Direct Oral Anticoagulants Versus Warfarin in Patients With Atrial Fibrillation: Patient-Level Network Meta-Analyses of Randomized Clinical Trials With Interaction Testing by Age and Sex
This article has been corrected.
VIEW CORRECTIONAbstract
Background:
Direct oral anticoagulants (DOACs) are preferred over warfarin for stroke prevention in atrial fibrillation. Meta-analyses using individual patient data offer substantial advantages over study-level data.
Methods:
We used individual patient data from the COMBINE AF (A Collaboration Between Multiple Institutions to Better Investigate Non-Vitamin K Antagonist Oral Anticoagulant Use in Atrial Fibrillation) database, which includes all patients randomized in the 4 pivotal trials of DOACs versus warfarin in atrial fibrillation (RE-LY [Randomized Evaluation of Long-Term Anticoagulation Therapy], ROCKET AF [Rivaroxaban Once Daily Oral Direct Factor Xa Inhibition Compared With Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation], ARISTOTLE [Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation], and ENGAGE AF-TIMI 48 [Effective Anticoagulation With Factor Xa Next Generation in Atrial Fibrillation–Thrombolysis in Myocardial Infarction 48]), to perform network meta-analyses using a stratified Cox model with random effects comparing standard-dose DOAC, lower-dose DOAC, and warfarin. Hazard ratios (HRs [95% CIs]) were calculated for efficacy and safety outcomes. Covariate-by-treatment interaction was estimated for categorical covariates and for age as a continuous covariate, stratified by sex.
Results:
A total of 71 683 patients were included (29 362 on standard-dose DOAC, 13 049 on lower-dose DOAC, and 29 272 on warfarin). Compared with warfarin, standard-dose DOACs were associated with a significantly lower hazard of stroke or systemic embolism (883/29 312 [3.01%] versus 1080/29 229 [3.69%]; HR, 0.81 [95% CI, 0.74–0.89]), death (2276/29 312 [7.76%] versus 2460/29 229 [8.42%]; HR, 0.92 [95% CI, 0.87–0.97]), and intracranial bleeding (184/29 270 [0.63%] versus 409/29 187 [1.40%]; HR, 0.45 [95% CI, 0.37–0.56]), but no statistically different hazard of major bleeding (1479/29 270 [5.05%] versus 1733/29 187 [5.94%]; HR, 0.86 [95% CI, 0.74–1.01]), whereas lower-dose DOACs were associated with no statistically different hazard of stroke or systemic embolism (531/13 049 [3.96%] versus 1080/29 229 [3.69%]; HR, 1.06 [95% CI, 0.95–1.19]) but a lower hazard of intracranial bleeding (55/12 985 [0.42%] versus 409/29 187 [1.40%]; HR, 0.28 [95% CI, 0.21–0.37]), death (1082/13 049 [8.29%] versus 2460/29 229 [8.42%]; HR, 0.90 [95% CI, 0.83–0.97]), and major bleeding (564/12 985 [4.34%] versus 1733/29 187 [5.94%]; HR, 0.63 [95% CI, 0.45–0.88]). Treatment effects for standard- and lower-dose DOACs versus warfarin were consistent across age and sex for stroke or systemic embolism and death, whereas standard-dose DOACs were favored in patients with no history of vitamin K antagonist use (P=0.01) and lower creatinine clearance (P=0.09). For major bleeding, standard-dose DOACs were favored in patients with lower body weight (P=0.02). In the continuous covariate analysis, younger patients derived greater benefits from standard-dose (interaction P=0.02) and lower-dose DOACs (interaction P=0.01) versus warfarin.
Conclusions:
Compared with warfarin, DOACs have more favorable efficacy and safety profiles among patients with atrial fibrillation.
Clinical Perspective
What Is New?
•
When individual patient data from randomized trials of direct oral anticoagulants (DOACs) versus warfarin are analyzed collectively, standard-dose DOAC use results in lower incidence of stroke, death, and intracranial hemorrhage with no difference in major bleeding.
•
The relative benefits of standard-dose DOACs over warfarin for stroke prevention were consistent across nearly all subgroups, including across the entire continuous spectrum of age, with no evidence of interaction by sex. These benefits may be greater in patients with lower creatinine clearance.
•
For major bleeding, younger patients and patients with lower body weight may derive a greater benefit from standard-dose DOAC over warfarin.
What Are the Clinical Implications?
•
The totality of efficacy and safety data from randomized clinical trials supports the use of standard-dose DOACs over warfarin for stroke prevention in nonvalvular atrial fibrillation, regardless of age or sex.
Editorial, see p 256
Direct oral anticoagulants (DOACs) are recommended by both European and North American guidelines as first-line therapy for prevention of ischemic stroke in patients with atrial fibrillation (AF).1,2 Four DOACs (dabigatran, rivaroxaban, apixaban, and edoxaban) have obtained regulatory approval and guideline recommendations for stroke prevention in patients with AF on the basis of data from 4 pivotal randomized trials comparing DOAC versus warfarin.3–6 These trials excluded patients with moderate or severe mitral stenosis and with mechanical prosthetic valves (ie, valvular AF), leading to a product-labeled indication for all DOACs for nonvalvular AF that will henceforth be referred to as AF in the current report.
Previously published trial-level meta-analyses used aggregate data from the 4 pivotal trials and demonstrated that DOAC use is associated with significant reductions in stroke, intracranial hemorrhage, and death compared with warfarin, with no statistically different risk of major bleeding.7 Study-level meta-analyses, however, are subject to important limitations.8 Meta-analyses using individual patient data offer important advantages over study-level data. Individual patient data meta-analyses allow for analyses of individual patient-level time-to-event censored survival data and application of consistent follow-up time across trials, rather than simply pooling study-level hazard ratios (HRs) that were estimated under different settings across individual trials. Individual patient-level meta-analyses also allow for analyses of continuous variables and a more thorough assessment of treatment effect heterogeneity.9–11
The COMBINE AF (A Collaboration Between Multiple Institutions to Better Investigate Non-Vitamin K Antagonist Oral Anticoagulant Use in Atrial Fibrillation) database contains individual patient data from the 4 pivotal trials of DOACs versus warfarin in patients with AF.12 We used data from the COMBINE AF database to perform network meta-analyses, aimed at assessing the overall safety and efficacy of DOACs versus warfarin, including 2 different DOAC treatment strategies (standard dose and lower dose). In these network meta-analyses, we aimed to leverage the strengths of individual patient data and estimate treatment effects by standardizing follow-up duration for time-to-event outcomes and study population across trials and to assess effect modification with a Cox regression model as well as across the spectrum of age as a continuous covariate.
Methods
Anonymized data from the COMBINE AF database are shared by members of the COMBINE AF executive committee and their corresponding institutions; however, these data are unable to be shared outside of these institutions because of preexisting data privacy restrictions. Individual investigators may reach out directly to a member of the COMBINE AF executive committee to discuss opportunities for collaboration.
Study Selection, Data Sources, and Treatment Strategies
The design of and rationale for COMBINE AF have been published previously12 and a list of COMBINE AF investigators can be found in Table S1. COMBINE AF contains individual patient data from RE-LY (Randomized Evaluation of Long-Term Anticoagulation Therapy),3 ROCKET AF (Rivaroxaban Once Daily Oral Direct Factor Xa Inhibition Compared With Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation),4 ARISTOTLE (Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation),5 and ENGAGE AF-TIMI 48 (Effective Anticoagulation With Factor Xa Next Generation in Atrial Fibrillation–Thrombolysis in Myocardial Infarction 48),6 which studied dabigatran, rivaroxaban, apixaban, and edoxaban, respectively, versus warfarin in patients with AF. All individual trials were performed in accordance with local data protection regulations that were in place at the time of study conduct, were approved by the local institutional review board, and all study participants provided written informed consent. Creation of the COMBINE AF database and the statistical analyses for this article were approved by the Duke University institutional review board. COMBINE AF is registered with PROSPERO (URL: https://www.crd.york.ac.uk/prospero/; Unique identifier: CRD42020178771).
For these analyses, a standard-dose DOAC treatment strategy was defined as dabigatran 150 mg twice daily (RE-LY), rivaroxaban 20 mg (or 15 mg if dose reduction criteria were met) once daily (ROCKET AF), apixaban 5 mg (or 2.5 mg if dose reduction criteria were met) twice daily (ARISTOTLE), or edoxaban 60 mg (or 30 mg if dose reduction criteria were met) once daily (ENGAGE AF-TIMI 48). A lower-dose DOAC treatment strategy was defined as dabigatran 110 mg twice daily (RE-LY) or edoxaban 30 mg (or 15 mg if dose reduction criteria were met) once daily (ENGAGE AF-TIMI 48). Patients in our meta-analyses were analyzed according to their randomization group regardless of whether they were treated with dose reduction by individual trial criteria.
Outcomes
All outcomes were adjudicated as described in the individual trials and were assessed as time to first event. Charter definitions of the adjudicated outcomes were similar across each individual trial. The primary efficacy outcome was a composite outcome of any stroke (ischemic, hemorrhagic, or other) or systemic embolism (ie, stroke or systemic embolism). Secondary efficacy outcomes included all-cause death, cardiovascular death, ischemic stroke, systemic embolism, hemorrhagic stroke, any stroke, and a composite efficacy outcome consisting of ischemic stroke, systemic embolism, or cardiovascular death.
The primary safety outcome was major bleeding as defined by the International Society on Thrombosis and Haemostasis.13 Secondary safety outcomes included fatal bleeding, major or clinically relevant nonmajor bleeding, any bleeding (including fatal, major, clinically relevant nonmajor, or minor bleeding), intracranial bleeding, and gastrointestinal bleeding (adjudicated major bleeding events determined to be from gastrointestinal bleeding events only).
Two net clinical benefits were assessed. These were the composite of any stroke, systemic embolism, all-cause death, or major bleeding and the composite of any stroke, systemic embolism, all-cause death, or intracranial bleeding.
Study Population
Efficacy outcomes were assessed using an intention-to-treat population; safety outcomes and net clinical benefits were assessed in the safety population as defined by the individual trials. A common definition of safety populations across trials was patients who received ≥1 dose of study drug and were followed for events occurring between the dates the patient began study drug and ≤2 days (or ≤3 in ENGAGE-AF TIMI 48) after the last dose of study drug. To account for differences in follow-up duration between trials, patients were censored at 32 months, which was the point at which <10% of patients remained at risk across all studies (Table S2).
Statistical Analyses
We performed network meta-analyses to compare time-to-event measures of efficacy and safety outcomes and net clinical benefits for 3 treatment strategies (standard-dose DOAC, lower-dose DOAC, and warfarin).14 For the primary analyses, Kaplan-Meier curves were generated for key outcomes and univariable stratified Cox proportional hazard models were fitted including treatment strategy as an independent variable. Cox models were stratified by trial allowing random effects to account for cross-trial heterogeneity. To evaluate treatment strategies, we compared HRs with 95% CIs for standard-dose or lower-dose DOACs versus warfarin. Between-trial heterogeneity was assessed by the estimated SD of random effects. A larger standard error of the HR, estimate is expected with large between-study heterogeneity compared with that estimated under fixed effect Cox models. We evaluated the proportional hazards assumption using graphical approach of Kaplan-Meier curves and the global Schoenfeld test.15 There was no strong evidence of violation of the proportional hazard assumption for any of the examined outcomes. To report trial-specific HRs, we fitted a Cox model to individual trials. The secondary analyses assessed effect modification by fitting stratified Cox proportional hazards models with random effects including baseline covariate-by-treatment interaction. For categorical covariates, HRs for standard-dose or lower-dose DOACs compared with warfarin were calculated and the associated 95% CIs were reported with P values for interaction.
Event rates (% per year) were calculated by categorical baseline body weight and creatinine clearance for each randomized treatment arm. To assess treatment effect variation across age, we fitted a separate model and calculated change in HR, per unit change in age with associated 95% CIs and interaction P values. In addition, we fitted an extended model including a 3-way interaction of age, sex, and treatment to further assess whether the age-by-treatment interaction differs by sex. We did not adjust interaction P values for multiplicity because these analyses were not confirmatory but rather exploratory.16 When assessing an interaction, we considered a P value <0.1 to indicate potentially meaningful evidence of effect modification, because testing for interactions has limited statistical power.16,17
We conducted sensitivity analyses limiting the study population only to trials of factor Xa inhibitors (ie, excluding all patients from RE-LY; Figure S1). For these analyses, the lower-dose edoxaban arm from ENGAGE AF-TIMI 48 was not analyzed, because these analyses would have served only to replicate findings from the individual trial.
Results
Patient Characteristics
A total of 71 683 patients were included in these analyses (Figure S1; n=29 362 randomized to standard-dose DOAC, n=13 049 randomized to lower-dose DOAC, and n=29 272 randomized to warfarin). After censoring at 32 months, the median (25th, 75th percentile) follow-up duration was 26.6 (18.9, 32.0) months. Baseline characteristics by treatment strategy can be found in Table 1. No clinically meaningful differences were observed across randomized treatment groups. Baseline characteristics by trial and extent of missing baseline variables from individual trials have been published previously.12
| Demographics | All patients (n=71 683) | Pooled by treatment strategy | ||
|---|---|---|---|---|
| Standard-dose DOAC (n=29 362) | Lower-dose DOAC (n=13 049) | Warfarin (n=29 272) | ||
| Age, y | 72 (65–77) | 72 (65–77) | 72 (66–77) | 72 (65–77) |
| <65 | 24.0 | 24.7 | 21.4 | 24.6 |
| 65–75 | 37.4 | 36.9 | 39.1 | 37.2 |
| ≥75 | 38.6 | 38.5 | 39.5 | 38.2 |
| Female | 37.3 | 37.4 | 37.4 | 37.1 |
| Race/ethnicity | ||||
| White | 80.2 | 80.7 | 77.8 | 80.7 |
| Black | 1.2 | 1.3 | 1.1 | 1.2 |
| Asian | 14.2 | 14.1 | 14.8 | 14.2 |
| Other | 4.4 | 3.9 | 6.2 | 4.0 |
| Hispanic | 9.7 | 11.3 | 2.7 | 11.3 |
| Vital signs | ||||
| Weight, kg | 81 (70–95) | 81 (70–95) | 81 (70–95) | 81 (70–94) |
| <60 | 9.3 | 9.4 | 9.0 | 9.5 |
| 60–120 | 85.7 | 85.6 | 86.1 | 85.6 |
| ≥120 | 4.8 | 4.9 | 4.9 | 4.8 |
| Systolic BP, mm Hg | 130 (120–140) | 130 (120–140) | 130 (120–140) | 130 (120–140) |
| BMI, kg/m2 | 28.3 (25.2–32.2) | 28.4 (25.2–32.2) | 28.4 (25.2–32.2) | 28.3 (25.1–32.2) |
| <25 | 23.7 | 23.6 | 23.5 | 23.9 |
| 25–30 | 38.1 | 37.9 | 38.7 | 38.2 |
| ≥30 | 37.9 | 38.3 | 37.6 | 37.6 |
| Alcohol use* | ||||
| None/rare | 60.2 | 62.6 | 48.4 | 63.0 |
| Light/moderate | 13.2 | 15.3 | 4.6 | 15.0 |
| Heavy | 1.3 | 1.4 | 0.9 | 1.4 |
| Medical history | ||||
| Diabetes | 30.8 | 31.1 | 30.3 | 30.8 |
| Hypertension | 87.7 | 87.8 | 86.7 | 88.0 |
| CAD | 29.9 | 29.7 | 30.9 | 29.8 |
| MI | 14.6 | 14.7 | 14.0 | 14.9 |
| CABG* | 5.2 | 5.5 | 3.7 | 5.4 |
| PCI* | 6.0 | 6.5 | 4.0 | 6.5 |
| Heart failure | 46.4 | 46.8 | 45.3 | 46.6 |
| Stroke or TIA | 28.1 | 28.8 | 24.5 | 29.0 |
| Paroxysmal AF | 23.2 | 21.6 | 28.8 | 22.2 |
| Smoking (ever) | 43.6 | 43.4 | 46.3 | 42.6 |
| CHADS2 score | ||||
| 0–1 | 16.7 | 17.2 | 15.1 | 16.9 |
| 2 | 34.5 | 32.6 | 41.3 | 33.3 |
| ≥3 | 48.8 | 50.2 | 43.6 | 49.8 |
| Previous GI bleeding* | 2.8 | 3.0 | 1.9 | 3.1 |
| Previous non-GI bleeding* | 5.6 | 6.0 | 4.0 | 5.8 |
| Baseline medications | ||||
| Previous VKA use (ever) | 68.2 | 67.6 | 71.5 | 67.4 |
| Aspirin | 33.8 | 33.6 | 33.9 | 33.9 |
| Thienopyridine | 3.0 | 2.8 | 4.0 | 2.9 |
| β-blocker | 64.3 | 64.3 | 64.8 | 64.1 |
| Calcium channel blocker | 30.5 | 30.1 | 31.9 | 30.3 |
| NSAID | 4.1 | 4.3 | 2.8 | 4.5 |
| Digoxin | 31.9 | 32.1 | 29.4 | 32.7 |
| PPI | 11.9 | 12.2 | 10.5 | 12.3 |
| Amiodarone | 10.6 | 10.6 | 10.9 | 10.6 |
| Laboratory studies | ||||
| Creatinine clearance, mL/min | 70.0 (54.0–90.3) | 70.0 (54.0–91.0) | 69.5 (53.6–90.0) | 70.0 (54.0–90.0) |
| ≤50 | 19.6 | 19.6 | 19.8 | 19.4 |
| 51–80 | 44.4 | 44.0 | 45.1 | 44.5 |
| >80 | 35.8 | 36.1 | 35.1 | 35.9 |
| LVEF* | ||||
| Normal | 46.5 | 47.2 | 42.5 | 47.5 |
| Mild | 11.6 | 11.9 | 10.1 | 12.0 |
| Moderate | 8.2 | 8.6 | 6.8 | 8.5 |
| Severe | 3.8 | 3.8 | 3.2 | 3.9 |
Continuous variables listed as median (25th–75th percentile). Categorical variables listed as %. Creatinine clearance calculated using Cockcroft-Gault equation. Standard-dose direct oral anticoagulant (DOAC) includes the dabigatran 150 mg twice daily arm from RE-LY (Randomized Evaluation of Long-Term Anticoagulation Therapy), the rivaroxaban arm from ROCKET AF (Rivaroxaban Once Daily Oral Direct Factor Xa Inhibition Compared With Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation), the apixaban arm from ARISTOTLE (Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation), and the standard-dose edoxaban arm from ENGAGE AF-TIMI 48 (Effective Anticoagulation With Factor Xa Next Generation in Atrial Fibrillation–Thrombolysis in Myocardial Infarction 48). Lower-dose DOAC includes the dabigatran 110 mg twice daily arm from RE-LY and the lower-dose edoxaban arm from ENGAGE AF-TIMI 48. AF indicates atrial fibrillation; BMI, body mass index; CABG, coronary artery bypass graft; CAD, coronary artery disease; GI, gastrointestinal; LVEF, left ventricular ejection fraction; MI, myocardial infarction; NSAID, nonsteroidal anti-inflammatory drug; PCI, percutaneous coronary intervention; PPI, proton pump inhibitor; TIA, transient ischemic attack; and VKA, vitamin K antagonist.
*
Data missing from ≥1 of the individual trials. See reference 12 for detailed information regarding missing data.
In the study population, the median (25th, 75th percentile) age was 72 (65, 77) years, 37.3% were female, 34.5% had a CHADS2 score of 2, and 48.8% had a CHADS2 score ≥3 (Table 1). More than two-thirds of patients (68.2%) had used a vitamin K antagonist (VKA) before randomization and 33.8% were on aspirin at baseline. A total of 19.6% of patients had creatinine clearance ≤50 at baseline.
Efficacy Outcomes
All efficacy outcomes showed little or no between-trial heterogeneity, with estimated SD of random effects across trials close to 0 (Table S4). HRs for the primary efficacy outcome from each individual trial can be found in Table S5.
Compared with warfarin, patients randomized to standard-dose DOAC had a lower hazard of stroke or systemic embolism (883/29 312 [3.01%] versus 1080/29 229 [3.69%]; HR, 0.81 [95% CI, 0.74–0.89]) and a lower hazard of all-cause death, cardiovascular death, systemic embolism, hemorrhagic stroke, any stroke, and the composite efficacy outcome (ischemic stroke, systemic embolism, or cardiovascular death) over the duration of follow-up (Figure 1). Kaplan-Meier curves for key efficacy outcomes can be found in Figure 2.
Compared with warfarin, patients randomized to lower-dose DOAC (ie, dabigatran 110 mg or edoxaban 30/15 mg) had no statistically different hazard of stroke or systemic embolism (531/13 049 [4.07%] versus 1080/29 229 [3.69%]; HR, 1.06 [95% CI, 0.95–1.19]), but had a lower hazard of all-cause death, cardiovascular death, and hemorrhagic stroke (Figure 1). Patients randomized to lower-dose DOAC had a significantly higher hazard of ischemic stroke compared with warfarin (454/13 049 [3.48%] versus 685/29 229 [2.34%]; HR, 1.35 [95% CI, 1.19–1.54]).
HRs with 95% CIs for pairwise comparison of standard-dose DOAC versus lower-dose DOAC can be found in Table S3. Compared with lower-dose DOAC, patients randomized to standard-dose DOAC had a lower hazard of stroke or systemic embolism (883/29 312 [3.01%] versus 531/13 049 [4.07%]; HR, 0.76 [95% CI, 0.68–0.86]) and a lower hazard of any stroke, ischemic stroke, and the composite efficacy outcome.
Safety Outcomes and Net Clinical Benefits
All safety outcomes except fatal bleeding and hemorrhagic stroke showed moderate between-trial heterogeneity with estimated SD of random effects across trials larger than 0.1 (Table S4). HRs for the primary safety outcome from each individual trial can be found in Table S5.
Compared with warfarin, patients randomized to standard-dose DOAC had no statistically different hazard of major bleeding (1479/29 270 [5.05%] versus 1733/29 187 [5.94%]; HR, 0.86 [95% CI, 0.74–1.01]), but a lower hazard of fatal bleeding and intracranial bleeding (Figure 1). Patients randomized to standard-dose DOAC had a significantly higher hazard of major gastrointestinal bleeding compared with warfarin (744/29 270 [2.54%] versus 569/29 187 [1.95%]; HR, 1.31 [95% CI, 1.08–1.57]). Kaplan-Meier curves for key safety outcomes can be found in Figure 2.
Compared with warfarin, patients randomized to lower-dose DOAC had a lower hazard of major bleeding (564/12 985 [4.34%] versus 1733/29 187 [5.94%]; HR, 0.63 [95% CI, 0.45–0.88]) and a lower hazard of fatal bleeding, major or clinically relevant nonmajor bleeding, any bleeding, and intracranial bleeding (Figure 1). Patients randomized to lower-dose DOAC had no statistically different risk of major gastrointestinal bleeding compared with warfarin (271/12 985 [2.09%] versus 569/29 187 [1.95%]; HR, 0.85 [95% CI, 0.62–1.18]).
Compared with lower-dose DOAC, patients randomized to standard-dose DOAC had no statistically different hazard of major bleeding (1479/29 270 [5.05%] versus 564/12 985 [4.34%]; HR, 1.37 [95% CI, 0.95–1.96]) and fatal bleeding, but had a higher hazard of major or clinically relevant nonmajor bleeding, intracranial bleeding, hemorrhagic stroke, and gastrointestinal bleeding (Table S3).
Effect Modification by Categorical Baseline Covariates
For stroke or systemic embolism, potentially meaningful interactions suggesting a greater benefit of standard-dose DOAC over warfarin were observed in patients with no history of previous VKA use and in patients with lower creatinine clearance (Figure 3). Potentially meaningful interactions suggesting a greater benefit of lower-dose DOAC over warfarin were observed in patients with no history of previous VKA use, with no history of previous gastrointestinal bleeding, and with older age (Table S6). Stroke or systemic embolism event rates for standard-dose and lower-dose DOAC versus warfarin by categorical baseline body weight and creatinine clearance can be found in Table S7 and Table S8, respectively.
For major bleeding, potentially meaningful interactions suggesting a benefit of standard-dose DOAC over warfarin were observed in patients without diabetes, with no history of coronary artery disease, with no history of gastrointestinal bleeding, with younger age, and with lower weight (Figure 3). Potentially meaningful interactions suggesting a benefit of lower-dose DOAC over warfarin were observed in patients with no history of coronary artery disease, no history of heart failure, and younger age (Table S9). Major bleeding event rates for standard-dose and lower-dose DOAC versus warfarin by categorical baseline body weight and creatinine clearance can be found in Table S7 and Table S8, respectively.
Effect Modification by Continuous Age and Sex
Baseline age in the study population ranged from 19 to 101 years. The 5th and 95th percentiles for age were 54 and 84 years. For stroke or systemic embolism, no meaningful interaction was observed for standard-dose or lower-dose DOAC versus warfarin across the spectrum of ages (Table 2 and Figure 4). Male and female sex did not show any meaningful treatment-by-age interaction (Table 2 and Figure S2).
| Outcomes | % Change in HR per 10-year increase in age | P value | |||
|---|---|---|---|---|---|
| Primary efficacy outcome (stroke/systemic embolism) | |||||
| Standard-dose DOAC vs warfarin | 5.1% decrease (–5.1%, 14.3%) | 0.31 | |||
| Lower-dose DOAC vs warfarin | 9.1% decrease (–3.2%, 20.0%) | 0.14 | |||
| Primary safety outcome (major bleeding) | |||||
| Standard-dose DOAC vs warfarin | 10.2% increase (1.3%, 19.9%) | 0.02 | |||
| Lower-dose DOAC vs warfarin | 17.6% increase (3.4%, 33.7%) | 0.01 | |||
| All-cause death | |||||
| Standard-dose DOAC vs warfarin | 2.1% decrease (–4.7%, 8.4%) | 0.54 | |||
| Lower-dose DOAC vs warfarin | 1.5% increase (–7.0%, 10.7%) | 0.75 | |||
| Outcomes | % Change in HR per 10-year increase in age | ||||
| Male | P value | Female | P value | Interaction P value | |
| Primary efficacy outcome (stroke/systemic embolism) | |||||
| Standard-dose DOAC vs warfarin | 8.0% decrease (–4.7%, 19.1%) | 0.21 | 1.4% increase (–14.6%, 20.4%) | 0.88 | 0.38 |
| Lower-dose DOAC vs warfarin | 8.0% decrease (–8.0%, 21.7%) | 0.31 | 11.0% decrease (–9.0%, 27.3%) | 0.26 | 0.80 |
| Primary safety outcome (major bleeding) | |||||
| Standard-dose DOAC vs warfarin | 13.1% increase (2.1%, 25.3%) | 0.02 | 6.5% increase (–8.5%, 23.9%) | 0.42 | 0.52 |
| Lower-dose DOAC vs warfarin | 12.2% increase (–3.6%, 30.6%) | 0.14 | 32.8% increase (5.9%, 66.7%) | 0.01 | 0.22 |
| All-cause death | |||||
| Standard-dose DOAC vs warfarin | 0.1% decrease (–8.2%, 7.7%) | 0.98 | 6.6% decrease (–6.2%, 17.8%) | 0.30 | 0.39 |
| Lower-dose DOAC vs warfarin | 3.7% decrease (–6.7%, 13.1%) | 0.47 | 15.2% increase (–2.7, 36.4%) | 0.10 | 0.08 |
For hazard ratio (HR) point estimates, a decrease in HR signifies increasing favorability of direct oral anticoagulant (DOAC) over warfarin, whereas an increase in HR signifies decreasing favorability of DOAC over warfarin. For 95% CIs, a negative value signifies an opposite direction (decreasing/increasing) of the reported increasing/decreasing HR point estimate.
For major bleeding, a potentially meaningful interaction suggesting a greater benefit for standard-dose DOAC versus warfarin was observed in younger patients (HR for standard-dose DOAC versus warfarin increases by 10.2% [95% CI, 1.3%–19.9%] for every 10-year increase in age; P=0.02). A similar potentially meaningful interaction for major bleeding was observed for lower-dose DOAC versus warfarin in younger patients (HR for lower-dose DOAC versus warfarin increases by 17.6% [95% CI, 3.4%–33.7%] for every 10-year increase in age; P=0.01; Table 2). No meaningful interaction for major bleeding was observed after stratification by sex for either treatment strategy (Table 1 and Figure S2).
For all-cause death, no meaningful interaction was observed for standard-dose or lower-dose DOAC versus warfarin across the spectrum of age or after stratification by sex (Table 2).
Sensitivity Analyses
In the sensitivity analyses, after exclusion of patients from RE-LY and exclusion of patients from the lower-dose DOAC arm from ENGAGE AF-TIMI 48 so as to include only factor Xa inhibitors in the standard-dose DOAC treatment strategy groups, HRs for stroke or systemic embolism and major bleeding for standard-dose DOAC versus warfarin did not differ from those from the primary analyses (Figure S3). The estimated HR for major gastrointestinal bleeding with standard-dose DOAC versus warfarin was found to be lower in magnitude compared with the primary analyses. For standard-dose DOAC versus warfarin, the increase in major gastrointestinal bleeding observed in the primary analyses was no longer statistically significant (551/23 211 [2.37%] versus 436/23 189 [1.88%]; HR, 1.24 [95% CI, 0.98–1.58]). All other efficacy and safety outcomes were consistent with those from the primary analyses.
Discussion
In these results from network meta-analyses using individual patient-level data from the pivotal randomized trials of DOACs versus warfarin in patients with AF, we found that a standard-dose DOAC treatment strategy results in significant reductions in the risk of stroke or systemic embolism, intracranial bleeding, and all-cause death compared with warfarin. No statistically significant difference in major bleeding was observed, although moderate heterogeneity across trials was observed. The benefit of standard-dose DOACs over warfarin for stroke or systemic embolism was more pronounced in patients without previous VKA use and in patients with lower creatinine clearance, but was consistent across the entire range of patient age and was consistent after stratification by sex. For major bleeding, statistically significant interaction was observed suggesting a greater benefit for standard-dose DOAC over warfarin in patients with lower body weight and younger age, regardless of sex.
Results from these analyses provide the most robust evidence to date demonstrating the collective benefits of DOACs over warfarin in patients with AF. Whereas previous meta-analyses have examined the relative efficacy and safety of DOACs versus warfarin using aggregate published data,7 such results are subject to the previously noted limitations of study-level data.8 Our analyses leverage the strengths of individual patient data through consistent follow-up for time-to-event outcomes, assessment of effect modification in the Cox regression model in a consistent manner, and examining interactions across the spectrum of age as a continuous covariate.
Results from the primary analyses more clearly define the benefit of standard-dose DOACs over warfarin for reducing intracranial bleeding, one of the most feared and devastating complications of oral anticoagulants, with approximately as much benefit as the lower-dose DOAC treatment strategy. Standard-dose DOACs were associated with an increased hazard of major gastrointestinal bleeding compared with warfarin, a phenomenon that has been hypothesized to be attributable to incomplete DOAC gastrointestinal absorption and dose-dependent variation in the relative anticoagulant intensity produced at the local gastrointestinal mucosal surface by different DOACs.20,21 Gastrointestinal bleeding events in these analyses included only events that met criteria for major bleeding; thus these are clinically significant events. In the primary analyses, history of gastrointestinal bleeding was identified as an important driving factor for the increased hazard of major bleeding with standard-dose DOACs. Moderate between-trial heterogeneity for the major bleeding outcome was detected, which is consistent with previous reports based on study-level data.7 Whereas results from the primary and sensitivity analyses demonstrate a consistent pattern of increased risk for gastrointestinal bleeding with standard-dose DOACs, it is likely that drug class (direct thrombin inhibitor versus factor Xa inhibitor) and drug dose and exposure also influence an individual patient’s risk. The increased hazard of gastrointestinal bleeding with standard-dose DOACs is counterbalanced by substantial reductions in thromboembolism, intracranial bleeding, and fatal bleeding, which are of far greater consequence.
Several key findings in these analyses add to previously published data. Previous VKA use was identified as a significant effect modifier for stroke or systemic embolism. Whereas patients with and without previous VKA use were found to derive a benefit from standard-dose DOACs over warfarin for reduction in thromboembolic events, those with no previous VKA use derived a potentially greater benefit from standard-dose DOACs. Previous secondary analyses from individual randomized trials of DOACs versus warfarin with stratification by previous VKA use have yielded inconsistent results, with some trials demonstrating greater benefits of DOACs in patients without previous VKA use22 and others showing no significant interaction.23–25 The differential effect of standard-dose DOACs over warfarin based on previous VKA use is important given guideline recommendations for DOACs as first-line therapy in patients without contraindications.1,2 Data from these analyses confirm that clinicians should have no hesitancy to start DOACs in eligible patients regardless of previous VKA use.
We identified evidence of interaction favoring standard-dose DOACs over warfarin with respect to the major bleeding outcome for the subgroup of patients with low baseline body weight. Interaction testing from 3 of the 4 individual trials have shown no statistically significant interaction for major bleeding by baseline body weight,4,5,26 whereas 1 of the 4 individual trials showed findings similar to those from our meta-analyses with respect to a treatment interaction favoring DOACs in lower body weight.27 The interaction may relate to the finding that the incidence of major bleeding was higher among patients with lower body weight, which in turn is related to other factors such as older age and worse kidney function, both of which are associated with higher risk for major bleeding and tendency for greater safety with DOACs. Dedicated analyses from COMBINE AF analyzing body weight as a continuous variable are forthcoming.
We identified evidence of interaction favoring standard-dose DOACs over warfarin with respect to the stroke or systemic embolism outcome for the subgroup of patients with low baseline creatinine clearance. Previous study-level meta-analyses have similarly suggested a greater benefit of standard-dose DOACs over warfarin in patients with lower baseline creatinine clearance,28 but these analyses have been limited by the use of categorical creatinine clearance cutoffs that restrict the generalizability of the results. Dedicated analyses from COMBINE AF analyzing creatinine clearance as a continuous variable are forthcoming.
An important strength of these analyses is the ability to assess effect modification using continuous baseline variables. We demonstrate consistent benefits of standard-dose DOACs versus warfarin for stroke or systemic embolism across the continuous spectrum of age. For the major bleeding outcome, younger patients experienced a greater reduction in bleeding with standard-dose DOACs versus warfarin, perhaps because of a lower prevalence of competing comorbidities such as previous gastrointestinal bleeding or kidney dysfunction. Previous reports assessing treatment interaction by age are limited by the use of categorical data, with a typical age cut point of < or ≥75 years.7 There is generally more information in a continuous variable when assessed as such. Moreover, data derived from categorical cut points are challenging to interpret because within each category exists a wide spectrum of competing comorbidities, some of which are factors influencing DOAC dose reduction for 3 of the 4 individual trials. Although there was little or no between-trial heterogeneity detected in these analyses for the examined efficacy outcomes, moderate between-trial heterogeneity with respect to bleeding outcomes was detected, thus aggregate findings for bleeding outcomes must be interpreted with caution. One trial (RE-LY) showed an increase in major bleeding with higher-dose dabigatran (150 mg) versus warfarin in patients ≥80 years old,29 whereas the other 3 trials showed no significant interaction across age groups (with 2 of the 3 trials showing statistically significant reduction in major bleeding with DOACs).30 Heterogeneity may partly explain the findings from these analyses with respect to the bleeding interaction across continuous age. These data add depth to previous studies of DOACs in elderly patients, which are limited to subgroup analyses from individual trials,30 aggregate meta-analyses,7 and observational data.31 These findings reinforce guideline recommendations supporting preferential use of DOACs over warfarin in all age categories given consistent stroke reduction among elderly patients with at worst no difference in the incidence of major bleeding.
These analyses have several limitations. We performed univariable analyses for individual outcomes separately and tested each effect modifier individually. These analyses may therefore introduce type I error inflation attributable to multiplicity. Furthermore, because we did not use a multivariable approach in the effect modification analyses, potential multivariable collinearity between covariates (such as age and creatinine clearance and treatment effect) may not be accounted for. Patients in the COMBINE AF database are from randomized trials with specific inclusion and exclusion criteria and thus do not represent an unselected group of patients with AF in general practice. Randomization, however, offers reliable assessment of relative efficacy and safety, which cannot be reliably determined from nonrandomized observational comparisons. Although the individual trials within COMBINE AF used robust efficacy and safety outcomes, few patient-reported outcomes were collected, which may limit the application of our findings. Furthermore, data describing temporary discontinuation of study drug and study drug adherence were not included in the COMBINE AF database. Adherence in the randomized trials—for example, as measured by international normalized ratio time in therapeutic range—was at least as good as what has been described in most studies of patients in general practice. Last, aggregating trials with different study drugs and doses for meta-analyses may obscure subtle differences between outcomes that are specific to individual study drugs.
These analyses of individual patient data from the 4 pivotal trials of DOACs versus warfarin in patients with AF show that standard-dose DOACs reduce the risk of stroke or systemic embolism, death, and intracranial bleeding compared with warfarin with no significant difference in risk of major bleeding. For stroke or systemic embolism, standard-dose DOACs performed consistently better than warfarin across every examined categorical subgroup as well as across the continuous spectrum of age. For major bleeding, standard-dose DOACs performed better than warfarin in younger patients and were no different from warfarin in elderly patients, regardless of sex. These data reinforce and provide more granular detail describing the widely accepted beneficial effects of DOACs over warfarin in a broad population of patients with AF.
Article Information
Supplemental Material
Tables S1–S9
Figures S1–S3
Footnote
Nonstandard Abbreviations and Acronyms
- AF
- atrial fibrillation
- ARISTOTLE
- Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation
- COMBINE AF
- A Collaboration Between Multiple Institutions to Better Investigate Non-Vitamin K Antagonist Oral Anticoagulant Use in Atrial Fibrillation
- DOAC
- direct oral anticoagulant
- ENGAGE AF-TIMI 48
- Effective Anticoagulation With Factor Xa Next Generation in Atrial Fibrillation–Thrombolysis in Myocardial Infarction 48
- HR
- hazard ratio
- RE-LY
- Randomized Evaluation of Long-Term Anticoagulation Therapy
- ROCKET AF
- Rivaroxaban Once Daily Oral Direct Factor Xa Inhibition Compared With Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation
- VKA
- vitamin K antagonist
Supplemental Material
References
1.
January CT, Wann LS, Calkins H, Chen LY, Cigarroa JE, Cleveland JC, Ellinor PT, Ezekowitz MD, Field ME, Furie KL, et al. 2019 AHA/ACC/HRS focused update of the 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association task force on clinical practice guidelines and the Heart Rhythm Society in collaboration with the Society of Thoracic Surgeons. Circulation. 2019;140:e125–e151. doi: 10.1161/CIR.0000000000000665
2.
Hindricks G, Potpara T, Dagres N, Arbelo E, Bax JJ, Blomström-Lundqvist C, Boriani G, Castella M, Dan GA, Dilaveris PE, et al; ESC Scientific Document Group. 2020 ESC guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS): the task force for the diagnosis and management of atrial fibrillation of the European Society of Cardiology (ESC): developed with the special contribution of the European Heart Rhythm Association (EHRA) of the ESC. Eur Heart J. 2021;42:373–498. doi: 10.1093/eurheartj/ehaa612
3.
Connolly SJ, Ezekowitz MD, Yusuf S, Eikelboom J, Oldgren J, Parekh A, Pogue J, Reilly PA, Themeles E, Varrone J, et al; RE-LY Steering Committee and Investigators. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2009;361:1139–1151. doi: 10.1056/NEJMoa0905561
4.
Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, Breithardt G, Halperin JL, Hankey GJ, Piccini JP, et al; ROCKET AF Investigators. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011;365:883–891. doi: 10.1056/NEJMoa1009638
5.
Granger CB, Alexander JH, McMurray JJ, Lopes RD, Hylek EM, Hanna M, Al-Khalidi HR, Ansell J, Atar D, Avezum A, et al; ARISTOTLE Committees and Investigators. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2011;365:981–992. doi: 10.1056/NEJMoa1107039
6.
Giugliano RP, Ruff CT, Braunwald E, Murphy SA, Wiviott SD, Halperin JL, Waldo AL, Ezekowitz MD, Weitz JI, Špinar J, et al; ENGAGE AF-TIMI 48 Investigators. Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2013;369:2093–2104. doi: 10.1056/NEJMoa1310907
7.
Ruff CT, Giugliano RP, Braunwald E, Hoffman EB, Deenadayalu N, Ezekowitz MD, Camm AJ, Weitz JI, Lewis BS, Parkhomenko A, et al. Comparison of the efficacy and safety of new oral anticoagulants with warfarin in patients with atrial fibrillation: a meta-analysis of randomised trials. Lancet. 2014;383:955–962. doi: 10.1016/S0140-6736(13)62343-0
8.
Lyman GH, Kuderer NM. The strengths and limitations of meta-analyses based on aggregate data. BMC Med Res Methodol. 2005;5:14. doi: 10.1186/1471-2288-5-14
9.
Hong H, Fu H, Price KL, Carlin BP. Incorporation of individual-patient data in network meta-analysis for multiple continuous endpoints, with application to diabetes treatment. Stat Med. 2015;34:2794–2819. doi: 10.1002/sim.6519
10.
Ioannidis J. Next-generation systematic reviews: prospective meta-analysis, individual-level data, networks and umbrella reviews. Br J Sports Med. 2017;51:1456–1458. doi: 10.1136/bjsports-2017-097621
11.
Chaimani A. Conduct and reporting of individual participant data network meta-analyses need improvement. BMC Med. 2020;18:156. doi: 10.1186/s12916-020-01630-w
12.
Carnicelli AP, Hong H, Giugliano RP, Connolly SJ, Eikelboom J, Patel MR, Wallentin L, Morrow DA, Wojdyla D, Hua K, et al; COMBINE AF Investigators. Individual patient data from the pivotal randomized controlled trials of non-vitamin K antagonist oral anticoagulants in patients with atrial fibrillation (COMBINE AF): design and rationale: from the COMBINE AF (A Collaboration between Multiple institutions to Better Investigate Non-vitamin K antagonist oral anticoagulant use in Atrial Fibrillation) investigators. Am Heart J. 2021;233:48–58. doi: 10.1016/j.ahj.2020.12.002
13.
Schulman S, Kearon C; Subcommittee on Control of Anticoagulation of the Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis. Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients. J Thromb Haemost. 2005;3:692–694. doi: 10.1111/j.1538-7836.2005.01204.x
14.
Smith CT, Williamson PR, Marson AG. Investigating heterogeneity in an individual patient data meta-analysis of time to event outcomes. Stat Med. 2005;24:1307–1319. doi: 10.1002/sim.2050
15.
Schoenfeld D. Chi-squared goodness-of-fit tests for the proportional hazards regression model. Biometrika. 1980;67:145–153.
16.
Li G, Taljaard M, Van den Heuvel ER, Levine MA, Cook DJ, Wells GA, Devereaux PJ, Thabane L. An introduction to multiplicity issues in clinical trials: the what, why, when and how. Int J Epidemiol. 2017;46:746–755. doi: 10.1093/ije/dyw320
17.
Brookes ST, Whitely E, Egger M, Smith GD, Mulheran PA, Peters TJ. Subgroup analyses in randomized trials: risks of subgroup-specific analyses; power and sample size for the interaction test. J Clin Epidemiol. 2004;57:229–236. doi: 10.1016/j.jclinepi.2003.08.009
18.
Therneau, T. Coxme: mixed effects Cox models. R package version 2. 2–16. 2020.
19.
Therneau, T, Grambsch, P. Modeling Survival Data: Extending the Cox Model. Springer; 2000.
20.
Desai J, Kolb JM, Weitz JI, Aisenberg J. Gastrointestinal bleeding with the new oral anticoagulants–defining the issues and the management strategies. Thromb Haemost. 2013;110:205–212. doi: 10.1160/TH13-02-0150
21.
Vanassche T, Hirsh J, Eikelboom JW, Ginsberg JS. Organ-specific bleeding patterns of anticoagulant therapy: lessons from clinical trials. Thromb Haemost. 2014;112:918–923. doi: 10.1160/TH14-04-0346
22.
O’Donoghue ML, Ruff CT, Giugliano RP, Murphy SA, Grip LT, Mercuri MF, Rutman H, Shi M, Kania G, Cermak O, et al. Edoxaban vs. warfarin in vitamin K antagonist experienced and naive patients with atrial fibrillation. Eur Heart J. 2015;36:1470–1477. doi: 10.1093/eurheartj/ehv014
23.
Ezekowitz MD, Wallentin L, Connolly SJ, Parekh A, Chernick MR, Pogue J, Aikens TH, Yang S, Reilly PA, Lip GY, et al; RE-LY Steering Committee and Investigators. Dabigatran and warfarin in vitamin K antagonist-naive and -experienced cohorts with atrial fibrillation. Circulation. 2010;122:2246–2253. doi: 10.1161/CIRCULATIONAHA.110.973735
24.
Mahaffey KW, Wojdyla D, Hankey GJ, White HD, Nessel CC, Piccini JP, Patel MR, Berkowitz SD, Becker RC, Halperin JL, et al. Clinical outcomes with rivaroxaban in patients transitioned from vitamin K antagonist therapy: a subgroup analysis of a randomized trial. Ann Intern Med. 2013;158:861–868. doi: 10.7326/0003-4819-158-12-201306180-00003
25.
Garcia DA, Wallentin L, Lopes RD, Thomas L, Alexander JH, Hylek EM, Ansell J, Hanna M, Lanas F, Flaker G, et al. Apixaban versus warfarin in patients with atrial fibrillation according to prior warfarin use: results from the Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation trial. Am Heart J. 2013;166:549–558. doi: 10.1016/j.ahj.2013.05.016
26.
Eikelboom JW, Wallentin L, Connolly SJ, Ezekowitz M, Healey JS, Oldgren J, Yang S, Alings M, Kaatz S, Hohnloser SH, et al. Risk of bleeding with 2 doses of dabigatran compared with warfarin in older and younger patients with atrial fibrillation: an analysis of the Randomized Evaluation of Long-Term Anticoagulant Therapy (RE-LY) trial. Circulation. 2011;123:2363–2372. doi: 10.1161/CIRCULATIONAHA.110.004747
27.
Boriani G, Ruff CT, Kuder JF, Shi M, Lanz HJ, Antman EM, Braunwald E, Giugliano RP. Edoxaban versus warfarin in patients with atrial fibrillation at the extremes of body weight: an analysis from the ENGAGE AF-TIMI 48 trial. Thromb Haemost. 2021;121:140–149. doi: 10.1055/s-0040-1716540
28.
Del-Carpio Munoz F, Gharacholou SM, Munger TM, Friedman PA, Asirvatham SJ, Packer DL, Noseworthy PA. Meta-analysis of renal function on the safety and efficacy of novel oral anticoagulants for atrial fibrillation. Am J Cardiol. 2016;117:69–75. doi: 10.1016/j.amjcard.2015.09.046
29.
Lauw MN, Eikelboom JW, Coppens M, Wallentin L, Yusuf S, Ezekowitz M, Oldgren J, Nakamya J, Wang J, Connolly SJ. Effects of dabigatran according to age in atrial fibrillation. Heart. 2017;103:1015–1023. doi: 10.1136/heartjnl-2016-310358
30.
Schäfer A, Flierl U, Berliner D, Bauersachs J. Anticoagulants for stroke prevention in atrial fibrillation in elderly patients. Cardiovasc Drugs Ther. 2020;34:555–568. doi: 10.1007/s10557-020-06981-3
31.
Mitchell A, Watson MC, Welsh T, McGrogan A. Effectiveness and safety of direct oral anticoagulants versus vitamin K antagonists for people aged 75 years and over with atrial fibrillation: a systematic review and meta-analyses of observational studies. J Clin Med. 2019;8:E554. doi: 10.3390/jcm8040554
Information & Authors
Information
Published In
Copyright
© 2022 American Heart Association, Inc.
Versions
You are viewing the most recent version of this article.
History
Received: 27 June 2021
Accepted: 2 December 2021
Published online: 5 January 2022
Published in print: 25 January 2022
Keywords
Subjects
Authors
Disclosures
Disclosures Dr Carnicelli reports grants from the National Institutes of Health during the conduct of the study. Dr Connolly reports personal fees from BMS, Bayer, Boehringer Ingelheim, Daiichi Sankyo, and Portola during the conduct of the study. J. Eikelboom reports honoraria and grant support from Astra Zeneca, Bayer, Boehringer Ingelheim, Bristol Myers Squibb/Pfizer, Daiichi-Sankyo, GlaxoSmithKline, Janssen, Sanofi Aventis, and Eli Lilly as well as a personnel award from the Heart and Stroke Foundation. Dr Giugliano reports personal fees from Amarin, American College of Cardiology, AstraZeneca, Boehringer Ingelheim, Bristol Myers Squibb, CryoLife, CVS Caremark, Dr Reddy’s Laboratories, Eli Lilly and Company, Esperion, Gilead, GlaxoSmithKline, Janssen, Lexicon, Pfizer, St Lukes, SAJA Pharmaceuticals, Samsung, and Servier; grants and personal fees from Amgen, Daiichi-Sankyo, and Merck; and grants from Anthos Therapeutics outside the submitted work. Dr Morrow reports being a member of the TIMI (Thrombolysis in Myocardial Infarction) Study Group, which has received institutional research grant support through Brigham and Women’s Hospital from Abbott, Amgen, Anthos Therapeutics, AstraZeneca, Bayer HealthCare Pharmaceuticals, Inc, Daiichi-Sankyo, Eisai, Intarcia, MedImmune, Merck, Novartis, Pfizer, Quark Pharmaceuticals, Regeneron Pharmaceuticals, Inc, Roche, Siemens Healthcare Diagnostics, Inc, The Medicines Company, and Zora Biosciences. Dr Patel reports grants from Astra Zeneca, Bayer, Janssen, Procyrion, and Heartflow; and personal fees from Bayer, Janssen, Mytonomy, and Procyrion outside the submitted work. Dr Wallentin reports grants from AstraZeneca, Boehringer Ingelheim, Bristol Myers Squibb/Pfizer, GlaxoSmithKline, Merck & Co, and Roche Diagnostics and personal fees from Abbott outside the submitted work. Dr Alexander reports grants from Bayer and XaTek; grants and personal fees from Bristol Myers Squibb and CryoLife; and personal fees from Janssen, Pfizer, and Portola outside the submitted work. Dr Bahit reports other support from Pfizer, VIFOR, and CSL Behring outside the submitted work. Dr Bohula reports being a member of the TIMI Study Group, which has received institutional research grant support through Brigham and Women’s Hospital from Abbott, Amgen, Anthos Therapeutics, AstraZeneca, Bayer Healthcare Pharmaceuticals, Inc, Daiichi-Sankyo, Eisai, Intarcia, MedImmune, Merck, Novartis, Pfizer, Quark Pharmaceuticals, Regeneron Pharmaceuticals, Inc, Roche, Siemens Healthcare Diagnostics, Inc, The Medicines Company, and Zora Biosciences. L. Dyal reports funding from Boehringer Ingelheim during the conduct of the study. Dr Ezekowitz reports grants and other support from Boehringer Ingelheim; other support from Sanofi-Aventis, Boston Scientific, Alta Thera, Anthos, Biogen Idec, and Boston Scientific; grants from Pfizer, Johnson & Johnson, and Daiichi-Sankyo Pharma Development during the conduct of the study. K.A.A. Fox reports grants and personal fees from Bayer/Janssen, grants from AstraZeneca, and personal fees from Verseon outside the submitted work. Dr Halperin reports personal fees from Boehringer Ingelheim, Bayer Healthcare, Ortho-McNeil-Janssen, Pfizer, Bristol Myers Squibb, and Daiichi-Sankyo during the conduct of the study and personal fees from Boehringer Ingelheim, Ortho-McNeil Janssen, the ATLAS Group, Duke Clinical Research Institute, and the TIMI Group outside the submitted work. Dr Hijazi reports consulting and lecture fees from Boehringer Ingelheim and Pfizer/BMS and grants from The Swedish Society for Medical Research (grant S17-0133) and The Swedish Heart-Lung Foundation (grant 20200722) during the conduct of the study and fees paid to his institution for advisory boards and lectures from Roche Diagnostics outside the submitted work. Dr Hohnloser reports personal fees from Boehringer Ingelheim, BMS, Pfizer, Daiichi-Sankyo, Bayer Healthcare, Medtronic, Sanofi, and zoll outside the submitted work. Dr Hylek reports consulting fees from Anthos Therapeutics, Bristol Myers Squibb/Pfizer, Janssen, and Medtronic; honoraria from Boehringer Ingelheim and Bristol Myers Squibb/Pfizer; and advisory board fees from Anthos Therapeutics and CryoLife outside of the submitted work. Dr Kato reports personal fees from Daiichi-Sankyo, AstraZeneca, Bristol Myers Squibb, MSD KK, Pfizer, Tanabe-Mitsubishi, Bayer, Boehringer Ingelheim, Amgen, and Takeda; grants and personal fees from Ono Pharmaceutical; and grants from Abbott Japan during the conduct of the study. J. Kuder reports grants from Daiichi-Sankyo during the conduct of the study and is a member of the TIMI Study Group, which has received institutional research grant support through Brigham and Women’s Hospital from Abbott, Amgen, Anthos Therapeutics, AstraZeneca, Bayer Healthcare Pharmaceuticals, Inc, Daiichi-Sankyo, Eisai, Intarcia, MedImmune, Merck, Novartis, Pfizer, Quark Pharmaceuticals, Regeneron Pharmaceuticals, Inc, Roche, Siemens Healthcare Diagnostics, Inc, The Medicines Company, and Zora Biosciences. Dr Lopes reports personal fees from Bayer, Boehringer Ingelheim, Bristol Myers Squibb, Daiichi-Sankyo, GlaxoSmithKline, Medtronic, Merck, Pfizer, Portola, and Sanofi; and grants from Bristol Myers Squibb, GlaxoSmithKline, Medtronic, Pfizer, and Sanofi outside the submitted work. Dr Mahaffey reports grants from Afferent, the American Heart Association, Apple, Inc, Cardiva Medical, Inc, Eidos, Ferring, Gilead, Google (Verily), Luitpold, Medtronic, Merck, Sanifit, and St Jude; grants and personal fees from Amgen, AstraZeneca, Bayer, Johnson & Johnson, Novartis, and Sanofi; and personal fees from Anthos, Applied Therapeutics, CSL Behring, Elsevier, Inova, Intermountain Health, Medscape, Mount Sinai, Mundi Pharma, Myokardia, Novo Nordisk, Otsuka, Portola, SmartMedics, and Theravance outside the submitted work. Dr Oldgren reports fees to his institution for consultant/advisory boards (including study steering committees and data safety monitoring boards) and lectures from Alexion, AstraZeneca, Bayer, BMS, Boehringer Ingelheim, Daiichi-Sankyo, Janssen, Novartis, Pfizer, Roche Diagnostics, and Sanofi outside the submitted work. Dr Piccini reports grants from Johnson & Johnson and Bayer during the conduct of the study; grants and personal fees from Boston Scientific and Abbott; and personal fees from BMS outside the submitted work. Dr Ruff reports grants from Daiichi-Sankyo during the conduct of the study; grants and personal fees from Anthos and Boehringer Ingelheim; grants from Daiichi-Sankyo, AstraZeneca, and the National Institutes of Health; personal fees from Bayer, Bristol Myers Squibb, Janssen, Pfizer, and Portola outside the submitted work; and is a member of the TIMI Study Group, which has received institutional research grant support through Brigham and Women’s Hospital from Abbott, Amgen, Anthos Therapeutics, AstraZeneca, Bayer Healthcare Pharmaceuticals, Inc, Daiichi-Sankyo, Eisai, Intarcia, MedImmune, Merck, Novartis, Pfizer, Quark Pharmaceuticals, Regeneron Pharmaceuticals, Inc, Roche, Siemens Healthcare Diagnostics, Inc, The Medicines Company, and Zora Biosciences. Dr Steffel reports personal fees from Amgen, Astra Zeneca, Boehringer Ingelheim, Bristol Myers Squibb, Novartis, Pfizer, Portola/Alexion, Medscape, WebMD, Merck/MSD, Berlin Chemie/Menarini, Roche Diagnostics, SAJA Pharmaceuticals, and Servier; grants and personal fees from Bayer Healthcare, Biosense Webster, Boston Scientific, Daiichi-Sankyo, Medtronic, Abbott, and Biotronik; and other support from CorXL outside the submitted work. Dr Granger reports personal fees from Bayer and Boston Scientific; grants and personal fees from Boehringer Ingelheim, Bristol Myers Squibb, Janssen, and Pfizer; and grants from Daiichi-Sankyo during the conduct of the study; personal fees from AbbVie, Espero, Medscape, Medtronic Inc, Merck, the National Institutes of Health, Novo Nordisk, Roche, Rho Pharmaceuticals, CeleCor, Correvio, Philips, Abiomed, and Anthos Therapeutics; grants from Akros, AstraZeneca, the US Food and Drug Administration, Glaxo Smith Kline, Medtronic Foundation, and Apple; and grants and personal fees from Novartis and The Medicines Company outside the submitted work. The other authors report no conflicts.
Sources of Funding
RE-LY (Randomized Evaluation of Long-Term Anticoagulation Therapy) was funded by Boehringer Ingelheim. ROCKET AF (Rivaroxaban Once Daily Oral Direct Factor Xa Inhibition Compared With Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation) was funded by Johnson & Johnson and Bayer. ARISTOTLE (Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation) was funded by Bristol Myers Squibb and Pfizer. ENGAGE AF-TIMI 48 (Effective Anticoagulation With Factor Xa Next Generation in Atrial Fibrillation–Thrombolysis in Myocardial Infarction 48) was funded by Daiichi-Sankyo. No outside funding was obtained to support the creation of the COMBINE AF (A Collaboration Between Multiple Institutions to Better Investigate Non-Vitamin K Antagonist Oral Anticoagulant Use in Atrial Fibrillation) database or for these analyses.
Metrics & Citations
Metrics
Citations
Download Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Select your manager software from the list below and click Download.
- Anticoagulation use in rheumatic heart disease-associated atrial fibrillation: a study of real-world practice in Central Australia, IJC Heart & Vasculature, 59, (101711), (2025).https://doi.org/10.1016/j.ijcha.2025.101711
- Association of Race and Ethnicity With Stroke and Mortality Outcomes in Atrial Fibrillation, JACC: Advances, 4, 7, (101860), (2025).https://doi.org/10.1016/j.jacadv.2025.101860
- Review of the top 5 cardiology studies of 2023–24, Canadian Pharmacists Journal / Revue des Pharmaciens du Canada, (2025).https://doi.org/10.1177/17151635251339445
- Stroke prevention in atrial fibrillation: A narrative review of current evidence and emerging strategies, European Journal of Clinical Investigation, (2025).https://doi.org/10.1111/eci.70082
- Bleeding Risk in East Asian Patients with Atrial Fibrillation: Validation of the DOAC Score in a Japanese Multicenter Registry, Heart Rhythm O2, (2025).https://doi.org/10.1016/j.hroo.2025.06.003
- Real-world management, resource use, patient-reported outcomes and adherence in patients receiving direct oral anticoagulants for first stroke attributed to non-valvular atrial fibrillation in secondary care: A UK mixed-methods observational study, PLOS One, 20, 5, (e0321621), (2025).https://doi.org/10.1371/journal.pone.0321621
- Continued Versus Interrupted Oral Anticoagulation During Transcatheter Aortic Valve Replacement in Patients With Atrial Fibrillation: A Meta-Analysis, American Journal of Therapeutics, (2025).https://doi.org/10.1097/MJT.0000000000001979
- Relevance of selected pharmacogenetic polymorphisms to bleeding and thromboembolic risks in Chinese patients taking direct‐acting oral anticoagulants, British Journal of Clinical Pharmacology, (2025).https://doi.org/10.1002/bcp.70078
- Direct oral anticoagulants in stroke prevention of atrial fibrillation patients: can we rely on them in all clinical scenarios?, Future Cardiology, (1-3), (2025).https://doi.org/10.1080/14796678.2025.2506915
- Comparative analysis of therapeutic strategies in atrial fibrillation patients with left atrial appendage thrombus despite optimal NOAC therapy, Clinical Research in Cardiology, (2025).https://doi.org/10.1007/s00392-025-02665-w
- See more
Loading...
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginPurchase Options
Purchase this article to access the full text.
eLetters(0)
eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.
Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.