The nodes represent the antithrombotic treatment regimens that were compared, and the edges represent the observed direct comparisons in the included randomized clinical trials. The size of nodes is proportional to the number of patients assigned to the treatment regimen and the thickness of edges is proportional to the sample size of each study. AF indicates atrial fibrillation; DAPT, dual antiplatelet therapy; NOAC, non-VKA oral anticoagulant; PCI, percutaneous coronary intervention; VKA, vitamin K antagonists.
The safety outcomes assessed were thrombolysis in myocardial infarction (TIMI) major bleeding, TIMI major or minor bleeding, trial-defined primary safety outcome, and intracranial hemorrhage. A total of 11 430 patients were included in the network meta-analyses for all safety outcomes. The efficacy outcomes assessed were all-cause death, myocardial infarction, stroke, and stent thrombosis. A total of 11 501 patients were included in the network meta-analyses for all efficacy outcomes. Odds ratios and 95% credible intervals compared with vitamin K antagonist plus dual antiplatelet therapy (VKA+DAPT) (reference) were plotted for all outcomes. NOAC indicates non-VKA oral anticoagulant.
eTable 1. Definitions of Trial-defined Primary Bleeding and MACE Outcomes in Each Study
eTable 2. Search Code (PubMed/MEDLINE Search)
eTable 3. Risk of Bias of Included Trials Using the Cochrane Risk Assessment Tool
eTable 4. Characteristics of Included Trials
eTable 5. Treatment Strategies in Each Study
eTable 6. Updated Network Meta-Analysis Data: Sample Size and the Number of Participants Who Had Each Outcome in Each Study
eTable 7. SUCRA Values for Each Treatment Regimen and Outcomes
eFigure 1. Forest Plots for Efficacy Outcomes: (A) Trial-defined MACE, (B) Cardiovascular Death
eFigure 2. Odds Ratios for TIMI Major Bleeding and MACE
Customize your JAMA Network experience by selecting one or more topics from the list below.
Lopes RD, Hong H, Harskamp RE, et al. Optimal Antithrombotic Regimens for Patients With Atrial Fibrillation Undergoing Percutaneous Coronary Intervention: An Updated Network Meta-analysis. JAMA Cardiol. 2020;5(5):582–589. doi:10.1001/jamacardio.2019.6175
What is the optimal antithrombotic regimen in terms of major bleeding and ischemic risk for patients with atrial fibrillation undergoing percutaneous coronary intervention?
This network meta-analysis of 5 randomized controlled trials found that the use of a combination of a non-vitamin K antagonist oral anticoagulant and a P2Y12 inhibitor (discontinuing the aspirin regimen a few days after percutaneous coronary intervention) reduced bleeding complications, including intracranial bleeding, whereas the combination of a vitamin K antagonist and dual antiplatelet therapy resulted in the highest rates of bleeding. The risk of ischemic events was comparable among the 4 tested regimens.
The findings of this study may provide a rigorous and up-to-date evaluation of the safety and efficacy of available antithrombotic strategies to aid health care professionals in making informed treatment decisions.
Antithrombotic treatment in patients with atrial fibrillation (AF) and percutaneous coronary intervention (PCI) presents a balancing act with regard to bleeding and ischemic risks.
To evaluate the safety and efficacy of 4 antithrombotic regimens by conducting an up-to-date network meta-analysis and to identify the optimal treatment for patients with AF undergoing PCI.
Online computerized database (MEDLINE).
Five randomized studies were included (N = 11 542; WOEST, PIONEER AF-PCI, RE-DUAL PCI, AUGUSTUS, ENTRUST-AF PCI).
Data Extraction and Synthesis
The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were used in this network meta-analysis, in which bayesian random-effects models were applied. The data were analyzed from September 9 to 29, 2019.
Main Outcomes and Measures
The primary safety outcome was thrombolysis in myocardial infarction (TIMI) major bleeding and the primary efficacy outcome was trial-defined major adverse cardiovascular events (MACE).
The total number of participants included in the study was 11 532. The mean age of the participants ranged from 70 to 72 years, 69% to 83% were male, 20% to 26% were female, and the participants were predominantly white (>90%). Compared with vitamin K antagonists (VKA) plus dual antiplatelet therapy (DAPT) (reference), the odds ratios (ORs) (95% credible intervals) for TIMI major bleeding were 0.57 (0.31-1.00) for VKA plus P2Y12 inhibitor, 0.69 (0.40-1.16) for non-VKA oral anticoagulant (NOAC) plus DAPT, and 0.52 (0.35-0.79) for NOAC plus P2Y12 inhibitor. For MACE, using VKA plus DAPT as reference, the ORs (95% credible intervals) were 0.97 (0.64-1.42) for VKA plus P2Y12 inhibitor, 0.95 (0.64-1.39) for NOAC plus DAPT, and 1.03 (0.77-1.38) for NOAC plus P2Y12 inhibitor.
Conclusions and Relevance
The findings of this study suggest that an antithrombotic regimen of VKA plus DAPT should generally be avoided, because regimens in which aspirin is discontinued may lead to lower bleeding risk and no difference in antithrombotic effectiveness. The use of a NOAC plus a P2Y12 inhibitor without aspirin may be the most favorable treatment option and the preferred antithrombotic regimen for most patients with AF undergoing PCI.
Identifying an optimal antithrombotic regimen to prevent bleeding and ischemic events presents an unmet challenge to physicians treating patients with atrial fibrillation (AF) who require antiplatelet therapy for percutaneous coronary intervention (PCI) and/or acute coronary syndrome (ACS).1,2 Previous studies have compared various antithrombotic regimens.3-7 A 2019 network meta-analysis found that a regimen of non-vitamin K antagonist oral anticoagulants (NOACs) plus a P2Y12 inhibitor without aspirin was associated with lower rates of bleeding, including intracranial hemorrhage, compared with a regimen of vitamin K antagonists (VKA) plus dual antiplatelet therapy (DAPT).7 Since that publication, another randomized clinical trial (RCT) was completed, ENTRUST-AF PCI, which compared the safety of the use of edoxaban (trade names, Savaysa and Lixiana) plus a P2Y12 inhibitor with VKA plus DAPT in 1506 patients with AF who underwent PCI.8 Given the relative importance of this study, we have updated the previous literature search and network meta-analysis to provide readers with a current, state-of-the-art evidence base on antithrombotic regimens in this high-risk patient population.
A full description of the methodology was previously published.7 In short, 2 of us (R.D.L. and R.E.H.) performed an updated literature review using the PubMed search engine and searched for the following: (1) RCTs with 2 or more comparator arms, (2) in patients with ACS and/or PCI, (3) a combination of anticoagulation and antiplatelet therapy, and (4) reported major bleeding and major adverse cardiovascular events (MACE) with a follow-up of 6 or more months. The study was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guideline.
The primary safety outcome was major bleeding according to the thrombolysis in myocardial infarction (TIMI) criteria.9 The primary efficacy outcome was trial-defined MACE, which was usually defined as a combination of either all-cause or cardiovascular mortality, myocardial infarction (MI), stroke, and stent thrombosis (eTable 1 in the Supplement). The RCTs, WOEST,3 AUGUSTUS,6 and ENTRUST-AF PCI8 reported the number of participants who had definite or probable stent thrombosis. The RCT, PIONEER AF-PCI,5 reported the number of participants who had stent thrombosis without specifying its category, and RE-DUAL PCI4 reported the number of participants who had definite stent thrombosis. Secondary efficacy outcomes were the individual components of this composite MACE outcome.
Two of us (H.H. and J.L.) independently extracted data on the study design, baseline characteristics, interventions, and outcomes.
The data were analyzed from September 9 to 29, 2019. We fitted a bayesian random-effects network meta-analysis model to simultaneously compare multiple regimens. We estimated odds ratios (ORs) of the treatment effects of the 2 regimens and the associated 95% credible intervals (CrIs) using Markov chain Monte Carlo algorithms. To evaluate and rank regimens, we calculated rank probabilities (ie, probability of a regimen being the best, second-best, or worst for an outcome) and the Surface under the Cumulative Ranking (SUCRA). All analyses were conducted using the gemtc package (version 0.8-2) in R, version 3.6.1 (The R Foundation).
In an updated literature search, we found 38 unique studies that were published between April 19, 2019 (date of last search update) and September 14, 2019 (eTable 2 in the Supplement). Of them, 1 study (ENTRUST-AF PCI) was assessed in full-text and was found eligible.8 These data were combined with those of the 4 RCTs previously identified (WOEST, PIONEER AF-PCI, RE-DUAL PCI, and AUGUSTUS).3-6
Baseline characteristics of patients in each RCT are provided in the Table. A total of 11 542 patients were included in the network meta-analysis. The mean age of the participants ranged from 70 to 72 years, 69% to 83% were male, 20% to 26% were women, and the participants were predominantly white (>90%). The prevalence of ACS ranged from 25% to 28% and, except for AUGUSTUS, all patients underwent PCI. Most patients were at high risk for thromboembolic and bleeding complications (>3% per annum). Risk of bias assessment, characteristics of the trial design, treatment regimens, and main results are provided in the Supplement (eTables 3-6 in the Supplement).
We simultaneously compared the following 4 treatment regimens: VKA plus DAPT (reference), VKA plus P2Y12 inhibitor, NOAC plus DAPT, and NOAC plus P2Y12 inhibitor (Figure 1). We assumed a class effect, that is, all 4 NOAC agents and respective doses had comparable safety and efficacy.
Compared with the ORs for VKA plus DAPT, those for all safety outcomes, including intracranial hemorrhage, were significantly lower for NOAC plus P2Y12 inhibitor (Figure 2A-D). Compared with VKA plus DAPT (reference), the ORs for TIMI major bleeding were 0.57 (95% CrI, 0.31-1.00) for VKA plus P2Y12 inhibitor, 0.69 (95% CrI, 0.40-1.16) for NOAC plus DAPT, and 0.52 (95% CrI, 0.35-0.79) for NOAC plus P2Y12 inhibitor. For MACE, compared with VKA plus DAPT the ORs were 0.97 (95% CrI, 0.64-1.42) for VKA plus P2Y12 inhibitor, 0.95 (95% CrI, 0.64-1.39) for NOAC plus DAPT, and 1.03 (95% CrI, 0.77-1.38) for NOAC plus P2Y12 inhibitor. Discontinuing the aspirin regimen (either with NOAC or VKA) was associated with a lower risk of trial-defined bleeding compared with regimens including aspirin. Compared with VKA plus DAPT, the ORs for trial-defined bleeding were 0.46 (95% CrI, 0.22-0.95) for VKA plus P2Y12 inhibitor and 0.53 (95% CrI, 0.31-0.90) for NOAC plus P2Y12 inhibitor.
No differences were found among the antithrombotic regimens in terms of the composite of MACE as well as its individual components of (cause-specific) death, MI, stroke, or stent thrombosis (Figure 2E-H and eFigure 1 in the Supplement).
The SUCRA values for safety and efficacy outcomes are presented in eTable 7 in the Supplement. The performance of the tested regimens is shown in a forest plot of ORs (eFigure 2 in the Supplement). Regimens in which aspirin was omitted had the best performance (ie, highest SUCRA value) for treating bleeding complications. The combination of NOAC plus P2Y12 inhibitor was the best regimen for treating major bleeding (SUCRA value, 81.9), and any regimen with NOAC (plus DAPT [SUCRA value, 67.3] or plus P2Y12 inhibitor [SUCRA value, 91.8]) was preferred to using VKA (plus DAPT [SUCRA value, 28.8] or plus P2Y12 inhibitor [SUCRA value, 12.9]) for treating intracranial hemorrhage. No treatment regimen was clearly favored overall for efficacy outcomes.
In this updated, comprehensive network meta-analysis an antithrombotic regimen in which aspirin is discontinued a few days after PCI appears to be associated with fewer bleeding complications while preserving antithrombotic efficacy. The findings of the present study suggest that a regimen of NOAC plus P2Y12 inhibitor without aspirin had the best safety profile, with the lowest rates of intracranial bleeding and similar rates of ischemic events compared with other antithrombotic regimens that included persistent use of aspirin.
Most guideline recommendations in cardiology are based on low-quality evidence, and the field of antithrombotic therapy for AF and ACS and PCI is no exception.10,11 The recommendation for traditional antithrombotic triple therapy mostly relied on extrapolation and findings from observational studies.1,2,12-15 Remarkable advances have been made over the past few years, starting with the initial findings from WOEST followed by PIONEER AF-PCI, RE-DUAL PCI, AUGUSTUS, and ENTRUST-AF PCI.3-6,8 Combined, these studies encompass high-quality data from more than 11 000 patients that allow for meaningful observations with regard to bleeding and ischemic outcomes.
A question that remains is whether this network meta-analysis provides definitive answers for infrequent outcomes, such as stent thrombosis. Although we did not observe a statistically significant difference in the rates of stent thrombosis between regimens with and without aspirin, the numerical excess of stent thrombosis in patients when aspirin therapy was discontinued may be important, particularly for patients at high risk of stent thrombosis and those in whom the consequences of this condition would be severe. Clear guidance on how to identify such patients based on the available evidence is lacking, and we do not foresee the data to do so becoming available in the future.11 Although the use of a network meta-analysis allows for simultaneous comparisons and evidence-based grading to facilitate overall conclusions, we believe that it does not have the granularity to address specific subgroups of patients. Future studies with individual patient-level data analyses, whether trial-specific or pooled, may help to further refine which patients would benefit most from longer-term aspirin use in combination with a NOAC and a P2Y12 inhibitor.
Selecting the optimal antithrombotic regimen for patients with AF undergoing PCI presents an important unmet clinical need. We believe that the findings of this study support the use of regimens in which aspirin therapy is discontinued a few days after PCI. A regimen that includes a NOAC plus a P2Y12 inhibitor seems to be the most favorable treatment option and may be the preferred antithrombotic regimen for most of these patients.
Accepted for Publication: December 12, 2019.
Corresponding Author: Renato D. Lopes, MD, PhD, Division of Cardiology, Duke Clinical Research Institute, 200 Morris St, Durham, NC 27701 (firstname.lastname@example.org).
Published Online: February 26, 2020. doi:10.1001/jamacardio.2019.6175
Author Contributions: Drs Lopes and Hong had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Lopes, Hong, Harskamp, Mehran, Cannon, Granger, Verheugt, Goette, Gibson, Alexander.
Acquisition, analysis, or interpretation of data: Lopes, Hong, Harskamp, Bhatt, Mehran, Cannon, Li, ten Berg, Sarafoff, Vranckx, Goette, Gibson.
Drafting of the manuscript: Lopes, Hong, Harskamp, Mehran, Verheugt, Goette, Gibson.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Hong, Mehran, Cannon, Li, Goette.
Obtained funding: Lopes, Goette.
Administrative, technical, or material support: Lopes, Mehran, Cannon, Sarafoff, Goette, Alexander.
Supervision: Lopes, Hong, Mehran, Cannon, Verheugt, ten Berg, Vranckx, Goette, Gibson, Alexander.
Conflict of Interest Disclosures: Dr Lopes reported receiving consulting fees from Bayer, Boehringer Ingelheim, Daiichi Sankyo, Merck, and Portola Pharmaceuticals; and grants and consulting fees from Bristol-Myers Squibb, GlaxoSmithKline, Medtronic, Pfizer, and Sanofi outside the submitted work. Dr Harskamp reported receiving grants from the Dutch Research Council outside the submitted work. Dr Bhatt reported receiving grants from Amarin Corporation, AstraZeneca, Bristol-Myers Squibb, Eisai, Ethicon, Medtronic, Sanofi, The Medicines Company, PLx Pharma, Roche, Pfizer, Forest Laboratories/AstraZeneca, Ischemix, Amgen, Eli Lilly and Company, Chiesi, Ironwood Pharmaceuticals, Abbott Laboratories, Regeneron Pharmaceuticals, Idorsia, Synaptic Pharma Ltd, Fractyl, and Afimmune; consulting fees from FlowCo Inc, Takeda, Medscape Cardiology, Regado Biosciences, Boston VA Research Institute, Clinical Cardiology, Veterans Affairs, St. Jude Medical (now Abbott), Biotronik, Cardax, Boston Scientific, Merck, Svelte Pharma, Novo Nordisk, Cereno Scientific, and Cardiovascular Systems Inc; personal fees from Harvard Clinical Research Institute (now Baim Institute for Clinical Research), Duke Clinical Research Institute, Mayo Clinic, Population Health Research Institute, Belvoir Publications, Slack Publications, WebMD, Elsevier, HMP Global, Journal of the American College of Cardiology, Cleveland Clinic, Mount Sinai School of Medicine, TobeSoft, Bayer, Medtelligence/ReachMD, CSL Behring, and Ferring Pharmaceuticals; personal fees, travel reimbursement, and consulting fees from the American College of Cardiology; personal fees and nonfinancial support from the Society of Cardiovascular Patient Care; nonfinancial support from the American Heart Association; grants and consulting fees from PhaseBio; and personal fees and consulting fees from Boehringer Ingelheim outside the submitted work. Dr Mehran reported receiving grants from Abbott Laboratories, AstraZeneca, Bayer, Beth Israel Deaconess, Bristol-Myers Squibb, CSL Behring, Daiichi Sankyo Inc, Medtronic, Novartis Pharmaceuticals, and OrbusNeich; personal fees from Abbott Laboratories, Boston Scientific, Medscape/WebMD, Siemens Medical Solutions, PLx Opco Inc/dba PLx Pharma Inc, Roivant Sciences, Sanofi, Medtelligence (Janssen Scientific Affairs), Janssen Scientific Affairs, American College of Cardiology, and American Medical Association; consulting fees from Abbott Laboratories, Abiomed, The Medicines Company, Spectranetics/Philips/Volcano Corp, Bristol-Myers Squibb, Watermark Research Partners, Claret Medical, and Elixir Medical; and nonfinancial support and consulting fees from Regeneron Pharmaceuticals outside the submitted work. Dr Cannon reported receiving grants and personal fees from Boehringer Ingelheim, Bristol-Myers Squibb, Daiichi Sankyo, Janssen, and Pfizer during the conduct of the study; grants from Amgen and Merck; and personal fees from Aegerion, Alnylam, Amarin Corporation, Amgen, Applied Therapeutics, Ascendia, Corvidia, HLS Therapeutics, Innovent, Kowa, and Sanofi outside the submitted work. Dr Granger reported receiving 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, Merck, National Institutes of Health, Novo Nordisk, Roche, Eli Lilly and Company, Gilead Sciences, Hoffmann-La Roche, Sirtex, Verseon, Rho Pharmaceuticals, and CeleCor Therapeutics; grants from Akros, US Food and Drug Administration, GlaxoSmithKline, Medtronic Foundation, and Apple; and grants and personal fees from AstraZeneca, Novartis, Armetheon, and The Medicines Company outside the submitted work. Dr Verheugt reported receiving personal fees from Bayer HealthCare, Bristol-Myers Squibb/Pfizer, Boehringer Ingelheim, and Daiichi-Sankyo during the conduct of the study. Dr Sarafoff reported receiving personal fees from Boehringer Ingelheim, and nonfinancial support from Bayer and Pfizer outside the submitted work. Dr Vranckx reported receiving personal fees from Daiichi Sankyo during the conduct of the study; and personal fees from Bayer, Behring CLS, and AstraZeneca outside the submitted work. Dr Goette reported receiving personal fees from Daiichi Sankyo, Boston Scientific, Boehringer Ingelheim, Bayer, AstraZeneca, Medtronic, Omeicos, and Bristol-Myers Squibb/Pfizer outside the submitted work. Dr Gibson reported receiving grants and personal fees from Bayer, Janssen Pharmaceuticals, Johnson & Johnson Corporation, and Portola Pharmaceuticals; and grants from Bristol-Myers Squibb during the conduct of the study; grants and personal fees from Angel Medical Corporation and CSL Behring; personal fees from The Medicines Company, Boston Clinical Research Institute, Cardiovascular Research Foundation, Eli Lilly and Company, Gilead Sciences, Novo Nordisk, WebMD, UpToDate in Cardiovascular Medicine, Amarin Corporation, Amgen, Boehringer Ingelheim, Chiesi, Merck, PharmaMar, Sanofi, Somahlution, St. Francis Hospital, Verreseon Corporation, Boston Scientific, Duke Clinical Research Institute, Impact Bio, Ltd, MedImmune, Medtelligence, Microport, and PERT Consortium; consulting fees from nference; nonfinancial support from Baim Institute; grants from SCAD Alliance; and personal fees from GE Healthcare, Caladrius Bioscience, CeleCor Therapeutics, and Thrombolytic Science outside the submitted work. Dr Alexander reported receiving grants from AstraZeneca, Boehringer Ingelheim, CryoLife, US Food & Drug Administration, National Institutes of Health, Sanofi, and Volumetrix; and personal fees from AbbVie; grants and personal fees from Bristol-Myers Squibb, CSL Behring, Novo Nordisk, Pfizer, Portola Pharmaceuticals, Quantum Genomics, Teikoku Pharmaceuticals, VA Cooperative Studies Program, and Zafgen outside the submitted work. No other disclosures were reported.
Funding/Support: This study was supported by Duke Clinical Research Institute (DCRI).
Role of the Funder/Sponsor: The authors from the DCRI designed and conducted the study; collected, managed, analyzed, and interpreted the data; and prepared, reviewed, and approved the manuscript; and made the decision to submit the manuscript for publication.
Disclaimer: Dr Mehran is Associate Editor of JAMA Cardiology, but she was not involved in any of the decisions regarding review of the manuscript or its acceptance.