Vitamin K antagonist vs direct oral anticoagulants with antiplatelet therapy in dual or triple therapy after percutaneous coronary intervention or acute coronary syndrome in atrial fibrillation: Meta‐analysis of randomized controlled trials

Abstract Background The combination of vitamin K antagonists (VKA) for atrial fibrillation (AF) and antiplatelet agents following percutaneous coronary intervention (PCI) is associated with an increased bleeding risk. Hypothesis Direct oral anticoagulants (DOAC) are associated with a greater safety profile but the optimal antithrombotic treatment strategy, especially when considering ischemic events, is unclear. Methods We performed a meta‐analysis of randomized controlled trials comparing outcomes in AF patients following PCI and/or acute coronary syndrome (ACS) when treated with DOAC vs VKA, both in combination with one (dual) or two (triple) antiplatelet regimens. A systematic review was performed by searches of electronic databases MEDLINE (source PubMed) and the Cochrane Controlled Clinical Trials Register Database as well as Cardiology annual meetings. Three studies were finally included. Results Compared to VKA triple therapy, the use of DOAC was associated with a decreased risk of any bleeding (relative risk [RR] 0.68 [0.62; 0.74]), major bleeding (RR 0.61 [0.51; 0.75]) and intracranial bleeding (RR 0.33 [0.17; 0.66]) and similar rates of the composite efficacy endpoint (RR 1.0 [0.87; 1.14]) and its components. Similar and consistent results were observed with both dual and triple therapy including a DOAC compared to VKA. Conclusion Our meta‐analysis supports the use of dual therapy combining a DOAC and clopidogrel as the default regimen in most AF patients after PCI and/or ACS.


| INTRODUCTION
Atrial fibrillation (AF) has been reported in 3% to 12% of patients undergoing percutaneous coronary intervention (PCI). [1][2][3] The combination of vitamin K antagonists (VKA) with aspirin and clopidogrelcalled "triple therapy"-was associated with an increased risk of major bleeding up to 2.2% within the first month and 12% at 1 year in acute coronary syndromes (ACS) patients. 4,5 An alternative to triple therapy is the use of dual therapy associating VKA and clopidogrel alone assessed in the setting of all-coming PCI patients who required anticoagulation. 6 Dual therapy was associated with a significantly lower rate of bleeding at 1 year. On the other hand, direct oral anticoagulants (DOAC) were reported to be similarly efficient as warfarin in the prevention of stroke in patients with AF and associated with lower rates of major or intracranial bleeding. [7][8][9][10] The greater safety profile of DOAC associated with antiplatelet agents over triple therapy with VKA was confirmed in two recent randomized controlled trials (RCT) in patients with AF treated with PCI. 11,12 Neither trial was designed to assess whether greater safety profile was due to the use of the DOAC or to the removal of aspirin therapy. Even if rates of major adverse cardiac events (MACE) were not different between study arms, both studies were underpowered for the comparison of DOAC and VKA with respect to ischemic outcomes. Some clinicians still fear to stop aspirin after PCI and/or ACS in patient treated with DOAC. The optimal antithrombotic treatment strategy especially when considering ischemic events remains unclear.
We conducted a systematic review and meta-analysis of RCTs to compare the safety and efficacy of DOAC vs VKA use either in a dual or triple combination therapy in AF patients requiring such combination in the setting of PCI and/or ACS.

| Study selection
We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines for the systematic review and metaanalysis (see PRISMA checklist in Appendix S1). We conducted a systematic literature review by formal searches of the electronic databases

MEDLINE (source PubMed) and the Cochrane Controlled Clinical Trials
Register Database as well as European Society of Cardiology, American Heart Association, and American College of Cardiology annual meetings through March 18, 2018. The following search terms were used: "vitamin K antagonist" OR "VKA" OR "warfarin" OR "dabigatran" OR "rivaroxaban" OR "apixaban" OR "edoxaban" OR "novel oral anticoagulant" OR "new oral anticoagulant" OR "NOAC" OR "direct oral anticoagulant" OR "DOAC" AND "percutaneous coronary intervention" OR "PCI" OR "coronary stent implantation" OR "acute coronary syndrome" AND "atrial fibrillation" OR "AF" AND "randomized controlled trial" OR "randomized" OR "trials." References from reviews and selected articles were also reviewed for potential relevant citations. Studies were searched for and evaluated by two independent reviewers (VR and PA).
We restricted our analysis to the trials that met all of the following inclusion criteria: (a) randomized controlled comparison between DOAC and VKA, (b) in patients with AF following PCI and/or ACS, (c) available outcomes involving main adverse cardiac and cerebrovascular events (MACCEs) and major bleeding, and (d) English-language publications.

| Outcome definition
Any bleeding ranging from minor to severe and major bleeding were the safety endpoints. Major bleeding complications were reported as defined in each study. The efficacy endpoints were all-causes death, cardiovascular death, myocardial infarction, stent thrombosis, stroke, and the composite efficacy endpoint (MACCEs) as defined in each study. Outcomes were based on the longest follow-up available for each study.

| Statistical analysis
The total numbers of patients experiencing or not the outcomes of interest in each arm extracted directly from the publications were used for the analyses. Results are presented as relative risks (RR) with 95% confidence intervals (95% CI). Outcomes from individual studies were combined using the Mantel-Haenzel fixed and random-effect models. Heterogeneity across studies was studied by the Cochran's Q statistic with a P-value set at .1. The I 2 was also taken into account regardless of the P-value. An I 2 of ≥50% was pre-specified as the threshold considered too high to provide consistent analysis. The random-effect model was considered for the analysis. Tests were two-tailed and a P-value of <.05 was considered statistically significant. Funnel plots were used to assess publication bias. As AUGUS-TUS included patients with ACS but no PCI, we did a complementary analysis after excluding such patients. We also analyzed different subgroups of DOAC patients (dual or triple therapy, low or high dose) in comparison with VKA regimens. In an exploratory analysis, we also compared data between subgroups of DOAC: dual therapy vs triple therapy, low dose vs high dose in studies providing such data. R software version 3.0.0 (April 3, 2013) for MacOS (R Foundation for Statistical Computing) with Meta package was used for the statistical analysis. The methodological quality of the randomized trials was assessed by Cochrane's Collaboration tool for assessing risk of bias.

| RESULTS
Three RCT [11][12][13] representing 9463 patients were selected for the metaanalysis. The review and selection process is depicted in Figure 1. The endpoints were collected at 12 months in PIONEER AF-PCI, at 14 months in RE-DUAL PCI and at 6 months in AUGUSTUS study. The major characteristics of the patients of each study are detailed in Table 1.    Figure 2B) compared to VKA ( Figure S8). Funnel plots showed no publications bias ( Figure S9). All the trials were judged to be at low risk of bias via the Cochran's Collaboration tool for risk assessment (Table S1). Table S2 resumed the design and characteristics of the studies.

| DISCUSSION
Our meta-analysis shows, with strong consistency between studies, that in AF patients treated with PCI/ACS, the association of DOAC and antiplatelet agents was associated with a significant decrease in the risks of any, major and intracranial bleeding with similar rates of efficacy endpoints as compared to VKA-based regimen. The most important reduction of bleeding risk was observed with dual therapy with DOAC which was associated with similar efficacy as compared to VKA.
The major risk of combining oral anticoagulation and antiplatelet therapy in patients with AF treated with PCI is bleeding. There is a high, early, and long-term, risk of bleeding with the triple therapy including VKA, 5,14,15 with a 2-to 3-fold increase of bleeding complications compared to VKA. Such events are associated with a short-and long-term higher risk of mortality. 16,17 A study reported that half of the patients experiencing major bleeding with triple therapy died within 6 months; all of which as a consequence of intracranial bleeding. 18 The lower risk of major bleeding associated with dual therapy including VKA 6 and further with DOAC 12 appears as a major improvement of care in such patients. Our study showed that DOAC use reduces major bleeding by 39% and intracranial bleeding by 67% compared to VKA. When DOAC were used as dual therapy with a single antiplatelet agent, major bleeding and intracranial bleeding were reduced by 43% and 76%, respectively. Considering the stepwise increase of the risk of death associated with increased severity of bleeding, 19  with VKA and triple therapy may lead to antithrombotic treatment interruption and in turn promote ischemic events. 24 Bleeding can reduce oxygen delivery to the myocardium and promote platelet activation. 25 Major bleeding is associated with an increased risk of recurrent ischemic events including myocardial infarction and stroke in ACS patients. 17

| Limits
Our meta-analysis was not performed on individual patients' data.
Hence, analyses could not take in to account the individual levels of risk. A limited proportion of included patients presented with STelevation myocardial infarction. Knowing the higher risk of stent thrombosis associated with the latter condition, caution should be taken to extend the results in this setting. The doses of rivaroxaban used in the PIONEER AF-PCI study 12 were lower than those used for stroke prevention in the ROCKET-AF trial. 10 Although the results of PIONEER trial are in line with other trials included in our analysis, the effect of a 20 mg recommended dose remains un-assessed. The comparison between low and high doses of DOAC and between dual and triple therapy using DOAC are only exploratory as with the exception of AUGUSTUS trial 13 the included trials were not designed to assess antiplatelet regimens individually. Our research was limited to two main databases for studies retrieval. Substantial heterogeneity exists in between trials in terms of trial design as well as type and duration of antiplatelet/antithrombotic therapy used, which could affect interpretation of our results. Finally, only three randomized studies were included in the analysis which might limit the assessment of publication bias.

| CONCLUSION
Our study showed that the association of DOAC and antiplatelet agents after PCI and/or ACS in AF patients was associated with a lower risk of major bleeding, especially when considering dual therapy, while preventing thrombotic events similarly to VKA-based regimens.
Because of low rates of bleeding and no increase in risk of thrombotic events, dual therapy combining a DOAC and clopidogrel appears as the default regimen in most patients in this setting.