%; p < 0.0001).
In patients assigned to warfarin, the mean TTR was 56.5% in patients who received amiodarone and 63.0% in patients who did not receive amiodarone (p < 0.0001). For those patients who received amiodarone versus those who did not, the mean times of follow-up below range (INR <2) were 28.5% and 24.2%, respectively (p < 0.0001). The mean times of follow-up above range (INR >3) were 15% and 12.8%, respectively (p < 0.0001).
Apixaban versus warfarin
Of the 2,051 patients who received amiodarone at baseline, 1,009 patients were assigned to apixaban and 1,042 were assigned to warfarin. There were no significant differences in selected stroke risk factors or in selected bleeding risk factors between patients receiving either apixaban or warfarin (Table 1).
For patients who received amiodarone, the rate of stroke or systemic embolism was 1.24%/year for patients assigned to apixaban and 1.85%/year for patients assigned to warfarin (HR: 0.68, 95% CI: 0.40 to 1.145). In patients who did not receive amiodarone, the rate of stroke or systemic embolism was 1.29%/year when assigned to apixaban and 1.57%/year when assigned to warfarin (HR: 0.82; 95% CI: 0.68 to 1.00) (Figure 1, Table 1). The interaction p value was 0.4776. For patients who received amiodarone, the rate of all-cause death was 4.15%/year when assigned to apixaban and 5.65%/year when assigned to warfarin (HR: 0.74; 95% CI: 0.55 to 0.98). In patients who did not receive amiodarone, the rate of all-cause death was 3.43%/year when assigned to apixaban and 3.68%/year when assigned to warfarin (HR: 0.93; 95% CI: 0.83 to 1.05). The interaction p value was 0.1366. Other efficacy analyses, including rates of cardiovascular death and noncardiovascular death, are shown in Table 2.
In terms of safety endpoints, in patients who received amiodarone at baseline, the rate of major bleeding was 1.86%/year when assigned to apixaban and 3.06%/year when assigned to warfarin (HR: 0.61; 95% CI: 0.39 to 0.96). In patients who did not receive amiodarone at baseline, the rate of major bleeding was 2.18%/year when assigned to apixaban and 3.03%/year in patients assigned to warfarin (HR: 0.72; 95% CI: 0.62 to 0.84) (Figure 2). The interaction p value was 0.4894. The rates of combined major and clinically relevant nonmajor bleeding in patients who received and did not receive amiodarone at baseline are shown in Table 2.
Major events associated with the use of amiodarone
Using a propensity score–adjusted analysis, patients who received amiodarone had a significantly higher rate of stroke or systemic embolism compared with those who did not receive amiodarone. The rate of stroke or systemic embolism was 1.58%/year in patients who received amiodarone and 1.19% in those who did not receive amiodarone (adjusted HR: 1.47; 95% CI: 1.03 to 2.10; p = 0.0322). Patients who received amiodarone had numerically, but not significantly, higher rates of all-cause death (4.76% vs. 4.09%; adjusted HR: 1.16; 95% CI: 0.95 to 1.41), cardiovascular death (2.65% vs. 2.26%; adjusted HR: 1.2; 95% CI: 0.91 to 1.55), and noncardiovascular death (1.49% vs. 1.09%; HR: 1.27; 95% CI: 0.88 to 1.82). There were no significant differences in MI and major bleeding between those who received and those who did not receive amiodarone (Table 3).
This analysis from the ARISTOTLE study included 2,051 patients who received amiodarone, which represented one of the largest experiences to date with amiodarone for AF in a clinical trial. The main findings were as follows: 1) amiodarone use was associated with a significantly increased risk of stroke and systemic embolism, and a lower time in the TTR when used with warfarin; 2) apixaban consistently reduced the rate of stroke and systemic embolism, death, and major bleeding compared with warfarin in amiodarone patients and patients not on amiodarone (Central Illustration); and 3) the relative reductions in stroke and systemic embolism, mortality, and major bleeding with apixaban compared with warfarin were similar with and without amiodarone use.
The quality of anticoagulation, as assessed by the mean TTR during the trial, was 56.5% in patients treated with warfarin combined with amiodarone. In clinical trials, TTR values of 62% to 66% were achieved (19–22), and were accepted as an indication of effective anticoagulation. In ARISTOTLE, the mean TTR was 63% in patients who did not receive amiodarone and who were assigned to warfarin. Despite the difference in TTR, there was a consistent risk reduction in stroke and systemic embolism, mortality, and major bleeding with apixaban compared with warfarin, regardless of treatment with amiodarone. These findings were in accordance with our previous observation on the lack of interaction between INR control and the benefits of apixaban compared with warfarin in the overall ARISTOTLE trial (23). These data suggested that apixaban is an attractive choice to prevent stroke and systemic embolism, lower mortality, and reduce major bleeding, irrespective of whether the patient receives or does not receive amiodarone.
The use of class IA antiarrhythmic drugs in AF has been associated with excess mortality (24,25). Whether class IC or class III agents, including amiodarone, are associated with higher mortality in patients with AF has been controversial. Excess mortality has not been noted in a pooled analysis of class IC and class III drugs after cardioversion (26). However, antiarrhythmic drug therapy in the AFFIRM study, which included amiodarone, sotalol, and 6 class I antiarrhythmic drugs, was associated with excess mortality (27), and noncardiovascular mortality was increased in patients on amiodarone (15,16). Reports of increased cancer rates with amiodarone have been published (13,14).
Despite these concerns, the use of amiodarone in ARISTOTLE was substantial; 11% of patients received this drug. There was wide geographic variability in amidarone use; its use was more likely in Latin America and less likely in North America. The reason for these differences was uncertain, but it might represent differences in the patient population, regional differences in perceptions about the merits of rhythm control, or differences in the availability of catheter-based techniques for rhythm control.
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. There is no a priori reason for class III antiarrhythmic medications to be thrombogenic. In the AFFIRM study, amiodarone was the most frequently used medication in the rhythm control arm, and there were no differences in cerebrovascular events in patients assigned to the rhythm control arm compared with the rate control arm (15). The multichannel blocker, dronedarone, with class III properties and negligible effects on INR values, was associated with a reduction in stroke in patients with paroxysmal or persistent AF (28), but was also associated with an increased stroke risk in patients with permanent AF (29). These data suggest that the thromboembolic risk is likely related to characteristics of the AF population, rather than the specific antiarrhythmic drug. In the present analysis, the mean TTR in patients who received amiodarone and warfarin was low, and the majority of values were below the accepted therapeutic range. It is possible that less effective anticoagulation with amiodarone could contribute to the observed association with stroke or systemic embolism in the warfarin group, where the HR for stroke or systemic embolism with amiodarone was somewhat greater, although there was no significant interaction. These findings indicate that special efforts are needed to maintain an in-range INR in patients who receive warfarin and amiodarone. These findings provide an additional rationale to prefer apixaban more than warfarin in patients who require amiodarone.
The concern about cancer and amiodarone use in humans was based on the results from a meta-analysis of 15 trials, 4 of which reported cancer deaths. Of 1,609 patients who received amiodarone, 13 (0.8%) died of cancer. In comparison, of 1,597 patients who received placebo, 4 (0.3%) died of cancer (13). A more recent retrospective cohort study from Taiwan included patients who received amiodarone for >28 days without antecedent malignancy (14). Of 6,418 patients followed for a median of 2.57 years, 280 cancers developed, which was higher than would be expected. Cancer was not an endpoint of the ARISTOTLE trial, and the follow-up time was relatively brief for the detection of any possible association of amiodarone with cancer.
This was a retrospective evaluation of the ARISTOTLE database; therefore, it was subject to all the biases inherent to this type of analysis. Amiodarone use was defined at the time of randomization, and its use was not systematically collected throughout the study. The duration of amiodarone use before entry into the study was not determined. At baseline, patients were classified with either paroxysmal or persistent and/or permanent AF, and patients with persistent and permanent AF were not separated into different groups. Patients who did not receive amiodarone at the time of randomization might have received it in the past. The follow-up was too brief to detect important long-term problems with the medication, including the development of cancer. Finally, because patients who received or did not receive amiodarone were not randomized, there were important differences between them. Although we attempted to compensate for these differences with a propensity score method, assumptions were made in the development of this scoring system, and there might have been residual unmeasured confounding. Therefore, the possibility of confounding factors or even the play of chance as a cause for the associations between amiodarone use and outcomes could not be excluded.
In a contemporary clinical trial involving patients with AF at risk for stroke, amiodarone use was substantial and varied by geographic region. Anticoagulation quality, as assessed by TTR, was lower in warfarin-treated patients who received amiodarone, and amiodarone use was associated with a significantly higher risk of stroke and systemic embolism. The relative effects of apixaban versus warfarin on stroke and systemic embolism, mortality, and major bleeding were consistent in patients who received and patients who did not receive amiodarone.
COMPETENCY IN MEDICAL KNOWLEDGE 1: Amiodarone is effective in preventing AF, but it can interfere with warfarin metabolism and potentially increase the risk of thromboembolic events or bleeding. Newer target-specific oral anticoagulants have less interaction with amiodarone.
COMPETENCY IN MEDICAL KNOWLEDGE 2: Compared with warfarin, in the ARISTOTLE trial, the factor Xa inhibitor, apixaban, reduced rates of stroke and systemic embolism, death, and major bleeding in patients with nonvalvular AF, whether or not they were concurrently treated with amiodarone.
TRANSLATIONAL OUTLOOK: Randomized studies of amiodarone and other antiarrhythmic medications in anticoagulated patients with AF could clarify the generalizability of these findings and better guide treatment decisions in clinical practice.
This study was sponsored by Bristol-Myers Squibb and Pfizer. Dr. Flaker has received grants from Boehringer Ingelheim and Sanofi Aventis; and consulting fees from Bristol-Myers Squibb, Boehringer Ingelheim, Pfizer, Daiichi Sankyo, and Sanofi-Aventis. Dr. Lopes has received grants from Bristol-Myers Squibb, AstraZeneca, Boehringer Ingelheim, and Daiichi Sankyo; and consulting fees from Bristol-Myers Squibb, Bayer, Boehringer Ingelheim, GlaxoSmithKline, and Pfizer. Dr. Hylek has received consulting fees and honoraria from Bristol-Myers Squibb, Daiichi Sankyo, Boehringer Ingelheim, Johnson & Johnson, Pfizer, and Bayer; research grants from Bristol-Myers Squibb; and is a member of the advisory boards for Bayer Healthcare, Boehringer Ingelheim, Bristol-Myers Squibb, Daiichi Sankyo, Janssen, Pfizer, and Roche. Dr. Hohnloser has received grants from Sanofi Aventis and St. Jude Medical; consulting fees from Bayer Healthcare, Boehringer Ingelheim, Bristol-Myers Squibb, Boston Scientific, Cardiome, Gilead, Johnson & Johnson, Pfizer, Portola, Sanofi Aventis, Servier, St. Jude Medical, and Zoll; and lecture fees from Bayer Healthcare, Boehringer Ingelheim, Bristol-Myers Squibb, Pfizer, Sanofi Aventis, St. Jude Medical, and Medtronic. Dr. Garcia has received consulting fees from Boehringer Ingelheim, Bristol-Myers Squibb, Daiichi Sankyo, Pfizer, Bayer Healthcare, and CSL Behring; and fees for the development of educational materials from Boehringer Ingelheim. Dr. Harjola has received consulting fees from Bristol-Myers Squibb/Pfizer. Dr. Dorian has received consulting fees and honoraria from Sanofi Aventis, Bayer Healthcare, Bristol-Myers Squibb, Pfizer, Servier, and Boehringer Ingelheim. Dr. Avezum has received research grants from AstraZeneca, Bayer, Bristol-Myers Squibb, Pfizer, and Boehringer Ingelheim; and consulting fees and honoraria from AstraZeneca, Bayer, Bristol-Myers Squibb, Boehringer Ingelheim, GlaxoSmithKline, and Pfizer. Dr. Wallentin has received consulting fees and honoraria from Abbott, AstraZeneca, Bristol-Myers Squibb/Pfizer, GlaxoSmithKline, Regado Biosciences, Boehringer Ingelheim, Athera Biotechnologies, and AstraZeneca; and research grants from Boehringer Ingelheim, AstraZeneca, GlaxoSmithKline, Merck/Schering-Plough, and Bristol-Myers Squibb/Pfizer. Dr. Granger has received research grants and consulting fees from Boehringer Ingelheim, Bristol-Myers Squibb, Bayer Healthcare, Daiichi Sankyo, Sanofi Aventis, and Pfizer. All other authors have reported that they have no relationships relevant to this paper to disclose.
- Abbreviations and Acronyms
- atrial fibrillation
- congestive heart failure, hypertension, age, diabetes, and stroke
- confidence interval
- hazard ratio
- international normalized ratio
- left ventricular ejection fraction
- myocardial infarction
- transient ischemic attack
- time in therapeutic range
- Received March 13, 2014.
- Revision received June 17, 2014.
- Accepted July 8, 2014.
- American College of Cardiology Foundation
- AFFIRM First Antiarrhythmic Drug Substudy Investigators
- Camm A.J.,
- Lip G.Y.,
- De Caterina R.,
- et al.
- Flaker G.C.,
- Pogue J.,
- Yusuf S.,
- et al.,
- for the Atrial Fibrillation Clopidogrel Trial With Irbesartan for Prevention of Vascular Events (ACTIVE) Investigators
- Wallentin L.,
- Yusuf S.,
- Ezekowitz M.,
- et al.,
- for the RE-LY Investigators
- Wan Y.,
- Heneghan C.,
- Perera R.,
- et al.
- Connolly S.,
- Pogue J.,
- Eikelboom J.,
- et al.,
- ACTIVE W Investigators
- Greenspon A.J.,
- Kidwell G.A.,
- Hurley W.,
- et al.
- Sobol S.M.,
- Rakita L.
- Piccini J.P.,
- Berger J.S.,
- O’Connor C.M.
- Steinberg J.,
- Sadaniantz A.,
- Kron J.,
- et al.
- Saksena S.,
- Slee A.,
- Waldo A.L.,
- et al.
- ACTIVE Writing Group of the ACTIVE Investigators,
- Connolly S.,
- Pogue J.,
- Hart R.,
- et al.
- Olsson S.B.,
- Executive Steering Committee of the SPORTIF III Investigators
- Wallentin L.,
- Lopes R.D.,
- Hanna M.,
- et al.,
- for the Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation (ARISTOTLE) Investigators
- Coplen S.E.,
- Antman E.M.,
- Berlin J.A.,
- et al.
- Flaker G.C.,
- Blackshear J.L.,
- McBride R.,
- et al.
- Corley S.D.,
- Epstein A.E.,
- DiMarco J.P.,
- et al.,
- for the AFFIRM Investigators
- Connolly S.J.,
- Crijns H.J.,
- Torp-Pedersen C.,
- et al.,
- for the ATHENA Investigators