Non‐vitamin K antagonist oral anticoagulants (NOACs) post‐percutaneous coronary intervention: a network meta‐analysis

Samer Al Said; Samer Alabed; Klaus Kaier; Audrey R Tan; Christoph Bode; Joerg J Meerpohl; Daniel Duerschmied

doi:10.1002/14651858.CD013252.pub2

. 2019 Dec 20;2019(12):CD013252. doi: 10.1002/14651858.CD013252.pub2

Non‐vitamin K antagonist oral anticoagulants (NOACs) post‐percutaneous coronary intervention: a network meta‐analysis

Samer Al Said ^1,^✉, Samer Alabed ², Klaus Kaier ³, Audrey R Tan ⁴, Christoph Bode ⁵, Joerg J Meerpohl ⁶, Daniel Duerschmied ⁵

Editor: Cochrane Heart Group

PMCID: PMC6923523 PMID: 31858590

Abstract

Background

Clinicians must balance the risks of bleeding and thrombosis after percutaneous coronary intervention (PCI) in people with an indication for anticoagulation. The potential of non‐vitamin K antagonists (NOACs) to prevent bleeding complications is promising, but evidence remains limited.

Objectives

To review the evidence from randomised controlled trials assessing the efficacy and safety of non‐vitamin K antagonist oral anticoagulants (NOACs) compared to vitamin K antagonists post‐percutaneous coronary intervention (PCI) in people with an indication for anticoagulation.

Search methods

We identified studies by searching CENTRAL, MEDLINE, Embase, the Conference Proceedings Citation Index – Science and two clinical trials registers in February 2019. We checked bibliographies of identified studies and applied no language restrictions.

Selection criteria

We searched for randomised controlled trials (RCT) that compared NOACs and vitamin K antagonists for people with an indication for anticoagulation who underwent PCI.

Data collection and analysis

Two review authors independently checked the results of searches to identify relevant studies, assessed each included study, and extracted study data. We conducted random‐effects, pairwise analyses using Review Manager 5 and network meta‐analyses (NMA) using the R package 'netmeta'. We ranked competing treatments by P scores, which are derived from the P values of all pairwise comparisons, and allow ranking of treatments on a continuous 0 to 1 scale.

Main results

We identified nine RCTs that met the inclusion criteria, but four were ongoing trials, and were not included in this analysis. We included five RCTs, with 8373 participants, in the NMA (two RCTs compared apixaban to a vitamin K antagonist, two RCTs compared rivaroxaban to a vitamin K antagonist, and one RCT compared dabigatran to a vitamin K antagonist).

Very low‐ to moderate‐certainty evidence suggests little or no difference between NOACs and vitamin K antagonists in death from cardiovascular causes (not reported in the dabigatran trial), myocardial infarction, stroke, death from any cause, and stent thrombosis. Apixaban (RR 0.85, 95% CI 0.77 to 0.95), high dose rivaroxaban (RR 0.86, 95% CI 0.74 to 1.00), and low dose rivaroxaban (RR 0.80, 95% CI 0.68 to 0.92) probably reduce the risk of recurrent hospitalisation compared with vitamin K antagonists. No studies looked at health‐related quality of life.

Very low‐ to moderate‐certainty evidence suggests that NOACs may be safer than vitamin K antagonists in terms of bleeding. Both high dose dabigatran (RR 0.53, 95% CI 0.29 to 0.97), and low dose dabigatran (RR 0.38, 95% CI 0.21 to 0.70) may reduce major bleeding more than vitamin K antagonists. High dose dabigatran (RR 0.83, 95% CI 0.72 to 0.96), low dose dabigatran (RR 0.66, 95% CI 0.58 to 0.75), apixaban (RR 0.67, 95% Cl 0.51 to 0.88), high dose rivaroxaban (RR 0.66, 95% CI 0.52 to 0.83), and low dose rivaroxaban (RR 0.71, 95% CI 0.57 to 0.88) probably reduce non‐major bleeding more than vitamin K antagonists.

The results from the NMA were inconclusive between the different NOACs for all primary and secondary outcomes.

Authors' conclusions

Very low‐ to moderate‐certainty evidence suggests no meaningful difference in efficacy outcomes between non‐vitamin K antagonist oral anticoagulants (NOAC) and vitamin K antagonists following percutaneous coronary interventions (PCI) in people with non‐valvular atrial fibrillation. NOACs probably reduce the risk of recurrent hospitalisation for adverse events compared with vitamin K antagonists.

Low‐ to moderate‐certainty evidence suggests that dabigatran may reduce the rates of major and non‐major bleeding, and apixaban and rivaroxaban probably reduce the rates of non‐major bleeding compared with vitamin K antagonists.

Our network meta‐analysis did not show superiority of one NOAC over another for any of the outcomes. Head to head trials, directly comparing NOACs against each other, are required to provide more certain evidence.

Plain language summary

Non‐vitamin K antagonist oral anticoagulants (NOACs) after a heart vessel stent placement

Background

Choosing the optimal treatment for people on long term blood thinners (due to atrial fibrillation), who require a heart vessel stent placement (after a heart attack or angina) remains a challenge in clinical practice. They need to take blood thinners to prevent a stroke, and antiplatelet drugs to prevent blood clots in the stents. However, this combination increases the risk of potentially life‐threatening bleeding, and thus, the optimal treatment remains uncertain. The aim of this review was to investigate whether next‐generation blood thinners (NOACs) are safer and more effective than older blood‐thinning medications (such as warfarin) in this group of individuals.

Study characteristics

We identified nine studies that compared NOACs with warfarin, four of which were ongoing studies. We included five trials involving 8373 participants in this review. Evidence is current to February 2019.

Key results

There may be little or no difference in effect between NOACs and warfarin in people with atrial fibrillation, who underwent heart vessel stenting. However, NOACs probably reduce the need for hospitalisation compared to warfarin.

NOACs may be safer than warfarin. One of NOACs drugs (dabigatran) may reduce the rate of both major and non‐major bleeding. Other NOAC drugs (apixaban and rivaroxaban) probably reduce the rate of non‐major bleeding. There was no significant difference between NOACs agents in any primary or secondary outcomes.

Quality of evidence:

The evidence ranged from Very low‐ to moderate‐certainty, indicating the need for more research on this issue.

Summary of findings

Background

Description of the condition

Coronary artery disease (CAD) is the leading cause of death worldwide (Benjamin 2018; Finegold 2013). According to the World Health Organization (WHO), CAD is responsible for about 15% of all deaths globally (WHO 2015). It remains one of the world's largest health problems, despite dramatic medical advances over the last few decades.

CAD is usually caused by atherosclerosis, in which fatty deposits accumulate on the walls of the coronary arteries. In general, CAD is classified into chronic coronary syndromes and acute coronary syndrome, the latter of which has three forms: unstable angina, non‐ST‐elevation myocardial infarction, and ST‐elevation myocardial infarction (Ibanez 2018; Knuuti 2019; Roffi 2016)).

Percutaneous coronary intervention (PCI) is the most common procedure used in the invasive treatment of people with CAD (Khera 2016), and accounted for 3.3% of all operating room procedures performed in USA in 2014 (McDermott 2017). Every year, millions of people undergo PCI, which is a non‐surgical revascularisation technique used to make narrowed or blocked blood coronary arteries in order to restore or improve blood flow to the heart muscle (McGrath 1999; Peterson 2000). PCI is indicated in acute coronary syndrome, and may also be performed in people with chronic coronary syndromes who are dissatisfied with their quality of life, particularly if their symptoms are uncontrolled with medication (Khera 2016; Windecker 2014).

Antithrombotic therapy is required after PCI to reduce the risk of recurrent cardiovascular events and stent thrombosis (Leon 1998; Valgimigli 2017). Long‐term dual antiplatelet therapy (DAPT) with aspirin and a P2Y12 receptor antagonist has been shown to reduce mortality and morbidity post‐PCI (Atar 2014; Mauri 2014; Wallentin 2009; Wiviott 2007; Yusuf 2001). Although clopidogrel remains the most widely used P2Y12 inhibitor, the more potent P2Y12 inhibitors prasugrel and ticagrelor are favoured for DAPT without anticoagulation following PCI in people with a high thrombotic risk (Ibanez 2018; Roffi 2016).

Approximately 5% to 8% of people undergoing PCI have an indication for long‐term oral anticoagulation, most commonly due to atrial fibrillation (Pérez‐Gómez 2004; Rubboli 2007; Wang 2008). CAD shares similar risk factors (diabetes, hypertension, genetics) with atrial fibrillation (Kohli 2014; Neelankavil 2015). As the population ages, the incidence of CAD and atrial fibrillation that require intervention has increased (Cho 2015). DAPT is not sufficient to reduce the risk of ischaemic stroke in atrial fibrillation (Connolly 2006), therefore people with atrial fibrillation and an increased risk of stroke (CHA₂DS₂‐VASc score of ≥ 1) are treated with oral anticoagulation (Camm 2010; January 2014; Kirchhof 2016). However, triple therapy by combining DAPT with oral anticoagulation significantly increases the risk of bleeding (Paikin 2010).

Balancing the risk of bleeding and thrombosis post‐PCI in people with an indication for anticoagulation is a challenging clinical issue. Questions regarding the efficacy and safety of non‐vitamin K antagonist oral anticoagulants (NOACs) in combination with antiplatelet agents therefore need to be answered.

Description of the intervention

NOACs are a group of oral anticoagulants that inhibit specific coagulation factors: thrombin or activated factor Xa. NOACs have pharmacologic advantages over vitamin K antagonists, including absence of food interactions, limited drug interactions, and predictable pharmacokinetics. Therefore, NOACs can be given at fixed dosing without the need for dietary restrictions or routine coagulation monitoring. NOACs have a rapid onset of action, within 0.5 to 4 hours, a faster offset of action, and shorter half‐life, therefore, unlike vitamin K antagonists, NOAC therapy can be initiated without temporary concurrent parenteral anticoagulant, such as low molecular weight heparin (Bauer 2013; Mekaj 2015; Zirlik 2017).

While vitamin K antagonists reduce the synthesis of functional vitamin K‐depending clotting factors II, VII, IX, X, as well as protein C and protein S, NOACs directly inhibit an activated clotting factor, either FIIa or FXa. Dabigatran is currently the only direct thrombin inhibitor, and was the first NOAC, approved in 2010. Factor Xa inhibitors include rivaroxaban, apixaban, and edoxaban (Mekaj 2015). Some NOACs have not been approved, due to safety concerns or ineffectiveness, such as the direct thrombin inhibitor ximelagatran and direct factor Xa inhibitors darexaban and letaxaban (Ahrens 2012; Steg 2011; Wallentin 2003).

NOACs are approved for the prevention of venous thromboembolism in orthopaedic patients undergoing hip or knee replacement surgery, the treatment of deep vein thrombosis or pulmonary embolism, and for stroke prevention in non‐valvular atrial fibrillation. Rivaroxaban has also been approved for secondary prophylaxis post‐acute coronary syndrome in Europe (Bayer Pharma AG 2016; European Medicines Agency 2013; Kearon 2016; Kirchhof 2016; Konstantinides 2014). Vitamin K antagonists remain the medication of choice in people with valvular atrial fibrillation (due to mitral stenosis), mechanical heart valves, venous thromboembolism associated with antiphospholipid syndrome, and unstable cancer patients. NOACs are not approved for these indications due to insufficient data on their efficacy and safety (Lee 2016).

NOACs are at least as effective as vitamin K antagonists for the prevention of stroke and systemic embolism in people with non‐valvular atrial fibrillation, and they are associated with a significant reduction of intracranial haemorrhage and mortality (Connolly 2009; Giugliano 2013; Granger 2011; Miller 2012; Patel 2011; Ruff 2014).

NOACs are excreted renally, therefore, they are contraindicated in renal failure, and a reduced dose should be used with people with renal impairment. Dose adjustments are also necessary in people with low or very high weight. The development of NOAC reversal agents will help to effectively manage patients with life‐threatening bleeding or requiring urgent surgery. Idarucizumab was licensed as a reversal agent for dabigatran, and andexanet alfa has been recently approved as a reversal agent for rivaroxaban and apixaban (Cuker 2019; Glund 2015; Pollack 2015).

Some limitations to the use of NOACs include higher cost compared to vitamin K antagonists, and the lack of laboratory monitoring to objectively assess compliance.

The role of plasmatic coagulation in the atherothrombotic clot formation has recently been gaining importance. The thrombotic complications after PCI are not only due to platelet aggregation, but also to plasmatic coagulation with thrombin‐mediated fibrin formation (Merlini 1994). The addition of vitamin K antagonists (e.g. warfarin) to aspirin to reduce atherothrombotic recurrences after acute coronary syndrome increases bleeding complications (Andreotti 2006; Rothberg 2005). As a result, studies have tested novel therapies targeting thrombin‐mediated pathways, including NOACs, to prevent thrombotic complications of acute coronary syndrome. Although trials with dabigatran and apixaban failed to show a significant clinical benefit post‐acute coronary syndrome (Alexander 2009; Alexander 2011; Oldgren 2011), rivaroxaban, in addition to DAPT has shown a significant reduction of cardiovascular and overall mortality (Gibson 2011; Mega 2012).

The antithrombotic strategy post‐PCI in people with an indication of oral anticoagulation (most commonly atrial fibrillation) poses a dilemma, as they require both antiplatelet and anticoagulation therapy. As a result, recent trials have assessed different antithrombotic treatment regimens and drug combinations to maintain anticoagulation, but minimise bleeding complications. One of these strategies is to withdraw aspirin. The combination of oral anticoagulation and clopidogrel was at least equal to, or better than, the combination of oral anticoagulation, clopidogrel, and aspirin, in efficacy and safety (Dewilde 2013; Lamberts 2013). Recent randomised controlled trials have evaluated alternative therapies, such as NOACs. The PIONEER AF‐PCI study showed that using rivaroxaban with a single P2Y12 antagonist or with DAPT lowers rates of clinically significant bleeding, compared to conventional triple therapy with warfarin plus DAPT (Gibson 2016a). Compared to warfarin, dabigatran and a P2Y12 antagonist without aspirin given post‐PCI to people with non‐valvular atrial fibrillation, is associated with a reduced risk of bleeding, and non‐inferior efficiency in preventing thromboembolic events (RE‐DUAL PCI 2017). The AUGUSTUS trial evaluated the role of apixaban compared to warfarin among people with atrial fibrillation undergoing coronary revascularisation (AUGUSTUS 2019). The ENTRUST‐AF PCI trial assessed the safety of edoxaban compared to VKAin patients with atrial fibrillation who had PCI (Vranckx 2018).

To date, combination therapy with a NOAC and an antiplatelet agent has not been addressed in people with venous thromboembolism, and it is unclear whether data from atrial fibrillation trials can be extrapolated when they undergo PCI.

How the intervention might work

Thrombotic complications after PCI have been primarily considered a platelet‐mediated process. However, rates of recurrent cardiovascular ischaemic events still remain high, despite use of DAPT regimens. This could be due to increased thrombin generation and an associated increase in clot formation (Merlini 1994; Moon 2017). NOACs selectively target thrombin (IIa) or activated factor X (Xa), which play a role in the coagulation cascade that leads to fibrin formation and resultant thrombosis. Therefore, NOACs may be beneficial post‐PCI in preventing thrombus formation (Mackman 2008).

Why it is important to do this review

The number of interventional coronary procedures has increased dramatically in recent years (Camm 2010). Current guidelines recommend triple therapy, comprising aspirin, clopidogrel, and oral anticoagulation post‐PCI, for people with an indication for anticoagulation and at a high risk for ischaemic complications, in order to reduce their risk of stent thrombosis and thrombotic arterial or venous events (January 2014; Levine 2016; Lip 2014; Valgimigli 2017). However, this regimen is associated with a high rate of bleeding complications, and the optimal anticoagulation therapy remains uncertain (Paikin 2010). The role of NOACs in these individuals is still not fully understood, and treatment decisions must rely on limited evidence.

We reviewed the evidence on the efficacy and safety of NOACs in people undergoing PCI, with an indication for anticoagulation. Given the complexity of the condition, and in the absence of randomised controlled trials comparing different types of NOACs against each other, it was essential to carry out a comprehensive and comparative evaluation of all available treatment options within the framework of a network meta‐analysis, to determine which treatments, if any, were the most effective and safe.

Network meta‐analysis has many advantages over conventional pairwise meta‐analysis, as the technique allows one to compare all interventions, including those for which head‐to‐head comparisons have not been conducted (Jansen 2008; Lu 2004). It also helps to understand the amount of available evidence for each treatment and treatment comparison (Mills 2013).

This review includes a network meta‐analysis that evaluates the efficacy and safety of different NOACs and vitamin K antagonists (Figure 1; Figure 2; Figure 3; Figure 4; Figure 5; Figure 6; Figure 7; Figure 8). To our knowledge, no other systematic review and network meta‐analysis has assessed the safety and efficacy of NOACs in people undergoing PCI with an indication for anticoagulation. We aimed to present the most current and best available evidence to patients, clinicians, policymakers, and researchers.

Network diagram for death from cardiovascular causes. Nodes are weighted according to the number of studies that includes the respective intervention. Edges are weighted according to the number of patients included in the respective comparison. Numbers on the lines represent the number of trials and participants for each comparison.

Network diagram for myocardial infarction. Nodes are weighted according to the number of studies that includes the respective intervention. Edges are weighted according to the number of patients included in the respective comparison. Numbers on the lines represent the number of trials and participants for each comparison.

Network diagram for stroke. Nodes are weighted according to the number of studies that includes the respective intervention. Edges are weighted according to the number of patients included in the respective comparison. Numbers on the lines represent the number of trials and participants for each comparison.

Network Diagram for major bleeding. Nodes are weighted according to the number of studies that includes the respective intervention. Edges are weighted according to the number of patients included in the respective comparison. Numbers on the lines represent the number of trials and participants for each comparison.

Network diagram for death from any cause. Nodes are weighted according to the number of studies that includes the respective intervention. Edges are weighted according to the number of patients included in the respective comparison. Numbers on the lines represent the number of trials and participants for each comparison.

Network diagram for stent thrombosis. Nodes are weighted according to the number of studies that includes the respective intervention. Edges are weighted according to the number of patients included in the respective comparison. Numbers on the lines represent the number of trials and participants for each comparison.

Network diagram for any non‐major bleeding. Nodes are weighted according to the number of studies that includes the respective intervention. Edges are weighted according to the number of patients included in the respective comparison. Numbers on the lines represent the number of trials and participants for each comparison.

Network diagram for recurrent hospitalisation. Nodes are weighted according to the number of studies that includes the respective intervention. Edges are weighted according to the number of patients included in the respective comparison. Numbers on the lines represent the number of trials and participants for each comparison.

Objectives

To review the evidence from randomised controlled trials assessing the efficacy and safety of non‐vitamin K antagonist oral anticoagulants compared to vitamin K antagonists post‐percutaneous coronary intervention, in people with an indication for anticoagulation.

Methods

Criteria for considering studies for this review

Types of studies

We included parallel‐arm randomised controlled trials (RCT).

Types of participants

We included adult participants (≥ 18 years) who had percutaneous coronary intervention (PCI) for acute coronary syndrome or chronic coronary syndromes, and who had an indication for non‐vitamin K antagonist oral anticoagulants (NOAC), for the prevention of stroke in non‐valvular atrial fibrillation or venous thromboembolism).

We excluded participants with the following comorbidities and characteristics.

High risk of bleeding or anticoagulation contraindication (e.g. active or recent clinically significant bleeding, thrombocytopenia, prior intracranial haemorrhage)
Prior stroke or transient ischaemic attack
Prior stent thrombosis or stent‐within‐a‐stent
Atrial fibrillation from reversible conditions (e.g. pulmonary embolism, recent surgery, thyroid abnormalities)
Active malignancy, or life expectancy less than one year

In trials with mixed populations, that is trials where some participants met the inclusion criteria and others did not, we attempted to include only the eligible participants, if this information was reported separately or could be obtained from trial authors. Otherwise, we included studies with a mixed population if the majority (> 50%) of the participants met the eligibility criteria.

Types of interventions

We included trials comparing any type of NOAC (dabigatran, rivaroxaban, apixaban, edoxaban) with placebo, vitamin K antagonists, or a different type of NOAC.

We included the following co‐interventions, even if they were part of the randomised treatment: single or dual antiplatelet therapy.

We excluded NOACs that were not licensed by the US Food and Drug Administration or European Medicines Agency because of lack of safety or effectiveness evidence (e.g. ximelagatran, darexaban, and letaxaban), as they are not clinically relevant.

Types of outcome measures

Primary outcomes

Death from cardiovascular causes
Myocardial infarction
Stroke
Major bleeding, as defined by the Thrombolysis In Myocardial Infarction (TIMI) criteria (Mehran 2011)

Secondary outcomes

Death from any cause
Stent thrombosis
Any non‐major TIMI bleeding
Recurrent hospitalisation
Health‐related quality of life (HRQL), assessed using validated instruments (e.g. 36‐Item Short Form Health Survey (SF‐36), EuroQol (EQ‐5D))

We assessed mortality outcomes, myocardial infarction, stroke, stent thrombosis, bleeding outcomes, and recurrent hospitalisation at the latest point of follow‐up for each trial.

Definitions of clinical events (e.g. myocardial infarction, stroke, and stent thrombosis) outcomes were according to the individual trialists.

Search methods for identification of studies

Electronic searches

We identified trials through systematic searches of the following bibliographic databases on 18 February 2019:

Cochrane Central Register of Controlled Trials (CENTRAL; 2019, Issue 2) in the Cochrane Library;
Epub Ahead of Print, In‐Process & Other Non‐Indexed Citations, MEDLINE Daily and MEDLINE Ovid (1946 to February 15, 2019);
Embase Classic and Embase Ovid (1947 to 2019 February 15);
Conference Proceedings Citation Index – Science (CPCI‐S) on Web of Science Clarivate Analytics (1990 to 15 February 2019).

Appendix 1 shows the search strategies. We applied the sensitivity‐maximising version of the Cochrane RCT filter to our MEDLINE search, and applied terms as recommended in the Cochrane Handbook for Systematic Reviews of Interventions to our Embase search (Lefebvre 2011). We applied an adaptation of the Cochrane RCT filter to CPCI‐S. We imposed no restrictions on the date or language of publication.

Searching other resources

We searched the World Health Organization (WHO) International Clinical Trials Registry Platform Search Portal (apps.who.int/trialsearch/), and Clinicaltrials.gov (clinicaltrials.gov), for ongoing and unpublished trials, on 18 February 2019. We checked the reference lists of all included trials identified by the above methods.

Data collection and analysis

Selection of studies

Two review authors (SAS, SA) independently screened the titles and abstracts of all the records identified as a result of the search, and coded them as 'retrieve' (eligible or potentially eligible or unclear) or 'do not retrieve'. They asked a third review author (DD) to arbitrate in case of disagreement. We retrieved the full‐text reports and publications of the 'retrieve' records. Two review authors (SAS, SA) independently screened the full texts and identified studies for inclusion, and identified and recorded reasons for exclusion of the ineligible studies. They resolved disagreements through discussion, or by consulting a third review author (DD). We identified and excluded duplicates and collated multiple reports of the same study so that each study, rather than each report, was the unit of interest in the review. We recorded the selection process in sufficient detail and completed a PRISMA flow diagram and 'Characteristics of excluded studies' table (Liberati 2009).

Data extraction and management

Two review authors (SAS, SA) independently extracted data from the included trials. We extracted and collated data using a standardised, agreed upon, data extraction form. Data collected included:

Methods: study design, total duration of study, number of study centres and location, study setting, and date of study.
Participants: N randomised, N lost to follow‐up or withdrawn, N analysed, mean age, age range, gender, inclusion criteria, and exclusion criteria.
Interventions: intervention, comparison, concomitant medications, and excluded medications.
Outcomes: primary and secondary outcomes specified and collected, and time points reported.
Notes: funding for trial, and notable conflicts of interest of trial authors.

One review author (SA) transferred the data into the Review Manager 5 file (RevMan 2014). A second review author (SAS) double‐checked that the data were entered correctly, by comparing the data presented in the systematic review with the data extraction form, and spot‐checked study characteristics for accuracy against the trial report.

Data on potential effect modifiers

All included studies used very similar inclusion criteria and found comparable baseline characteristics. From each study, we extracted the following characteristics that may have acted as effect modifiers: age, gender, lipid levels, BMI, co‐morbidities, and embolic risk.

Assessment of risk of bias in included studies

Two review authors (SAS, SA) independently assessed the risk of bias for each study using the criteria outlined in Chapter 8 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2017). We resolved disagreements by discussion, or by involving a third review author (DD). We assessed risk of bias according to the following domains.

Random sequence generation
Allocation concealment
Blinding of participants and personnel
Blinding of outcome assessment
Incomplete outcome data
Selective outcome reporting
Other bias (e.g. conflicts of interest or imbalance in baseline characteristics between study arms)

We classified each potential source of bias as high, low, or unclear, and provided a quote from the study report together with a justification for our judgement in the 'Risk of bias' tables. We summarised the 'Risk of bias' judgements across different studies for each of the domains listed.

When considering treatment effects, one of the factors we took into account was the risk of bias in the studies that contributed to that outcome.

Assessment of bias in conducting the systematic review

We conducted the review according to the published protocol, and reported any deviations from it in the 'Differences between protocol and review' section.

Measures of treatment effect

We analysed dichotomous data as risk ratios with 95% confidence intervals (CI) and continuous data as mean differences with 95% CIs. We entered data measured on a scale with a consistent direction of effect.

Unit of analysis issues

Our unit of analysis was the participant. We did not include cross‐over trials. We did not encounter any cluster‐randomised trials.

Dealing with missing data

In the case of missing data for participants, or missing statistics (such as standard deviations), we had intended to contact the trial authors; however, this was not required.

Assessment of heterogeneity

We used the I² statistic to measure heterogeneity among the trials in each analysis, and followed the recommendations for thresholds outlined in Section 9.5.2 of the Cochrane Handbook for Systematic Reviews of Interventions (Deeks 2017; Higgins 2017).

In network meta‐analyses, inconsistency occurs when direct and indirect estimates do not agree. Heterogeneity and inconsistency were measured in the network meta‐analyses by decomposing the Q statistics, using the R package 'netmeta', and assessing differences between direct and indirect estimates (Rücker 2017). Specifically, we planned to use the R package 'netsplit' to split network estimates into the contribution of direct and indirect evidence, and to test for local inconsistencies in our network meta‐analyses. We also planned to create net heat plots (Jackson 2012; Krahn 2013), but did not, due to the limited number of included studies. The assessment of heterogeneity for pairwise comparisons included visual inspection of the forest plot. In the case of relevant heterogeneity in the network meta‐analyses for the primary outcomes, we had planned to explore possible sources, and conduct subgroup and sensitivity analyses, based on factors described in the subgroup and sensitivity analysis sections.

Assessment of intransitivity across treatment comparisons

We considered transitivity by assessing clinical and methodological comparability. We concluded, that given the very similar inclusion criteria and comparable included populations in the various RCTs, the transitivity assumption holds, assuming the following: 1) the common treatment used to compare different NOACs indirectly was similar when it appeared in different trials; 2) No relevant variation in effect modifiers were identified between trials.

Assessment of reporting biases

To assess the risk of publication bias, we had planned to construct funnel plots with at least 10 trials, to explore possible small‐study biases for the primary outcomes; however, this was not possible owing to the limited number of included studies.

Data synthesis

Methods for direct treatment comparisons

We conducted pairwise meta‐analyses using random‐effects models in Review Manager 5 and R for every treatment comparison with at least two studies. We selected the random‐effects model as the model of choice. With a random‐effects model, the true effect size may or may not vary from study to study, and thus does not assume that either is the case. As part of the analysis, the amount of variance in true effects is estimated across studies, and the estimate may or may not be zero. With a fixed‐effect model, the true effect size does not vary from study to study. Therefore, the fixed‐effect model is more restrictive. It imposes a constraint that is neither necessary nor plausible.

Methods for indirect and network comparisons

We evaluated the geometry of the network diagrams to determine whether a network meta‐analysis was appropriate. The diagrams demonstrated that the evidence was conducive for a network meta‐analysis, so we performed a multivariate meta‐analysis within a frequentist framework, using a random‐effects model, using the R package 'netmeta' (Rücker 2017). This technique is appropriate for multi‐arm studies.

We performed network meta‐analyses for all efficacy and safety outcomes at the latest point of follow‐up for each trial. We estimated the relative rankings for the main outcomes included in the 'Sumary of findings' tables (death from cardiovascular causes, myocardial infarction, stroke, major bleeding, death from any cause, and stent thrombosis).

The nodes of the network are the interventions specified in the review inclusion criteria; we did not combine any. We added a network plot for each outcome (Figure 1; Figure 2; Figure 3; Figure 4; Figure 5; Figure 6; Figure 7; Figure 8).

We presented all results as summary relative effects (risk ratios) for each possible pair of treatments.

Competing treatments are ranked by P scores, which allow ranking of treatments on a continuous 0 to 1 scale, and are derived from the P values of all pairwise comparisons. P scores are based solely on the point estimates and standard errors of the frequentist network meta‐analysis estimates under normality assumption. P scores measure the mean extent of certainty that a treatment is better than the competing treatments. This interpretation is comparable to that of the surface under the cumulative ranking curve (SUCRA (Rücker 2015)).

Subgroup analysis and investigation of heterogeneity

We planned to carry out the following subgroup analyses for the primary outcomes:

Indication of NOAC: non‐valvular atrial fibrillation and venous thromboembolism.
Indication for PCI: ST‐elevation myocardial infarction, non‐ST‐elevation myocardial infarction, unstable and stable angina.
Risk of stroke: CHA₂DS₂‐VASc score < 2, 2 to 4, > 4.
Kind of coronary stents: dual‐therapy stent, bioresorbable vascular scaffold, bio‐engineered stent, drug‐eluting stent, bare‐metal stent.
Evaluation of the involved coronary vessel.
Mean age of the participants in each trial: elderly (≥ 75 years of age) versus non‐elderly (< 75 years of age).

We had planned to investigate heterogeneity through subgroup analyses, but abandoned the plan due to the limited number of included studies.

Sensitivity analysis

We had planned to conduct sensitivity analyses to assess the effect of excluding studies judged as being at unclear or high risk of bias in any of the 'Risk of bias' domains. This was not possible, as all included studies had at least one domain at a high risk of bias.

Summary of findings and assessment of the certainty of the evidence

'Summary of findings' table

We created 'Summary of findings' tables with the following outcomes: death from cardiovascular causes, myocardial infarction, stroke, major bleeding, death from any cause, and stent thrombosis. We used the five GRADE considerations (study limitations, consistency of effect, imprecision, indirectness, and publication bias) to assess the quality of the body of evidence as it related to the studies that contributed data to the meta‐analyses for each outcome. We applied the four‐step approach, presented by Puhan and colleagues, to rate the certainty of evidence in each of the direct, indirect, and network meta‐analysis estimates (Puhan 2014).

Overall, we used methods and recommendations described in Section 8.5 and Chapter 12 of the Cochrane Handbook for Systematic Reviews of Interventions, and GRADEpro GDT software (GRADEpro GDT; Higgins 2017; Schünemann 2017). In the 'Summary of findings' tables, we presented direct evidence, indirect evidence, and results from the network meta‐analyses (Brignardello‐Petersen 2017; Brignardello‐Petersen 2018a; Puhan 2014).

Two review authors (SAS, AT) independently made judgements about the quality of the evidence, with disagreements resolved by discussion. They justified and documented their judgements, and incorporated them into the reporting of results for each outcome.

League table

We created league tables using the following outcomes: death from cardiovascular causes, myocardial infarction, stroke, major bleeding, death from any cause, and stent thrombosis. League tables use a matrix structure, where the upper triangle presents the results from direct (pairwise) meta‐analyses, and the lower triangle presents the results from the network meta‐analyses (Chaimani 2019). Comparisons between treatments are read from left to right; the estimate is in the cell in common between the column‐defining treatment and the row‐defining treatment. We presented results as risk ratios (95% CI), where a risk ratio of < 1 favours the row‐defined treatment.

Results

Description of studies

Results of the search

We identified 380 references. We screened 315 titles and abstracts after removing duplicates. We excluded most studies on abstract review as they were not randomised controlled trials (RCT), or did not compare non‐vitamin K antagonist oral anticoagulants (NOACs) with a vitamin K antagonist. We assessed 60 full‐text records and excluded 30 studies. Four studies (five references) were ongoing trials (Gao 2015; NCT02789917; NCT03536611; Vranckx 2018). We included five studies (25 references) in the meta‐analysis (RE‐DUAL PCI 2017; PIONEER AF‐PCI 2016; Kopin 2018; AUGUSTUS 2019; Sherwood 2016). See Figure 9 for details.

Included studies

The Characteristics of included studies table shows a detailed description of the trials.

Study designs

All included trials used a parallel‐group design. All were international, multicentred trials. The follow‐up of the trials ranged between six months (AUGUSTUS 2019), and 2.2 years (Sherwood 2016). Two trials had three arms (PIONEER AF‐PCI 2016; RE‐DUAL PCI 2017). Two trials had two arms (Kopin 2018; Sherwood 2016), one trial had a 2 x 2 factorial design (AUGUSTUS 2019).

The PIONEER AF‐PCI trial had three arms. Group 1: rivaroxaban 15 mg once daily (or a dose of 10 mg once daily in cases of moderate renal impairment) plus single antiplatelet therapy (clopidogrel, ticagrelor, or prasugrel) for 12 months. Group 2: a very low dose of rivaroxaban (2.5 mg twice daily) plus dual antiplatelet therapy (low dose aspirin and either clopidogrel, ticagrelor, or ticagrelor) for 1, 6, or 12 months. After completion of one or six months of dual antiplatelet therapy (DAPT), participants received rivaroxaban 15 mg daily plus low dose aspirin. Group 3: triple therapy with vitamin K antagonists plus dual antiplatelet therapy (low dose aspirin and either clopidogrel, ticagrelor, or prasugrel) for 1, 6, or 12 months. After completion of one or six months of DAPT, participants received vitamin K antagonists plus low dose aspirin (PIONEER AF‐PCI 2016).

The RE‐DUAL PCI trial had three arms. Group 1: a low dose of dabigatran (110 mg twice daily) plus single antiplatelet therapy (either clopidogrel or ticagrelor). Group 2: a high dose of dabigartran (150 mg twice daily) plus single antiplatelet therapy (either clopidogrel or ticagrelor). Group 3: triple therapy with warfarin plus dual antiplatelet therapy (low dose aspirin and either clopidogrel or ticagrelor). The aspirin was discontinued after one month in participants in whom a bare‐metal stent was implanted, and after three months in participants in whom a drug‐eluting stent was implanted (RE‐DUAL PCI 2017).

The AUGUSTUS trial used a 2 x 2 factorial design. Participants were randomised to receive open‐label apixaban or a vitamin K antagonist and to receive double‐blind aspirin or matching placebo. All participants received a P2Y12 inhibitor (clopidogrel in most of the cases). Four antithrombotic regimens were compared: 1. apixaban + P2Y12 inhibitor + placebo; 2. apixaban + P2Y12 inhibitor + aspirin; 3. vitamin K antagonist + P2Y12 inhibitor + placebo; 4. vitamin K antagonist + P2Y12 inhibitor + aspirin. The AUGUSTUS trial was the only trial that compared triple therapy (NOAC + aspirin + P2Y12 inhibitor) with triple therapy (vitamin K antagonists + aspirin + P2Y12 inhibitor (AUGUSTUS 2019)).

Kopin 2018 is a small post hoc subgroup analysis of the participants in the ARISTOTLE trial who underwent percutaneous coronary intervention (PCI) during follow‐up. ARISTOLTLE compared apixaban to vitamin K antagonists. Antiplatelet therapy use post‐PCI was variable. DAPT was used in 35% of participants, 23% received aspirin only, 13% received a P2Y12 inhibitor only, and 29% received no antiplatelet therapy.

Sherwood 2016 is a small post hoc subgroup analysis of the participants in the ROCKET AF trial who underwent PCI during follow‐up. The ROCKET AF trial compared rivaroxaban to vitamin K antagonists. Antiplatelet therapy use post‐PCI was variable. DAPT was used in 37%, 34% received single antiplatelet therapy, and 15% of participants received no antiplatelet therapy.

The PIONEER AF‐PCI, RE‐DUAL PCI, and AUGUSTUS trials were not designed to be large enough to detect small but potentially meaningful differences in the incidence of ischaemic events.

Study design characteristics of the included studies can be found in the Characteristics of included studies tables.

Sample sizes

Included trials randomised 8373 participants in total. The AUGUSTUS trial randomised 4614 participants, the PIONEER AF‐PCI trial randomly assigned 2124, and the RE‐DUAL PCI trial randomly assigned 2725. Kopin 2018 was a subgroup study of 316 participants, and Sherwood 2016 was a subgroup study of 153 participants.

Participants

Male participants were predominant in four trials (AUGUSTUS 2019; Kopin 2018; PIONEER AF‐PCI 2016; RE‐DUAL PCI 2017). One trial did not report the gender of the participants (Sherwood 2016). All studies were conducted in adults. All participants had non‐valvular atrial fibrillation as the indication for long‐term oral anticoagulation.

Interventions

Two RCTs compared apixaban with vitamin K antagonists (AUGUSTUS 2019; Kopin 2018). Two RCTs compared rivaroxaban with vitamin K antagonists (PIONEER AF‐PCI 2016; Sherwood 2016). One RCT compared dabigatran with vitamin K antagonists (RE‐DUAL PCI 2017).

Outcomes

All trials reported rates of myocardial infarction and stroke. Four trials reported death from cardiovascular causes (AUGUSTUS 2019; Kopin 2018; PIONEER AF‐PCI 2016; Sherwood 2016). The trial examining dabigatran did not report this outcome (RE‐DUAL PCI 2017). All trials except Sherwood 2016 reported death from any cause. Three trials provided rates of stent thrombosis, non‐major bleeding, and recurrent hospitalisation (AUGUSTUS 2019; PIONEER AF‐PCI 2016; RE‐DUAL PCI 2017). No trials assessed health‐related quality of life.

Excluded studies

We excluded 30 studies after assessing the full texts: 12 RCTs used the wrong study design, 5 measured the wrong outcomes, 5 had the wrong population, 5 had a wrong comparator, 2 had a wrong intervention, and 1 had a wrong indication. See Characteristics of excluded studies table. Studies were excluded based on the outcomes reported, if they were substudies of included studies but did not contain any information helpful to our review, or addressed outcomes not specified in our review.

Ongoing trials

We identified four ongoing trials. Gao 2015 is a randomised study comparing rivaroxaban and warfarin in people with atrial fibrillation and coronary artery disease undergoing PCI – the RT‐AF trial; NCT02789917 is an open‐label trial comparing apixaban 5 mg/dL and phrenprocoumon in people with acute coronary syndrome and atrial fibrillation – the APPROACH‐ACS‐AF trial; NCT03536611 is a randomised study comparing dabigatran etexilate and warfarin in Chinese patients with non‐valvular atrial fibrillation undergoing PCI with stenting; and Vranckx 2018 is an open‐label trial examining the effects of edoxaban versus warfarin in people with atrial fibrillation undergoing PCI – the ENTRUST‐AF PCI trial. See Characteristics of ongoing studies table for details.

Risk of bias in included studies

The risk of bias in the included trials is summarised in Figure 10 and Figure 11, and detailed in the 'Characteristics of included studies' table.

Risk of bias graph: review authors’ judgements about each risk of bias item presented as percentages across all included studies

Risk of bias summary: review authors’ judgements about each risk of bias item for each included study

Allocation

Randomization was performed via an interactive voice‐response system in four trials (AUGUSTUS 2019; Kopin 2018; PIONEER AF‐PCI 2016; Sherwood 2016). RE‐DUAL PCI 2017 used a random number generator. We assessed the risk of bias arising from the method of generation of the allocation sequence to be low in all included studies.

Blinding

We assessed the risk of performance bias due to lack of blinding of participants and personnel as high in three trials (AUGUSTUS 2019; PIONEER AF‐PCI 2016; RE‐DUAL PCI 2017) and low in two trials (Kopin 2018; Sherwood 2016). We assessed the risk of detection bias as low in all included trials.

Incomplete outcome data

In Kopin 2018, 13 randomised participants (4%) were excluded from the efficacy outcome analysis and 43 (13.6%) were excluded from the safety outcome analysis. The study did not specify from which trial arm participants were missing in the final analysis. The rest of the trials analysed all randomised participants according to their allocated groups. Therefore, we assessed the risk of attrition bias as high in Kopin 2018 and low in all other included trials.

Selective reporting

All trials published a priori protocols, and reported all the outcomes prespecified in their protocols. However, two trials were post hoc sub‐group analysis and thus were subject to selection bias (Kopin 2018; Sherwood 2016). Therefore, we assessed the risk of selective outcome reporting bias as low in three trials (AUGUSTUS 2019; PIONEER AF‐PCI 2016; RE‐DUAL PCI 2017) for all assessed outcomes, and high in two trials (Kopin 2018; Sherwood 2016).

Other potential sources of bias

We assessed the risk of other bias as high in two trials as they were post hoc subgroup analysis, and thus were subject to both measured and unmeasured confounding (Kopin 2018; Sherwood 2016). We rated the risk of potential sources of bias as low for the other three included studies (AUGUSTUS 2019; PIONEER AF‐PCI 2016; RE‐DUAL PCI 2017).

Effects of interventions

See: Table 1; Table 2; Table 3; Table 4; Table 5; Table 6

Summary of findings 1. Summary of findings – death from cardiovascular causes.

Non‐vitamin K antagonist oral anticoagulants (NOACs) post‐percutaneous coronary intervention in people with an indication for anticoagulation
Patient or population: adults who underwent percutaneous coronary intervention and who have an indication for anticoagulation Settings: secondary care Intervention: NOACs (apixaban, rivaroxaban) Comparison: vitamin K antagonists (VKA – warfarin) Outcome: death from cardiovascular causes
Comparisons	Direct evidence			Indirect evidence		Network meta‐analysis		Anticipated absolute effects for NMA estimate***
Comparisons	No. of participants (studies)	RR (95% CI)	Certainty of the evidence	RR (95% CI)	Certainty of the evidence*	RR (95% CI)**	Certainty of the evidence	Risk with VKA	Risk with intervention	Risk difference with intervention
APX vs VKA Follow‐up: 0.5 to 1.8 years	4930 (2 RCTs)	1.06 (0.74 to 1.51)	⊕⊕⊕⊝ Moderate^a^,b	‐	‐	1.06 (0.41 to 2.75)	⊕⊕⊕⊝ Moderate^a^,b	21 per 1000	22 per 1000 (9 to 57)	1 more per 1000 (from 12 fewer to 36 more)
RVX (high dose, 10 mg to 20 mg) vs VKA Follow up: 1 to 2.2 years	1542 (2 RCTs)	0.69 (0.14 to 3.33)	⊝⊝⊝⊝ Very low^c^,d,e	‐	‐	0.79 (0.28 to 2.26)	⊝⊝⊝⊝ Very low^c^,d,e	21 per 1000	16 per 1000 (6 to 47)	4 fewer per 1000 (from 15 fewer to 26 more)
RVX (low dose, 2.5 mg) vs VKA Follow up: 1 year	1399 (1 RCT)	1.26 (0.57 to 2.75)	⊕⊕⊝⊝ Low^c^,f	0.06 (0.00 to 3.29)	⊝⊝⊝⊝ Very low^c^,d,e	0.94 (0.27 to 3.24)	⊕⊝⊝⊝ Very low^g	21 per 1000	2 per 1000 (6 to 68)	1 fewer per 1000 (from 15 fewer to 47 more)
The rating of the indirect estimate is based on the lowest of the two direct comparisons forming the most dominant first‐order loop (Brignardello‐Petersen 2018b). Any discrepancies between the direct estimate and NMA estimate (when there is no indirect estimate) have arisen due to differing analysis methods ‐ the pairwise meta‐analysis was conducted using Mantel‐Haenszel random effects models, while the NMA was conducted with the inverse variance method. The assumed risks in the warfarin group are based on the median baseline risks from the studies with warfarin groups in the network meta‐analysis. The corresponding risks in the apixaban, rivaroxaban and dabigatran groups (and their 95% CI) are based on the assumed risk in the warfarin group and the relative effect of the individual NOAC when compared with warfarin (and its 95% CI) derived from the network meta‐analysis. APX: apixaban; CI: confidence interval; RR: risk ratio; RVX: rivaroxaban; VKA**: vitamin K antagonist (warfarin)
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect

Open in a new tab

^aOne study included in the meta‐analysis stated that intention‐to‐treat analysis was used, but excluded participants (n for each arm not specified) from efficacy and safety analyses due to event occurring after safety censoring date. However, study only accounted for approximately 5% of weighting in the meta‐analysis. The other study was open‐label, but outcome assessors were blinded to treatment allocation and outcomes were objective. Overall bias unlikely, therefore certainty of evidence was not downgraded for methodological limitations.  ^bThe 95% CI includes no effect and includes default values for appreciable harm (i.e. CI > 1.25), appreciable benefit (i.e. CI < 0.75), or both; therefore, the certainty of evidence was downgraded by 1 level for imprecision. ^cThe 95% CI includes no effect and includes default values for appreciable harm (i.e. CI > 1.25),appreciable benefit (i.e. CI < 0.75), or both, and the optimal information size was not met (i.e. sample size < 2000 participants); therefore, the certainty of evidence was downgraded by 2 levels for imprecision.  ^dOne study in the meta‐analysis was a very small post‐hoc subgroup analysis of the patients in the ROCKET AF trial who underwent PCI during follow‐up, and thus, was at risk of significant confounding; therefore certainty of evidence was downgraded by 1 for methodological limitations. ^eThere was a degree of variability between studies of 50% to 90%, which might represent substantial heterogeneity; therefore, the certainty of evidence was downgraded by 1 level for inconsistency. ^fThough study was open‐label, outcome assessors were blinded to treatment allocation and outcomes were objective, so bias unlikely, therefore, certainty of evidence was not downgraded for methodological limitations. ^gRating was downgraded 1 level from low (the highest rating between the direct and indirect rating) due to incoherence from risk of bias in one of the direct comparisons (RVX high dose, 10 mg to 20 mg vs VKA) forming the most dominant first‐order loop

Summary of findings 2. Summary of findings – myocardial infarction.

Non‐vitamin K antagonist oral anticoagulants (NOACs) post‐percutaneous coronary intervention in people with an indication for anticoagulation
Patient or population: adults who underwent percutaneous coronary intervention and who have an indication for anticoagulation Settings: secondary care Intervention: NOACs (apixaban, dabigatran, rivaroxaban) Comparison: vitamin K antagonists (VKA – warfarin) Outcome: myocardial infarction
Comparisons	Direct evidence			Indirect evidence		Network meta‐analysis		Anticipated absolute effects for NMA estimate***
Comparisons	No. of participants (studies)	RR (95% CI)	Certainty of the evidence	RR (95% CI)	Certainty of the evidence*	RR (95% CI)**	Certainty of the evidence	Risk with VKA	Risk with intervention	Risk difference with intervention
APX vs VKA Follow up: 0.5 to 1.8 years	4930 (2 RCTs)	0.89 (0.65 to 1.20)	⊕⊕⊕⊝ Moderate^a^,b	‐	‐	0.89 (0.65 to 1.20)	⊕⊕⊕⊝ Moderate^a^,b	30 per 1000	27 per 1000 (20 to 36)	30 fewer 1000 (from 11 fewer to 6 more)
RVX (high dose, 10 mg to 20 mg) vs VKA Follow up: 1 to 2.2 years	1542 (2 RCTs)	0.94 (0.53 to 1.65)	⊕⊝⊝⊝ Very low^c^,d	‐	‐	0.94 (0.53 to 1.65)	⊕⊝⊝⊝ Very low^c^,d	30 per 1000	29 per 1000 (16 to 50)	20 fewer per 1000 (from 14 fewer to 20 more)
RVX (low dose, 2.5 mg) vs VKA Follow up: 1 year	1399 (1 RCT)	0.80 (0.43 to 1.50)	⊕⊕⊝⊝ Low^c^,e	1.31 (0.04 to 41.78)	⊕⊝⊝⊝ Very low^c.d	0.81 (0.44 to 1.51)	⊕⊕⊝⊝ Low^f	30 per 1000	25 per 1000 (13 to 46)	6 fewer per 1000 (from 17 fewer to 15 more)
DAB (low dose, 110 mg) vs VKA Follow up: 14 months	1962 (1 RCT)	1.52 (0.96 to 2.40)	⊕⊕⊝⊝ Low^c^,e	‐	‐	1.52 (0.96 to 2.40)	⊕⊕⊝⊝ Low^c^,e	30 per 1000	46 per 1000 (29 to 73)	16 more per 1000 (from 1 fewer to 42 more)
DAB (high dose, 150 mg) vs VKA Follow up: 14 months	1527 (1 RCT)	1.18 (0.68 to 2.07)	⊕⊕⊝⊝ Low^c^,e	‐	‐	1.15 (0.68 to 1.94)	⊕⊕⊝⊝ Low^c^,e	30 per 1000	36 per 1000 (21 to 63)	5 more per 1000 (from 10 fewer to 32 more)
The rating of the indirect estimate is based on the lowest of the two direct comparisons forming the most dominant first‐order loop (Brignardello‐Petersen 2018b). Any discrepancies between the direct estimate and NMA estimate (when there is no indirect estimate) have arisen due to differing analysis methods ‐ the pairwise meta‐analysis was conducted using Mantel‐Haenszel random effects models, while the NMA was conducted with the inverse variance method. The assumed risks in the warfarin group are based on the median baseline risks from the studies with warfarin groups in the network meta‐analysis. The corresponding risks in the apixaban, rivaroxaban and dabigatran groups (and their 95% CI) are based on the assumed risk in the warfarin group and the relative effect of the individual NOAC when compared with warfarin (and its 95% CI) derived from the network meta‐analysis. APX: apixaban; CI: confidence interval; DAB: dabigatran; RR: risk ratio; RVX: rivaroxaban; VKA**: vitamin K antagonist (warfarin)
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect

Open in a new tab

^aOne study included in the meta‐analysis stated that intention‐to‐treat analysis was used, but excluded patients (n for each arm not specified) from efficacy and safety analyses due to event occurring after safety censoring date. However, study only accounted for approximately 5% of weighting in the meta‐analysis. The other study was open‐label, but outcome assessors were blinded to treatment allocation and outcomes were objective. Overall bias unlikely, therefore, certainty of evidence was not downgraded for methodological limitations. ^bThe 95% CI includes no effect, and includes default values for appreciable harm (i.e. CI > 1.25), appreciable benefit (i.e. CI < 0.75), or both; therefore, the certainty of evidence was downgraded by 1 level for imprecision. ^cThe 95% CI includes no effect, and includes default values for appreciable harm (i.e. CI > 1.25), appreciable benefit (i.e. CI < 0.75), or both, and the optimal information size was not met (i.e. sample size < 2000 participants); therefore, the certainty of evidence was downgraded by 2 levels for imprecision. ^dOne study in the meta‐analysis was a very small post‐hoc subgroup analysis of the patients in the ROCKET AF trial who underwent PCI during follow‐up, and thus, was at risk of significant confounding; therefore, certainty of evidence was downgraded by 1 level for imprecision. ^eThough study was open‐label, outcome assessors were blinded to treatment allocation and outcomes were objective, therefore certainty of evidence was not downgraded for methodological limitations. ^fAs per GRADE guidance, if there is incoherence between the direct and indirect evidence, but the dominant estimate is similar to the network estimate, the certainty of the network estimate should not be rated down for incoherence (Brignardello‐Petersen 2019).

Summary of findings 3. Summary of findings – stroke.

Non‐vitamin K antagonist oral anticoagulants (NOACs) post‐percutaneous coronary intervention in people with an indication for anticoagulation
Patient or population: adults who underwent percutaneous coronary intervention and who have an indication for anticoagulation Settings: secondary care Intervention: NOACs (apixaban, dabigatran, rivaroxaban) Comparison: vitamin K antagonist (VKA – warfarin) Outcome: stroke
Comparisons	Direct evidence			Indirect evidence		Network meta‐analysis		Anticipated absolute effects for NMA estimate***
Comparisons	No. of participants (studies)	RR (95% CI)	Certainty of the evidence	RR (95% CI)	Certainty of the evidence*	RR (95% CI)**	Certainty of the evidence	Risk with VKA	Risk with intervention	Risk difference with intervention
APX vs VKA Follow up: 0.5 to 1.8 year	4930 (2 RCTs)	0.68 (0.27 to 1.76)	⊕⊕⊝⊝ Low^a^,b,d	‐	‐	0.65 (0.29 to 1.45)	⊕⊕⊝⊝ Low^a^,b,d	12 per 1000	8 per 1000 (4 to 18)	4 fewer per 1000 (from 9 fewer to 6 more)
RVX (high dose, 10 mg to 20 mg) vs VKA Follow up: 1 to 2.2 years	1546 (2 RCTs)	0.94 (0.38 to 2.34)	⊕⊝⊝⊝ Very low^c^,e	‐	‐	0.90 (0.31 to 2.58)	⊕⊝⊝⊝ Very low^c^,e	12 per 1000	11 per 1000 (4 to 32)	1 fewer per 1000 (from 8 fewer to 19 more)
RVX (low dose, 2.5 mg) vs VKA Follow up: 1 year	1399 (1 RCT)	1.41 (0.54 to 3.68)	⊕⊕⊝⊝ Low^c^,f	0.20 (0.00 to 15.96)	⊕⊝⊝⊝ Very low^c^,e	1.24 (0.41 to 3.79)	⊕⊕⊝⊝ Low^g	12 per 1000	15 per 1000 (5 to 46)	3 more per 1000 (from 7 fewer to 34 more)
DAB (low dose, 110 mg) vs VKA Follow up: 14 months	1962 (1 RCT)	1.31 (0.64 to 2.68)	⊕⊕⊝⊝ Low^c^,f	‐	‐	1.31 (0.50 to 3.42)	⊕⊕⊝⊝ Low^c^,f	12 per 1000	16 per 1000 (6 to 42)	4 more per 1000 (from 6 fewer to 30 more)
DAB (high dose, 150 mg) vs VKA Follow up: 14 months	1527 (1 RCT)	1.13 (0.44 to 2.90)	⊕⊕⊝⊝ Low^c^,f	‐	‐	0.89 (0.31 to 2.57)	⊕⊕⊝⊝ Low^c^,f	12 per 1000	11 per 1000 (4 to 32)	1 fewer per 1000 (from 8 fewer to 19 more)
The rating of the indirect estimate is based on the lowest of the two direct comparisons forming the most dominant first‐order loop (Brignardello‐Petersen 2018b). Any discrepancies between the direct estimate and NMA estimate (when there is no indirect estimate) have arisen due to differing analysis methods – the pairwise meta‐analysis was conducted using the Mantel‐Haenszel random‐effects model, while the NMA was conducted with the inverse variance method. The assumed risks in the warfarin group are based on the median baseline risks from the studies with warfarin groups in the network meta‐analysis. The corresponding risks in the apixaban, rivaroxaban and dabigatran groups (and their 95% CI) are based on the assumed risk in the warfarin group and the relative effect of the individual NOAC when compared with warfarin (and its 95% CI), derived from the network meta‐analysis. APX: apixaban; CI: confidence interval; DAB: dabigatran; RR: risk ratio; RVX: rivaroxaban; VKA**: vitamin K antagonist (warfarin)
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect

Open in a new tab

^aOne study included in the meta‐analysis stated that intention‐to‐treat analysis was used, but excluded patients (n for each arm not specified) from efficacy and safety analyses due to event occurring after safety censoring date. However, the study only accounted for approximately 5% of weighting in the meta‐analysis. The other study was open‐label, but outcome assessors were blinded to treatment allocation and outcomes were objective. Overall bias unlikely, therefore certainty of evidence was not downgraded for methodological limitations. ^bThe 95% CI includes no effect, and includes default values for appreciable harm (i.e. CI > 1.25), appreciable benefit (i.e. CI < 0.75), or both; therefore, the certainty of evidence was downgraded by 1 level for imprecision. ^cThe 95% CI includes no effect and includes default values for appreciable harm (i.e. CI > 1.25), appreciable benefit (i.e. CI < 0.75), or both, and the optimal information size was not met (i.e. sample size < 2000 participants); therefore, the certainty of evidence was downgraded by 2 levels for imprecision. ^dThere was a degree of variability between studies of 30% to 60%, which might represent moderate heterogeneity, therefore, certainty of evidence was downgraded by 1 level for inconsistency. ^eOne study in the meta‐analysis was a very small post‐hoc subgroup analysis of the patients in the ROCKET AF trial who underwent PCI during follow‐up, and thus, was at risk of significant confounding; therefore, certainty of evidence was downgraded by 1 for methodological limitations. ^fThough study was open‐label, outcome assessors were blinded to treatment allocation and outcomes were objective, therefore, certainty of evidence was not downgraded for methodological limitations.  ^gAs per GRADE guidance, if there is incoherence between the direct and indirect evidence, but the dominant estimate is similar to the network estimate, the certainty of the network estimate should not be rated down for incoherence (Brignardello‐Petersen 2019).

Summary of findings 4. Summary of findings – major bleeding.

Non‐vitamin K antagonist oral anticoagulants (NOACs) post‐percutaneous coronary intervention in people with an indication for anticoagulation
Patient or population: adults who underwent percutaneous coronary intervention and who have an indication for anticoagulation Settings: secondary care Intervention: NOACs (apixaban, dabigatran, rivaroxaban) Comparison: vitamin K antagonist (VKA – warfarin) Outcome: major bleeding as defined by the Thrombolysis In Myocardial Infarction (TIMI) criteria
Comparisons	Direct evidence			Indirect evidence		Network meta‐analysis		Anticipated absolute effects for NMA estimate***
Comparisons	No. of participants (studies)	RR (95% CI)	Certainty of the evidence	RR (95% CI)	Certainty of the evidence*	RR (95% CI)**	Certainty of the evidence	Risk with VKA	Risk with intervention	Risk difference with intervention
APX vs VKA Follow up: 0.5 to 1.8 years	4865 (2 RCTs)	0.82 (0.55 to 1.21)	⊕⊕⊕⊝ Moderate^a^,b	‐	‐	0.82 (0.55 to 1.21)	⊕⊕⊕⊝ Moderate^a^,b	38 per 1000	31 per 1000 (21 to 46)	7 fewer per 1000 (from 17 fewer to 80 more)
RVX (high dose, 10 mg to 20 mg) vs VKA Follow up: 1 to 2.2 years	1546 (2 RCTs)	0.91 (0.45 to 1.85)	⊕⊝⊝⊝ Very low^c^,d	‐	‐	0.87 (0.49 to 1.54)	⊕⊝⊝⊝ Very low^c^,d	38 per 1000	33 per 1000 (18 to 58)	5 fewer per 1000 (from 19 fewer to 20 more)
RVX (low dose, 2.5 mg) vs VKA Follow up: 1 year	1403 (1 RCT)	0.59 (0.29 to 1.20)	⊕⊕⊝⊝ Low^c.e	4.64 (0.16 to 133.28)	⊕⊝⊝⊝ Very low^c^,d	0.65 (0.32 to 1.29)	⊕⊕⊝⊝ Low^f	38 per 1000	24 per 1000 (12 to 49)	13 fewer per 1000 (from 26 fewer to 11 more)
DAB (low dose, 110 mg) vs VKA Follow up: 14 months	1962 (1 RCT)	0.38 (0.21 to 0.70)	⊕⊕⊕⊝ Moderate^e^,g	‐	‐	0.38 (0.21 to 0.70)	⊕⊕⊕⊝ Moderate^e^,g	38 per 1000	14 per 1000 (8 to 26)	23 fewer per 1000 (from 30 fewer to 11 fewer)
DAB (high dose, 150 mg) vs VKA Follow up: 14 months	1527 (1 RCT)	0.53 (0.29 to 0.97)	⊕⊕⊝⊝ Low^c^,e	‐	‐	0.56 (0.31 to 0.99)	⊕⊕⊝⊝ Low^c^,e	38 per 1000	21 per 1000 (12 to 37)	17 fewer per 1000 (from 26 fewer to 0 fewer)
The rating of the indirect estimate is based on the lowest of the two direct comparisons forming the most dominant first‐order loop (Brignardello‐Petersen 2018b). Any discrepancies between the direct estimate and NMA estimate (when there is no indirect estimate) have arisen due to differing analysis methods – the pairwise meta‐analysis was conducted using the Mantel‐Haenszel random‐effects model, while the NMA was conducted with the inverse variance method. The assumed risks in the warfarin groups are based on the median baseline risks from the studies with warfarin groups in the network meta‐analysis. The corresponding risks in the apixaban, rivaroxaban, and dabigatran groups (and their 95% CI) are based on the assumed risk in the warfarin group and the relative effect of the individual NOAC when compared with warfarin (and its 95% CI) derived from the network meta‐analysis. APX: apixaban; CI: confidence interval; DAB: dabigatran; RR: risk ratio; RVX: rivaroxaban; VKA**: vitamin K antagonist (warfarin)
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect

Open in a new tab

^aOne study included in the meta‐analysis stated that intention‐to‐treat analysis was used, but excluded patients (n for each arm not specified) from efficacy and safety analyses due to event occurring after safety censoring date. However, the study only accounted for approximately 5% of weighting in the meta‐analysis. The other study was open‐label, but outcome assessors were blinded to treatment allocation and outcomes were objective. Overall bias unlikely, therefore, certainty of evidence was not downgraded for methodological limitations. ^bThe 95% CI includes no effect, and includes default values for appreciable harm (i.e. CI > 1.25), appreciable benefit (i.e. CI < 0.75), or both; therefore, the certainty of evidence was downgraded by 1 level for imprecision. ^cThe 95% CI includes no effect, and includes default values for appreciable harm (i.e. CI > 1.25), appreciable benefit (i.e. CI < 0.75), or both, and the optimal information size was not met (i.e. sample size < 2000 participants); therefore, the certainty of evidence was downgraded by 2 levels for imprecision. ^dOne study in the meta‐analysis was a very small post‐hoc subgroup analysis of the patients in the ROCKET AF trial who underwent PCI during follow‐up, and thus, was at risk of significant confounding; therefore, certainty of evidence was downgraded by 1 level for methodological limitations. ^eThough study was open‐label, outcome assessors were blinded to treatment allocation and outcomes were objective, therefore, certainty of evidence was not downgraded for methodological limitations. ^fAs per GRADE guidance, if there is incoherence between the direct and indirect evidence, but the dominant estimate is similar to the network estimate, the certainty of the network estimate should not be rated down for incoherence (Brignardello‐Petersen 2019). ^gThe optimal information size was not met (i.e. sample size < 2000 participants); therefore, the certainty of evidence was downgraded by 1 level for imprecision.

Summary of findings 5. Summary of findings – death from any cause.

Non‐vitamin K antagonist oral anticoagulants (NOACs) post‐percutaneous coronary intervention in people with an indication for anticoagulation
Patient or population: adults who underwent percutaneous coronary intervention and who have an indication for anticoagulation Settings: secondary care Intervention: NOACs (apixaban, dabigatran, rivaroxaban) Comparison: vitamin K antagonist (VKA – warfarin) Outcome: death from any cause
Comparisons	Direct evidence			Indirect evidence		Network meta‐analysis		Anticipated absolute effects for NMA estimate**
Comparisons	No. of participants (studies)	RR (95% CI)	Certainty of the evidence	RR (95% CI)	Certainty of the evidence	RR (95% CI)*	Certainty of the evidence	Risk with VKA	Risk with intervention	Risk difference with intervention
APX vs VKA Follow up: 0.5 to 1.8 years	4930 (2 RCTs)	1.18 (0.74 to 1.87)	⊕⊕⊕⊝ Moderate^a^,b	‐	‐	1.18 (0.74 to 1.87)	⊕⊕⊕⊝ Moderate^a^,b	43 per 1000	5 per 1000 (32 to 80)	8 more per 1000 (from 11 fewer to 37 more)
RVX (high dose, 10 mg to 20 mg) vs VKA Follow up: 1 year	1393 (1 RCT)	1.23 (0.60 to 2.54)	⊕⊕⊝⊝ Low^c^,d	‐	‐	1.23 (0.53 to 2.84)	⊕⊕⊝⊝ Low^c^,d	43 per 1000	53 per 1000 (23 to 121)	10 more per 1000 (from 20 fewer to 79 more)
RVX (low dose, 2.5 mg) vs VKA Follow up: 1 year	1403 (1 RCT)	1.29 (0.63 to 2.64)	⊕⊕⊝⊝ Low^c^,d	‐	‐	1.29 (0.56 to 2.95)	⊕⊕⊝⊝ Low^c^,d	43 per 1000	55 per 1000 (24 to 126)	12 more per 1000 (from 19 fewer to 83 more)
DAB (low dose, 110 mg) vs VKA Follow up: 14 months	1962 (1 RCT)	1.15 (0.79 to 1.67)	⊕⊕⊝⊝ Low^c^,d	‐	‐	1.15 (0.65 to 2.01)	⊕⊕⊝⊝ Low^c^,d	43 per 1000	49 per 1000 (28 to 86)	6 more per 1000 (from 15 fewer to 43 more)
DAB (high dose, 150 mg) vs VKA Follow up: 14 months	1527 (1 RCT)	0.86 (0.53 to 1.38)	⊕⊕⊝⊝ Low^c^,d	‐	‐	0.80 (0.44 to 1.48)	⊕⊕⊝⊝ Low^c^,d	43 per 1000	34 per 1000 (19 to 36)	8 fewer per 1000 (from 24 fewer to 21 more)
Any discrepancies between the direct estimate and NMA estimate (when there is no indirect estimate) have arisen due to differing analysis methods – the pairwise meta‐analysis was conducted using the Mantel‐Haenszel random‐effects model, while the NMA was conducted with the inverse variance method. The assumed risks in the warfarin group are based on the median baseline risks from the studies with warfarin groups in the network meta‐analysis. The corresponding risks in the apixaban, rivaroxaban, and dabigatran groups (and their 95% CI) are based on the assumed risk in the warfarin group and the relative effect of the individual NOAC when compared with warfarin (and its 95% CI) derived from the network meta‐analysis. APX: apixaban; CI: confidence interval; DAB: dabigatran; RR: risk ratio; RVX: rivaroxaban; VKA*: vitamin K antagonist (warfarin)
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect

Open in a new tab

Summary of findings 6. Summary of findings – stent thrombosis.

Non‐vitamin K antagonist oral anticoagulants (NOACs) post‐percutaneous coronary intervention in people with an indication for anticoagulation
Patient or population: adults who underwent percutaneous coronary intervention and who have an indication for anticoagulation Settings: secondary care Intervention: NOACs (apixaban, dabigatran, rivaroxaban) Comparison: vitamin K antagonists (VKA – warfarin) Outcome: stent thrombosis
Comparisons	Direct evidence			Indirect evidence		Network meta‐analysis		Anticipated absolute effects for NMA estimate**
Comparisons	No. of participants (studies)	RR (95% CI)	Certainty of the evidence	RR (95% CI)	Certainty of the evidence	RR (95% CI)*	Certainty of the evidence	Risk with VKA	Risk with intervention	Risk difference with intervention
APX vs VKA Follow up: 6 months	4614 (1 RCT)	0.78 (0.39 to 1.56)	⊕⊕⊕⊝ Moderate^a^,b	‐	‐	0.78 (0.39 to 1.56)	⊕⊕⊕⊝ Moderate^a^,b	8 per 1000	6 per 1000 (3 to 12)	2 fewer per 1000 (from 5 fewer to 4 more)
RVX (high dose, 10 mg to 20 mg) vs VKA Follow up: 1 year	1389 (1 RCT)	1.25 (0.34 to 4.64)	⊕⊕⊝⊝ Low^c^,d	‐	‐	1.25 (0.34 to 4.64)	⊕⊕⊝⊝ Low^c^,d	8 per 1000	1 per 1000 (3 to 37)	2 more per 1000 (from 5 fewer to 29 more)
RVX (low dose, 2.5 mg) vs VKA Follow up: 1 year	1399 (1 RCT)	1.48 (0.42 to 5.22)	⊕⊕⊝⊝ Low^c^,d	‐	‐	1.48 (0.42 to 5.22)	⊕⊕⊝⊝ Low^c^,d	8 per 1000	12 per 1000 (3 to 42)	4 more per 1000 (from 5 fewer to 34 more)
DAB (low dose, 110 mg) vs VKA Follow up: 14 months	1962 (1 RCT)	1.88 (0.80 to 4.40)	⊕⊕⊝⊝ Low^c^,d	‐	‐	1.88 (0.80 to 4.40)	⊕⊕⊝⊝ Low^c^,d	8 per 1000	15 per 1000 (6 to 35)	7 more per 1000 (from 2 fewer to 27 more)
DAB (high dose, 150 mg) vs VKA Follow up: 14 months	1527 (1 RCT)	1.00 (0.35 to 2.84)	⊕⊕⊝⊝ Low^c^,d	‐	‐	1.13 (0.41 to 3.09)	⊕⊕⊝⊝ Low^c^,d	8 per 1000	9 per 1000 (3 to 25)	1 more per 1000 (from 5 fewer to 17 more)
Any discrepancies between the direct estimate and NMA estimate (when there is no indirect estimate) have arisen due to differing analysis methods – the pairwise meta‐analysis was conducted using the Mantel‐Haenszel random‐effects model, while the NMA was conducted with the inverse variance method. The assumed risks in the warfarin group are based on the median baseline risks from the studies with warfarin groups in the network meta‐analysis. The corresponding risks in the apixaban, rivaroxaban, and dabigatran groups (and their 95% CI) are based on the assumed risk in the warfarin group and the relative effect of the individual NOAC when compared with warfarin (and its 95% CI) derived from the network meta‐analysis. APX: apixaban; CI: confidence interval; DAB: dabigatran; RR: risk ratio; RVX: rivaroxaban; VKA*: vitamin K antagonist (warfarin)
GRADE Working Group grades of evidence High certainty: We are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect Very low certainty: We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect

Open in a new tab

^aThough study was open‐label, outcome assessors were blinded to treatment allocation and outcomes were objective; therefore, certainty of evidence was not downgraded for methodological limitations. ^bThe 95% CI includes no effect, and includes default values for appreciable harm (i.e. CI > 1.25), appreciable benefit (i.e. CI < 0.75), or both; therefore, the certainty of evidence was downgraded by 1 level. ^cThough study was open‐label, outcome assessors were blinded to treatment allocation and outcomes were objective. Study arms received different treatment regimens, but bias unlikely, therefore, certainty of evidence was not downgraded for methodological limitations. ^dThe 95% CI includes no effect, and includes default values for appreciable harm (i.e. CI > 1.25), appreciable benefit (i.e. CI < 0.75), or both; the optimal information size was not met (i.e. sample size < 2000 participants); therefore, the certainty of evidence was downgraded by 2 levels for imprecision.

Death from cardiovascular events

Direct evidence (NOACs vs vitamin K antagonists)

Two RCTs compared apixaban with vitamin K antagonists (AUGUSTUS 2019; Kopin 2018). There is probably little or no difference in the rate of death from cardiovascular events between apixaban and vitamin K antagonists (risk ratio (RR) 1.06, 95% confidence interval (CI) 0.74 to 1.51; 2 studies, 4930 participants; I² = 0%; moderate‐certainty evidence; Analysis 1.1).

1.1 — Comparison 1: Apixaban vs warfarin, Outcome 1: Death from cardiovascular causes

Two RCTs compared rivaroxaban to vitamin K antagonists (PIONEER AF‐PCI 2016; Sherwood 2016). The PIONEER AF‐PCI trial investigated two dosing regimens of rivaroxaban versus vitamin K antagonists. There may be little or no difference in the rate of death from cardiovascular events between high and low doses of rivaroxaban (RR 1.09, 95% CI 0.53 to 2.23; 1 study, 1398 participants; low‐certainty evidence; Analysis 2.1); between high dose rivaroxaban and vitamin K antagonists (RR 0.69, 95% CI 0.14 to 3.33; 2 studies, 1542 participants; I² = 73%; very‐low certainty evidence; Analysis 3.1); or between low dose rivaroxaban and vitamin K antagonists (RR 1.26, 95% CI 0.57 to 2.75; 1 study, 1399 participants; low‐certainty evidence; Analysis 4.1).

2.1 — Comparison 2: High dose rivaroxaban vs low dose rivaroxaban, Outcome 1: Death from cardiovascular causes

3.1 — Comparison 3: Rivaroxaban (high dose) vs warfarin, Outcome 1: Death from cardiovascular causes

4.1 — Comparison 4: Rivaroxaban (low dose) vs warfarin, Outcome 1: Death from cardiovascular causes

No trial compared death from cardiovascular event between dabigatran and vitamin K antagonists.

Network meta‐analysis (NOACs compared to each other)

Low‐certainty evidence suggests little or no difference in the rate of death from cardiovascular events between high and low doses of rivaroxaban (RR 0.84, 95% CI 0.25 to 2.81); between apixaban and high dose rivaroxaban (RR 1.34, 95% CI 0.33 to 5.53), or between apixaban and low dose rivaroxaban (RR 1.13, 95% CI 0.24 to 5.36; Figure 1; Table 7; Table 1).

1. League table – death from cardiovascular causes.

Pairwise meta‐analysis
Vitamin K antagonist	1.06 (0.74 to 1.51)	0.69 (0.14 to 3.33)	1.26 (0.57 to 2.75)
1.06 (0.41 to 2.75)	Apixaban	‐	‐
0.79 (0.28 to 2.26)	1.34 (0.33 to 5.53)	Rivaroxaban (high dose)	1.09 (0.53 to 2.23)
0.94 (0.27 to 3.24)	1.13 (0.24 to 5.36)	0.84 (0.25 to 2.82)	(Rivaroxaban low dose)
Network meta‐analysis
Comparisons between treatments should be read from left to right and the estimate is in the cell in common between the column‐defining treatment and the row‐defining treatment. The upper triangle presents the results from direct (pairwise) meta‐analyses and the lower triangle presents the results from the network meta‐analyses. Results are presented as RR (95% CI), where a RR of < 1 favours the row‐defined treatment. The order of treatments in the diagonal is arbitrary and does not reflect ranking. CI: confidence interval; RR: risk ratio

Open in a new tab

Myocardial infarction

Direct evidence (NOACs vs vitamin K antagonists)

Two RCTs compared apixaban with vitamin K antagonists (AUGUSTUS 2019; Kopin 2018). There is probably little or no difference in the rate of myocardial infarction between apixaban and vitamin K antagonists (RR 0.89, 95% CI 0.65 to 1.20; 2 studies, 4930 participants; I² = 0%; moderate‐quality evidence; Analysis 1.2).

1.2 — Comparison 1: Apixaban vs warfarin, Outcome 2: Myocardial infarction

Two RCTs compared rivaroxaban to vitamin K antagonists (PIONEER AF‐PCI 2016; Sherwood 2016). One RCT compared high to low dose of rivaroxaban (PIONEER AF‐PCI 2016). There may be little or no difference in the rate of myocardial infarction between high and low doses of rivaroxaban (RR 1.13, 95% CI 0.59 to 2.16; 1 study, 1398 participants; low‐certainty evidence; Analysis 2.2); between high dose rivaroxaban and vitamin K antagonists (RR 0.94, 95% CI 0.53 to 1.64; 2 studies, 1542 participants; I² = 0%; very low‐certainty evidence; Analysis 3.2); or between low dose rivaroxaban and vitamin K antagonists (RR 0.80, 95% CI 0.43 to 1.50; 1 study, 1399 participants; low‐certainty evidence; Analysis 4.2).

2.2 — Comparison 2: High dose rivaroxaban vs low dose rivaroxaban, Outcome 2: Myocardial infarction

3.2 — Comparison 3: Rivaroxaban (high dose) vs warfarin, Outcome 2: Myocardial infarction

4.2 — Comparison 4: Rivaroxaban (low dose) vs warfarin, Outcome 2: Myocardial infarction

One RCT compared high and low doses of dabigatran and vitamin K antagonists (RE‐DUAL PCI 2017). There may be little or no difference in the rate of myocardial infarction between high and low dose dabigatran (RR 1.32, 95% CI 0.82 to 2.12; 1 study, 1744 participants; low‐certainty evidence; Analysis 5.1); high dose dabigatran and vitamin K antagonists (RR 1.18, 95% CI 0.68 to 2.07; 1 study, 1527 participants; low‐certainty evidence; Analysis 6.1); or low dose dabigatran and vitamin K antagonists (RR 1.52, 95% CI 0.96 to 2.40; 1 study, 1962 participants; low‐certainty evidence; Analysis 7.1).

5.1 — Comparison 5: Low dose dabigatran vs high dose dabigatran, Outcome 1: Myocardial infarction

6.1 — Comparison 6: Dabigatran (high dose) vs warfarin, Outcome 1: Myocardial infarction

7.1 — Comparison 7: Dabigatran (low dose) vs warfarin, Outcome 1: Myocardial infarction

Network meta‐analysis (NOACs compared to each other)

Low‐certainty evidence suggests little or no difference in the rate of myocardial infarction between apixaban and high dose rivaroxaban (RR 0.95, 95% CI 0.50 to 1.80), or between apixaban and low dose rivaroxaban (RR 1.09, 95% CI 0.55 to 2.18; Figure 2).

Low‐certainty evidence suggests little or no difference between high dose rivaroxaban and high dose dabigatran (RR 0.81, 95% CI 0.37 to 1.75), or high dose rivaroxaban and low dose dabigatran (RR 0.62, 95% C 0.30 to 1.28); low dose rivaroxaban and high dose dabigatran (RR 0.70, 95% CI 0.31 to 1.58), or low dose rivaroxaban and low dose dabigatran (RR 0.54, 95% CI 0.25 to 1.16; Figure 2)

Low‐certainty evidence suggests little or no difference in the rate of infarction between high and low dose rivaroxaban (RR 1.15, 95% CI 0.61 to 2.17), or between high and low dose dabigatran (RR 1.32, 95% CI 0.82 to 2.12; Figure 2; Table 8; Table 2).

2. League table – myocardial infarction.

Pairwise meta‐analysis
Vitamin K antagonist	0.89 (0.65 to 1.20)	0.94 (0.53 to 1.64)	0.80 (0.43 to 1.50)	1.52 (0.96 to 2.40)	1.18 (0.68 to 2.07)
0.89 (0.65 to 1.20)	Apixaban	‐	‐	‐	‐
0.94 (0.53 to 1.65)	0.95 (0.50 to 1.80)	Rivaroxaban (high dose)	1.13 (0.59 to 2.16)	‐	‐
0.81 (0.44 to 1.51)	1.09 (0.55 to 2.18)	1.15 (0.61 to 2.17)	Rivaroxaban (low dose)	‐	‐
1.52 (0.96 to 2.40)	1.71 (0.98 to 2.97)	0.62 (0.30 to1.28)	0.54 (0.25 to 1.16)	Dabigatran (low dose)	1.32 (0.82 to 2.12)
1.15 (0.68 to 1.94)	1.30 (0.71 to 2.38)	0.81 (0.38 to 1.75)	0.70 (0.31 to 1.58)	1.32 (0.82 to 2.12)	Dabigatran (high dose)
Network meta‐analysis
Comparisons between treatments should be read from left to right and the estimate is in the cell in common between the column‐defining treatment and the row‐defining treatment. The upper triangle presents the results from direct (pairwise) meta‐analyses and the lower triangle presents the results from the network meta‐analyses. Results are presented as RR (95% CI), where a RR of < 1 favours the row‐defined treatment. The order of treatments in the diagonal is arbitrary and does not reflect ranking. CI: confidence interval; RR: risk ratio

Open in a new tab

Stroke

Direct evidence (NOACs vs vitamin K antagonists)

Two RCTs compared apixaban with vitamin K antagonists (AUGUSTUS 2019; Kopin 2018). There may be little or no difference in the rate of stroke between apixaban and vitamin K antagonists (RR 0.68, 95% CI 0.27 to 1.76; 2 studies, 4930 participants; I² = 39%; low‐certainty evidence; Analysis 1.3).

1.3 — Comparison 1: Apixaban vs warfarin, Outcome 3: Stroke

Two RCTs compared rivaroxaban to vitamin K antagonists (PIONEER AF‐PCI 2016, Sherwood 2016). One RCT compared high to low dose of rivaroxaban (PIONEER AF‐PCI 2016). There may be little or no difference in the rate of stroke between high and low doses of rivaroxaban (RR 0.81, 95% CI 0.32 to 2.04; 1 study, 1398 participants; low‐certainty evidence; Analysis 2.3); between high dose rivaroxaban and vitamin K antagonists (RR 0.94, 95% CI 0.38 to 2.34; 2 studies, 1542 participants; I² = 0%; very low‐certainty evidence; Analysis 3.3); or between low dose rivaroxaban and vitamin K antagonists (RR 1.41, 95% CI 0.54 to 3.68; 1 study, 1399 participants; low‐certainty evidence; Analysis 4.3).

2.3 — Comparison 2: High dose rivaroxaban vs low dose rivaroxaban, Outcome 3: Stroke

3.3 — Comparison 3: Rivaroxaban (high dose) vs warfarin, Outcome 3: Stroke

4.3 — Comparison 4: Rivaroxaban (low dose) vs warfarin, Outcome 3: Stroke

One RCT compared high and low doses of dabigatran with each other, and with vitamin K antagonists (RE‐DUAL PCI 2017). There may be little or no difference in the rate of stroke between low and high doses of dabigatran (RR 1.47, 95% CI 0.66 to 3.28; 1 study, 1744 participants; low‐certainty evidence; Analysis 5.2); between high dose dabigatran and vitamin K antagonists (RR 1.13, 95% CI 0.44 to 2.90; 1 study, 1527 participants; low‐certainty evidence; Analysis 6.2); or between low dose dabigatran and vitamin K antagonists (RR 1.31, 95% CI 0.64 to 2.68; 1 study, 1962 participants; low‐certainty evidence; Analysis 7.2).

5.2 — Comparison 5: Low dose dabigatran vs high dose dabigatran, Outcome 2: Stroke

6.2 — Comparison 6: Dabigatran (high dose) vs warfarin, Outcome 2: Stroke

7.2 — Comparison 7: Dabigatran (low dose) vs warfarin, Outcome 2: Stroke

Network meta‐analysis (NOACs compared to each other)

Low‐certainty evidence suggests little or no difference in the rate of stroke between apixaban and high dose rivaroxaban (RR 0.72, 95% CI 0.19 to 2.72), or between apixaban and low dose rivaroxaban (RR 0.52, 95% CI 0.13 to 2.07; Figure 3).

Low‐certainty evidence suggests little or no difference in the rate of stroke between high dose rivaroxaban and high dose dabigatran (RR 1.01, 95% CI 0.23 to 4.51); high dose rivaroxaban and low dose dabigatran (RR 0.69, 95% CI 0.16 to 2.87); low dose rivaroxaban and high dose dabigatran (RR 1.40, 95% CI 0.30 to 6.51); or low dose rivaroxaban and low dose dabigatran (RR 0.95, 95% CI 0.22 to 4.15; Figure 3).

Low‐certainty evidence suggests little or no difference in the rate of stroke between high and low dose rivaroxaban (RR 0.72, 95% CI 0.24 to 2.16), or high and low dose dabigatran (RR 1.47, 95% CI 0.53 to 4.10; Figure 3; Table 9; Table 3).

3. League table – stroke.

Pairwise meta‐analysis
Vitamin K antagonist	0.68 (0.27 to 1.76)	0.94 (0.38 to 2.34)	1.41 (0.54 to 3.68)	1.31 (0.64 to 2.68)	1.13 (0.44 to 2.90)
0.65 (0.29 to 1.45)	Apixaban	‐	‐	‐	‐
0.90 (0.31 to 2.58)	0.72 (0.19 to 2.72)	Rivaroxaban (high dose)	0.81 (0.32 to 2.04)	‐	‐
1.24 (0.41 to 3.79)	0.52 (0.13 to 2.07)	0.72 (0.24 to 2.16)	Rivaroxaban (low dose)	‐	‐
1.31 (0.50 to 3.42)	2.01 (0.58 to 7.04)	0.69 (0.16 to 2.87)	0.95 (0.22 to 4.15)	Dabigatran (low dose)	1.47 (0.66 to 3.28)
0.89 (0.31 to 2.57)	1.37 (0.36 to 5.18)	1.01 (0.23 to 4.51)	1.40 (0.30 to 6.51)	1.47 (0.53 to 4.10)	Dabigatran (high dose)
Network meta‐analysis
Comparisons between treatments should be read from left to right and the estimate is in the cell in common between the column‐defining treatment and the row‐defining treatment. The upper triangle presents the results from direct (pairwise) meta‐analyses and the lower triangle presents the results from the network meta‐analyses. Results are presented as RR (95% CI), where a RR of < 1 favours the row‐defined treatment. The order of treatments in the diagonal is arbitrary and does not reflect ranking. CI: confidence interval; RR: risk ratio

Open in a new tab

Major TIMI bleeding

Direct evidence (NOACs vs vitamin K antagonists)

Two RCTs compared apixaban with vitamin K antagonists (AUGUSTUS 2019; Kopin 2018). There is probably little or no difference in the rate of major bleeding between apixaban and vitamin K antagonists (RR 0.82, 95% CI 0.55 to 1.21; 2 studies, 4865 participants; I² = 0%; moderate‐certainty evidence; Analysis 1.4).

1.4 — Comparison 1: Apixaban vs warfarin, Outcome 4: Major Bleeding

Two RCTs compared rivaroxaban to vitamin K antagonists (PIONEER AF‐PCI 2016, Sherwood 2016). One RCT compared high to low dose rivaroxaban (PIONEER AF‐PCI 2016).

There may be little or no difference in the rate of major bleeding between high and low doses of rivaroxaban (RR 1.18, 95% CI 0.55 to 2.54; 1 study, 1402 participants; low‐certainty evidence; Analysis 2.4); between high dose rivaroxaban and vitamin K antagonists (RR 0.91, 95% CI 0.45 to 1.85; 2 studies, 1546 participants; I² = 28%; very low‐certainty evidence; Analysis 3.4); or between low dose rivaroxaban and vitamin K antagonists (RR 0.59, 95% CI 0.29 to 1.20; 1 study, 1403 participants; low‐certainty evidence; Analysis 4.4).

2.4 — Comparison 2: High dose rivaroxaban vs low dose rivaroxaban, Outcome 4: Major Bleeding

3.4 — Comparison 3: Rivaroxaban (high dose) vs warfarin, Outcome 4: Major Bleeding

4.4 — Comparison 4: Rivaroxaban (low dose) vs warfarin, Outcome 4: Major Bleeding

One RCT compared high and low doses of dabigatran and vitamin K antagonists (RE‐DUAL PCI 2017).

There may be little or no difference in the rate of major bleeding between high and low dose dabigatran (RR 0.68, 95% CI 0.33 to 1.39; 1 study, 1744 participants; low‐certainty evidence; Analysis 5.3).

5.3 — Comparison 5: Low dose dabigatran vs high dose dabigatran, Outcome 3: Major Bleeding

High dose dabigatran may reduce the rate of major bleeding more than vitamin K antagonists (RR 0.53, 95% CI 0.29 to 0.97; 1 study, 1527 participants; low‐certainty evidence; Analysis 6.3). Low dose dabigatran probably reduces the rate of major bleeding compared to vitamin K antagonists (RR 0.38, 95% CI 0.21 to 0.70; 1 study, 1962 participants; moderate‐certainty evidence; Analysis 7.3).

6.3 — Comparison 6: Dabigatran (high dose) vs warfarin, Outcome 3: Major bleeding

7.3 — Comparison 7: Dabigatran (low dose) vs warfarin, Outcome 3: Major Bleeding

Network meta‐analysis (NOACs compared to each other)

Low‐certainty evidence suggests little or no difference in major bleeding between apixaban and high dose rivaroxaban (RR 0.94, 95% CI 0.47 to 1.89); or low dose rivaroxaban (RR 1.27, 95% CI 0.57 to 2.80; Figure 4).

Low‐certainty evidence suggests little or no difference in major bleeding between high dose rivaroxaban and high dose dabigatran (RR 1.56, 95% CI 0.69 to 3.53); high dose rivaroxaban and low dose dabigatran (RR 2.29, 95% CI 0.99 to 5.29); low dose rivaroxaban and high dose dabigatran (RR 1.16, 95% CI 0.47 to 2.87); or low dose rivaroxaban and low dose dabigatran (RR 1.71, 95% CI 0.68 to 4.30; Figure 4).

Low‐certainty evidence suggests little or no difference in major bleeding between high and low dose of rivaroxaban (RR 1.34, 95% CI 0.64 to 2.80), or and between high and low dose of dabigatran (RR 0.68, 95% CI 0.33 to 1.39; Figure 4; Table 10; Table 4).

4. League table – major bleeding.

Pairwise meta‐analysis
Vitamin K antagonist	0.82 (0.55 to 1.21)	0.91 (0.45 to 1.85)	0.59 (0.29 to 1.20)	0.38 (0.21 to 0.70)	0.53 (0.29 to 0.97)
0.82 (0.55 to 1.21)	Apixaban	‐	‐	‐	‐
0.87 (0.49 to 1.54)	0.94 (0.47 to 1.89)	Rivaroxaban (high dose)	1.18 (0.55 to 2.54)	‐	‐
0.65 (0.32 to 1.29)	1.27 (0.57 to 2.80)	1.34 (0.64 to 2.80)	Rivaroxaban (low dose)	‐	‐
0.38 (0.21 to 0.70)	0.46 (0.22 to 0.95)	2.29 (0.99 to 5.29)	1.71 (0.68 to 4.30)	Dabigatran (low dose)	0.68 (0.33 to 1.39)
0.56 (0.31 to 0.99)	0.68 (0.34 to 1.37)	1.56 (0.69 to 3.53)	1.16 (0.47 to 2.87)	0.68 (0.33 to 1.39)	Dabigatran (high dose)
Network meta‐analysis
Comparisons between treatments should be read from left to right and the estimate is in the cell in common between the column‐defining treatment and the row‐defining treatment. The upper triangle presents the results from direct (pairwise) meta‐analyses and the lower triangle presents the results from the network meta‐analyses. Results are presented as RR (95% CI), where a RR of < 1 favours the row‐defined treatment. The order of treatments in the diagonal is arbitrary and does not reflect ranking. CI: confidence interval; RR: risk ratio

Open in a new tab

Death from any cause

Direct evidence (NOACs vs vitamin K antagonists)

Two RCTs compared apixaban with vitamin K antagonists (AUGUSTUS 2019; Kopin 2018). There is probably little or no difference in the rate of death from any cause between apixaban and vitamin K antagonists (RR 1.18, 95% CI 0.74 to 1.87; 2 studies, 4930 participants; I² = 28%; moderate‐certainty evidence; Analysis 1.5).

1.5 — Comparison 1: Apixaban vs warfarin, Outcome 5: Death from any cause

One RCT compared death from any cause between rivaroxaban and vitamin K antagonists and between high and low dose rivaroxaban (PIONEER AF‐PCI 2016). Low‐certainty evidence suggests little or no difference between high and low dose rivaroxaban (RR 0.95, 95% CI 0.49 to 1.87; 1 study, 1402 participants; Analysis 2.5); between high dose rivaroxaban and vitamin K antagonists (RR 1.23, 95% CI 0.60 to 2.54; 1 study, 1393 participants; Analysis 3.5); or between low dose rivaroxaban and vitamin K antagonists (RR 1.29, 95% CI 0.63 to 2.64; 1 study, 1403 participants; Analysis 4.5).

2.5 — Comparison 2: High dose rivaroxaban vs low dose rivaroxaban, Outcome 5: Death from any cause

3.5 — Comparison 3: Rivaroxaban (high dose) vs warfarin, Outcome 5: Death from any cause

4.5 — Comparison 4: Rivaroxaban (low dose) vs warfarin, Outcome 5: Death from any cause

One RCT compared high and low doses of dabigatran and vitamin K antagonists (RE‐DUAL PCI 2017). Low‐certainty evidence suggests little or no difference in the rate of death from any cause between low and high dose dabigatran (RR 1.43, 95% CI 0.92 to 2.20; 1 study, 1744 participants; Analysis 5.4); between high dose dabigatran and vitamin K antagonists (RR 0.86, 95% CI 0.53 to 1.38; 1 study, 1527 participants; Analysis 6.4); and between low dose dabigatran and vitamin K antagonists (RR 1.15, 95% CI 0.79 to 1.67; 1 study, 1962 participants; Analysis 7.4).

5.4 — Comparison 5: Low dose dabigatran vs high dose dabigatran, Outcome 4: Death from any cause

6.4 — Comparison 6: Dabigatran (high dose) vs warfarin, Outcome 4: Death from any cause

7.4 — Comparison 7: Dabigatran (low dose) vs warfarin, Outcome 4: Death from any cause

Network meta‐analysis (NOACs compared to each other)

Low‐certainty evidence suggests little or no difference in death from any cause between apixaban and high dose rivaroxaban (RR 0.96, 95% CI 0.37 to 2.49), or apixaban and low dose rivaroxaban (RR 0.91, 95% CI 0.35 to 2.36; Figure 5).

Low‐certainty evidence suggests little or no difference in death from any cause between high dose rivaroxaban and high dose dabigatran (RR 1.53, 95% CI 0.54 to 4.32); high dose rivaroxaban and low dose dabigatran (RR 1.08, 95% CI 0.39 to 2.95); low dose rivaroxaban and high dose dabigatran (RR 1.61, 95% 0.57 to 4.49); or between low dose rivaroxaban and low dose dabigatran (RR 1.13, 95% CI 0.41 to 3.06; Figure 5).

Low‐certainty evidence suggests little or no difference in death from any cause between high and low dose rivaroxaban (RR 0.95, 95% CI 0.43 to 2.11), or between high and low dose dabigatran (RR 1.43, 95% CI 0.78 to 2.61; Figure 5; Table 11; Table 5).

5. League table – death from any cause.

Pairwise meta‐analysis
Vitamin K antagonist	1.18 (0.74 to 1.87)	1.23 (0.60 to 2.54)	1.29 (0.63 to 2.64)	1.15 (0.79 to 1.67)	0.86 (0.53 to 1.38)
1.18 (0.74 to 1.87)	Apixaban	‐	‐	‐	‐
1.23 (0.53 to 2.84)	0.96 (0.37 to 2.49)	Rivaroxaban (high dose)	0.95 (0.49 to 1.87)	‐	‐
1.29 (0.56 to 2.95)	0.91 (0.35 to 2.36)	0.95 (0.43 to 2.11)	Rivaroxaban (low dose)	‐	‐
1.15 (0.65 to 2.01)	0.97 (0.47 to 2.01)	1.08 (0.39 to 2.95)	1.13 (0.41 to 3.06)	Dabigatran (low dose)	1.43 (0.92 to 2.20)
0.80 (0.44 to 1.48)	0.68 (0.32 to 1.47)	1.53 (0.54 to 4.32)	1.61 (0.57 to 4.49)	1.43 (0.78 to 2.61)	Dabigatran (high dose)
Network meta‐analysis
Comparisons between treatments should be read from left to right and the estimate is in the cell in common between the column‐defining treatment and the row‐defining treatment. The upper triangle presents the results from direct (pairwise) meta‐analyses and the lower triangle presents the results from the network meta‐analyses. Results are presented as RR (95% CI), where a RR of < 1 favours the row‐defined treatment. The order of treatments in the diagonal is arbitrary and does not reflect ranking. CI: confidence interval; RR: risk ratio

Open in a new tab

Stent thrombosis

Direct evidence (NOACs vs vitamin K antagonists)

One RCT compared apixaban with vitamin K antagonists (AUGUSTUS 2019). There is probably little or no difference in the rate of stent thrombosis between apixaban and vitamin K antagonists (RR 0.78, 95% CI 0.39 to 1.56; 1 study, 4614 participants; moderate‐certainty evidence; Analysis 1.6).

1.6 — Comparison 1: Apixaban vs warfarin, Outcome 6: Stent thrombosis

One RCT compared stent thrombosis between rivaroxaban and vitamin K antagonists and between high and low dose of rivaroxaban (PIONEER AF‐PCI 2016).

Low‐certainty evidence suggests little or no difference in the rate of stent thrombosis between high and low dose rivaroxaban (RR 0.85, 95% CI 0.26 to 2.76; 1 study, 1398 participants; Analysis 2.6); between high dose rivaroxaban and vitamin K antagonists (RR 1.25, 95% CI 0.34 to 4.64; 1 study, 1389 participants; Analysis 3.6); or between low dose rivaroxaban and vitamin K antagonists (RR 1.48, 95% CI 0.42 to 5.22; 1 study, 1399 participants; Analysis 4.6).

2.6 — Comparison 2: High dose rivaroxaban vs low dose rivaroxaban, Outcome 6: Stent thrombosis

3.6 — Comparison 3: Rivaroxaban (high dose) vs warfarin, Outcome 6: Stent thrombosis

4.6 — Comparison 4: Rivaroxaban (low dose) vs warfarin, Outcome 6: Stent thrombosis

One RCT compared high and low doses of dabigatran and vitamin K antagonists (RE‐DUAL PCI 2017). Low‐certainty evidence suggests little or no difference in the rate of stent thrombosis between high and low dose dabigatran (RR 1.67, 95% CI 0.68 to 4.07; 1 study, 1744 participants; Analysis 5.5); between high dose dabigatran and vitamin K antagonists (RR 1.00, 95% CI 0.35 to 2.84; 1 study, 1527 participants; Analysis 6.5); and between low dose dabigatran and vitamin K antagonists (RR 1.88, 95% CI 0.80 to 4.40; 1 study, 1962 participants; Analysis 7.5).

5.5 — Comparison 5: Low dose dabigatran vs high dose dabigatran, Outcome 5: Stent thrombosis

6.5 — Comparison 6: Dabigatran (high dose) vs warfarin, Outcome 5: Stent thrombosis

7.5 — Comparison 7: Dabigatran (low dose) vs warfarin, Outcome 5: Stent thrombosis

Network meta‐analysis (NOACs compared to each other)

Low‐certainty evidence suggests little or no difference in stent thrombosis between apixaban and high dose rivaroxaban (RR 0.62, 95% CI 0.14 to 2.74), or apixaban and low dose rivaroxaban (RR 0.53, 95% CI 0.12 to 2.22; Figure 6).

Low‐certainty evidence suggests little or no difference in stent thrombosis between high dose rivaroxaban and high dose dabigatran (RR 1.11, 95% CI 0.21 to 5.82); high dose rivaroxaban and low dose dabigatran (RR 0.67, 95% CI 0.14 to 3.19); low dose rivaroxaban and high dose dabigatran (RR 1.32, 95% 0.26 to 6.62); or low dose rivaroxaban and low dose dabigatran (RR 0.79, 95% CI 0.17 to 3.62; Figure 6).

Low‐certainty evidence suggests little or no difference in stent thrombosis between high and low dose rivaroxaban (RR 0.85, 95% CI 0.26 to 2.76); or between low and high dose dabigatran (RR 1.67, 95% CI 0.68 to 4.07; Figure 6; Table 12; Table 6).

6. League table – stent thrombosis.

Pairwise meta‐analysis
Vitamin K antagonist	0.78 (0.39 to 1.56)	1.25 (0.34 to 4.64)	1.48 (0.42 to 5.22)	1.88 (0.80 to 4.40)	1.00 (0.35 to 2.84)
0.78 (0.39 to 1.56)	Apixaban	‐	‐	‐	‐
1.25 (0.34 to 4.64)	0.62 (0.14 to 2.74)	Rivaroxaban (high dose)	0.85 (0.26 to 2.76)	‐	‐
1.48 (0.42 to 5.22)	0.53 (0.12 to 2.22)	0.85 (0.26 to 2.76)	Rivaroxaban (low dose)	‐	‐
1.88 (0.80 to 4.40)	2.41 (0.80 to 7.25)	0.67 (0.14 to 3.19)	0.79 (0.17 to 3.62)	Dabigatran (low dose)	1.67 (0.68 to 4.07)
1.13 (0.41 to 3.09)	1.45 (0.42 to 4.93)	1.11 (0.21 to 5.82)	1.32 (0.26 to 6.62)	1.67 (0.68 to 4.07)	Dabigatran (high dose)
Network meta‐analysis
Comparisons between treatments should be read from left to right and the estimate is in the cell in common between the column‐defining treatment and the row‐defining treatment. The upper triangle presents the results from direct (pairwise) meta‐analyses and the lower triangle presents the results from the network meta‐analyses. Results are presented as RR (95% CI), where a RR of < 1 favours the row‐defined treatment. The order of treatments in the diagonal is arbitrary and does not reflect ranking. CI: confidence interval; RR: risk ratio

Open in a new tab

Non‐major TIMI bleeding

Direct evidence (NOACs vs vitamin K antagonists)

One RCT compared apixaban with vitamin K antagonists (AUGUSTUS 2019). Apixaban probably reduces rates of non‐major bleeding compared to vitamin K antagonists (RR 0.67, 95% CI 0.51 to 0.88; 1 study, 4549 participants; moderate‐certainty evidence; Analysis 1.7).

1.7 — Comparison 1: Apixaban vs warfarin, Outcome 7: Non‐Major Bleeding

One RCT compared non‐major bleeding rates between rivaroxaban and vitamin K antagonists and between high and low doses of rivaroxaban (PIONEER AF‐PCI 2016). Low‐certainty evidence suggests little or no difference in the rate of non‐major bleeding between high and low doses of rivaroxaban (RR 0.93, 95% CI 0.72 to 1.20; 1 study, 1402 participants; Analysis 2.7). High dose rivaroxaban probably reduces non‐major bleeding compared to vitamin K antagonists (RR 0.66, 95% CI 0.52 to 0.83; 1 study, 1393 participants; moderate‐certainty evidence; Analysis 3.7); as does low dose rivaroxaban (RR 0.71, 95% CI 0.57 to 0.88; 1 study, 1403 participants; moderate‐certainty evidence; Analysis 4.7).

2.7 — Comparison 2: High dose rivaroxaban vs low dose rivaroxaban, Outcome 7: Non‐Major Bleeding

3.7 — Comparison 3: Rivaroxaban (high dose) vs warfarin, Outcome 7: Non‐Major Bleeding

4.7 — Comparison 4: Rivaroxaban (low dose) vs warfarin, Outcome 7: Non‐Major Bleeding

One RCT compared high and low doses of dabigatran and vitamin K antagonists (RE‐DUAL PCI 2017). Low‐certainty evidence suggests that high dose dabigatran may reduce the rate of non‐major bleeding compared to vitamin K antagonists (RR 0.83, 95% CI 0.72 to 0.96; 1 study, 1527 participants; Analysis 6.6), while low dose dabigatran probably reduces the rate of non‐major bleeding compared to vitamin K antagonists (RR 0.66, 95% CI 0.58 to 0.75; 1 study, 1962 participants; moderate‐certainty evidence; Analysis 7.6).

6.6 — Comparison 6: Dabigatran (high dose) vs warfarin, Outcome 6: Non‐Major Bleeding

7.6 — Comparison 7: Dabigatran (low dose) vs warfarin, Outcome 6: Non‐Major Bleeding

Network meta‐analysis (NOACs compared to each other)

Low‐certainty evidence suggests little or no difference in the rates of non‐major bleeding between apixaban and high dose rivaroxaban (RR 1.01, 95% CI 0.70 to 1.45); or between apixaban and low dose rivaroxaban (RR 0.94, 95% CI 0.66 to 1.34; Figure 7).

Low‐certainty evidence suggests little or no difference in rates of non‐major bleeding between high dose rivaroxaban and high dose dabigatran (RR 0.82, 95% CI 0.63 to 1.07); high dose rivaroxaban and low dose dabigatran (RR 1.00, 95% CI 0.77 to 1.30); low dose rivaroxaban and high dose dabigatran (RR 0.88, 95% 0.68 to 1.15); or between low dose rivaroxaban and low dose dabigatran (RR 1.07, 95% CI 0.83 to 1.39; Figure 7).

Low‐certainty evidence suggests little or no difference in rates of non‐major bleeding between high and low dose rivaroxaban (RR 0.93, 95% CI 0.72 to 1.19), or between high and low dose dabigatran (RR 0.82, 95% CI 0.70 to 0.95; Figure 7).

Recurrent hospitalisation

Direct evidence (NOACs vs vitamin K antagonists)

One RCT compared apixaban with vitamin K antagonists (AUGUSTUS 2019). Apixaban probably reduces the risk of recurrent hospitalisation for adverse events compared with vitamin K antagonists (RR 0.85, 95% CI 0.77 to 0.95; 1 study, 4614 participants; moderate‐certainty evidence; Analysis 1.8).

1.8 — Comparison 1: Apixaban vs warfarin, Outcome 8: Recurrent hospitalisation (at least one at one year)

One RCT compared recurrent hospitalisation between rivaroxaban and vitamin K antagonists (PIONEER AF‐PCI 2016).

Low‐certainty evidence suggests little or no difference in the rate of recurrent hospitalisation between high and low dose rivaroxaban (RR 1.08, 95% CI 0.92 to 1.27; 1 study, 1402 participants; Analysis 2.8).

2.8 — Comparison 2: High dose rivaroxaban vs low dose rivaroxaban, Outcome 8: Recurrent hospitalisation (at least one at one year)

High dose rivaroxaban likely reduces the rate of recurrent hospitalisation more than vitamin K antagonists (RR 0.86, 95% CI 0.74 to 1.00; 1 study, 1393 participants; moderate‐certainty evidence; Analysis 3.8), while low dose rivaroxaban may reduce the rate of recurrent hospitalisation more than vitamin K antagonists (RR 0.80, 95% CI 0.68 to 0.92; 1 study, 1403 participants; low‐certainty evidence; Analysis 4.8).

3.8 — Comparison 3: Rivaroxaban (high dose) vs warfarin, Outcome 8: Recurrent hospitalisation (at least one at one year)

4.8 — Comparison 4: Rivaroxaban (low dose) vs warfarin, Outcome 8: Recurrent hospitalisation (at least one at one year)

No trial compared the rate of recurrent hospitalisation between dabigatran and vitamin K antagonists.

Network meta‐analysis (NOACs compared to each other)

Low‐certainty evidence suggests little or no difference in the rate of recurrent hospitalisation between apixaban and high dose rivaroxaban (RR 0.99, 95% CI 0.83 to 1.18); or between apixaban and low dose rivaroxaban (RR 1.07, 95% CI 0.89 to 1.28; Figure 8).

Low‐certainty evidence suggests little or no difference in the rate of recurrent hospitalisation between high and low dose rivaroxaban (RR 1.08, 95% CI 0.92 to 1.26; Figure 8).

Health‐related quality of life

None of the included studies reported this outcome.

Subgroup analysis

We found insufficient data to carry out intended subgroup analyses.

Heterogeneity and inconsistency

We used R package 'netsplit' to split network estimates into the contribution of direct and indirect evidence, and to test for local inconsistencies in our network meta‐analyses. Inconsistency was present in two comparisons (high dose rivaroxaban versus low dose rivaroxaban, and low dose rivaroxaban versus vitamin K antagonists), and in cases when the Sherwood 2016 study was included in the respective network. P values for comparing the direct and indirect evidence are shown in Table 1; Table 2; Table 3; Table 4; Table 5; and Table 6, indicating no significant difference between direct and indirect evidence.

Ranking

Competing treatments were ranked by P scores, which allowed ranking of treatments on a continuous 0 to 1 scale, and were derived from the P values of all pairwise comparisons. P scores were based solely on the point estimates and standard errors of the frequentist network meta‐analysis estimates under normality assumption. P scores measure the mean extent of certainty that a treatment is better than the competing treatments. However, rankings are not measures of treatment effectiveness. See Table 13 and Figure 12.

7. Ranking.

Intervention	Rank (P value)
Intervention	Death from cardiovascular causes	Myocardial infarction	Stroke	Major bleeding	Death from any cause	Stent thrombosis
Vitamin K antagonist	3 (0.4475)	4 (0.5103)	4 (0.4673)	5 (0.1998)	2 (0.6192)	2 (0.6234)
Apixaban	4 (0.4108)	2 (0.7040)	1 (0.7810)	6 (0.0006)	4 (0.3977)	1 (0.7933)
Rivaroxaban (high dose)	1 (0.6461)	3 (0.5917)	3 (0.5614)	2 (0.8244)	5 (0.3947)	4 (0.4773)
Rivaroxaban (low dose)	2 (0.4955)	1 (0.7520)	5 (0.3346)	3 (0.6767)	6 (0.3486)	5 (0.3681)
Dabigratan (low dose)	‐	6 (0.0697)	6 (0.2871)	1 (0.8447)	3 (0.4242)	6 (0.1902)
Dabigratan (high dose)	‐	5 (0.3723)	2 (0.5686)	4 (0.4538)	1 (0.8156)	3 (0.5476)

Open in a new tab

Competing treatments are ranked by P‐scores, which allow ranking of treatments on a continuous 0 to 1 scale, and are derived from the P‐values of all pairwise comparisons

Discussion

Summary of main results

Our review aimed to assess the efficacy and safety of non‐vitamin K antagonist oral anticoagulants (NOACs) following a percutaneous coronary intervention (PCI) in people with an indication for anticoagulation. We included five trials, with 8373 participants with non‐valvular atrial fibrillation, comparing NOACs with vitamin K antagonists post‐PCI. We used the following outcomes to compare efficiency of these agents: death from cardiovascular causes, myocardial infarction, stroke, death from any cause, stent thrombosis, recurrent hospitalisation, and health‐related quality of life. To assess the safety of NOACs, we used two outcomes: major Thrombolysis In Myocardial Infarction criteria (TIMI) bleeding, and any non‐major TIMI bleeding. It is worth mentioning that non‐valvular atrial fibrillation was the indication for anticoagulation in all included studies of our review. We did not find any study that compared NOACs to vitamin K antagonists in people with venous thromboembolism who were undergoing PCI.

Efficacy of NOACs post‐PCI in people with an indication for anticoagulation

Very low‐ to moderate‐certainty evidence found inconclusive results between NOACs (specifically apixaban, rivaroxaban, and dabigatran) and vitamin K antagonists for the rate of myocardial infarction, stroke, death from any cause, and stent thrombosis. Low‐ to moderate‐certainty evidence suggests little or no difference between NOACs (specifically apixaban, rivaroxaban) and vitamin K antagonists in death from cardiovascular causes. The included trial on dabigatran did not report this outcome. Apixaban and rivaroxaban in either high and low doses probably reduces the risk of recurrent hospitalisation for adverse events compared with vitamin K antagonists. No studies were found that assessed health‐related quality of life.

Safety of NOACs post PCI in people with an indication for anticoagulation.

NOACs may be safer than vitamin K antagonists in terms of major bleeding. Dabigatran in both high and low doses may reduce both major and non‐major bleeding; while apixaban and rivaroxaban probably reduce non‐major bleeding compared to vitamin K antagonists.

Comparing NOACs against each other

We did not find any trials comparing NOACs directly against each other. Our network meta‐analysis compared NOACs agents indirectly against each other. We found inconclusive results between NOAC agents regarding all primary and secondary outcomes. We conducted the frequentist P score to rank the different treatment on a continuous 0 to 1 scale, based on P values of all pairwise comparisons. However, rankings are not measures of treatment effectiveness. According to the P score, participants taking a high dose of rivaroxaban may have the lowest rate of death from cardiovascular events; those taking a low dose of rivaroxaban may have the lowest rate of myocardial infarction; those taking apixaban may have the lowest rate of stroke; those taking a low dose of dabigatran may have the lowest rate of major bleeding; and participants taking a high dose of dabigatran may have the lowest rate of death from any cause.

Overall completeness and applicability of evidence

We aimed to evaluate the efficacy and safety of NOACs post‐PCI in people with an indication for anticoagulation. Given the complexity of the condition, and in the absence of randomised controlled trials comparing different types of NOACs against each other, we conducted a network meta analysis. This provided a comprehensive and comparative evaluation of all available treatment options in a coherent and methodologically robust way across efficacy and safety outcomes. We combined both direct and indirect evidence, thus increasing the statistical power and confidence in the results. The review suggests that NOACs may be safer than, and as effective as vitamin K antagonists post‐PCI in patients with an indication for oral anticoagulation due to non‐valvular atrial fibrillation. The rate of recurrent hospitalisation was probably lower with NOACs compared to vitamin K antagonists. Due to insufficient data, we were unable to identify which NOAC agent is safer and more effective than the rest. All relevant outcomes were reported by the included trials, except health‐related quality of life. The conclusions of this review are based on limited number of RCTs. There is need for high quality RCTs to give more conclusive evidence about the use of NOACs post PCI. Also, future studies should investigate the influence of NOACs on quality of life outcomes.

Quality of the evidence

The overall certainty of the evidence ranged from moderate to very low. We included five randomised controlled trials (RCT) with 8373 randomised participants. One of the reasons for downgrading the certainty of the evidence was imprecision of results with wide confidence intervals. The optimal information size was not met in four trials (RE‐DUAL PCI 2017; PIONEER AF‐PCI 2016; Kopin 2018; Sherwood 2016). In addition two trials in the meta‐analysis were very small post‐hoc subgroup analysis and thus were at risk of significant confounding (Kopin 2018; Sherwood 2016).

Potential biases in the review process

We performed a comprehensive systematic literature search to find all relevant trials for inclusion in this review. Two review authors independently checked the results of the searches to identify relevant studies, assessed each included study, extracted study data, and assessed risk of bias to, minimise review bias. We conducted the review according to the previously published protocol. However, there were some instances in which we deviated from the protocol during the review process; these are documented under Differences between protocol and review. Our network meta‐analysis (NMA) has limitations. First, we assessed the outcomes at the latest point of follow‐up for each trial. Therefore, caution is needed in interpreting the results of our review due to the differences in follow‐up (from 6 months to 2.2 years). Our NMA included two small post hoc subgroup analyses, which are, by definition, not randomised, and thus, were at risk of significant confounding. Indeed, there was heterogeneity across the included trials with respect to concomitant use of antiplatelet therapy; follow‐up time; and type, dose, and duration of antithrombotic therapy. That may also potentially affect the interpretation of our results. It should be noted that the very low dose of rivaroxaban (2.5 mg twice daily) is not approved for the prevention of thromboembolic events in atrial fibrillation. Morever, some outcome definitions, such as stent thrombosis, were not uniform across trials. Finally, individual participant data are not publicly available. An individual participant level data analysis would be more robust to explore who would benefit most from a given treatment combination.

Agreements and disagreements with other studies or reviews

Our findings agree with, and extend the findings of two previous systematic reviews. Brunetti 2018 performed a meta‐analysis of RCTs comparing NOACs with standard triple therapy in people with non‐valvular atrial fibrillation undergoing PCI. The review suggested that NOACs are safer than, and as effective as warfarin, when used in participants with atrial fibrillation undergoing PCI. However, this study only included two RCTs, did not include data on mortality in their meta‐analysis, and did not use a network meta analysis (PIONEER AF‐PCI 2016; RE‐DUAL PCI 2017). Lopes 2019a evaluated the safety and efficacy of different antithrombotic regimens in people with atrial fibrillation undergoing PCI, using a network meta‐analysis of five RCTs (AUGUSTUS 2019; Dewilde 2013; Fiedler 2015; PIONEER AF‐PCI 2016; RE‐DUAL PCI 2017). The study showed that administration of NOAC plus a P2Y12 inhibitor resulted in less bleeding compared with vitamin K antagonists plus dual antiplatelet therapy, without significant difference in efficacy outcomes. Howeover, two of the included studies in this review did not compare NOACs to vitamin K antagonists (Dewilde 2013; Fiedler 2015). Furthermore, they did not include the two trials comparing NOACs to vitamin K antagonists post PCI in participants with atrial fibrillation in their analysis, as we did (Kopin 2018; Sherwood 2016).

Authors' conclusions

Implications for practice.

We conclude that using non‐vitamin K antagonist oral anticoagulants (NOAC) in combination with antiplatelet therapy might be safer than triple therapy regimen of vitamin K antagonists plus dual antiplatelet therapy post‐percutaneous coronary intervention (PCI) in participants with non‐valvular atrial fibrillation. Low‐certainty evidence suggests little or no difference between NOACs and vitamin K antagonists regarding efficacy outcomes, suggesting that NOACs may not cause harm due to reduced antithrombotic potency. The rate of recurrent hospitalisation is probably lower with NOACs compared to vitamin K antagonists. We found no clear evidence to inform guidelines about which NOAC agent should be used in preference to another for the management of individuals with indication for anticoagulation after PCI.

Implications for research.

More randomised controlled trials are needed to assess the efficacy and safety of NOACs post‐PCI in people with an indication for anticoagulation. These trials should be well conducted randomised trials, designed to be large enough to detect differences in the incidence of ischaemic events. To date, combination therapy with a NOAC and an antiplatelet agent has not been addressed in people with venous thromboembolism, and it is unclear whether data from atrial fibrillation trials can be extrapolated for these individuals when they undergo PCI. Outcomes, such as health‐related quality of life should be evaluated. Studies comparing NOACs directly against each other are needed to determine which NOAC agent is preferable in dual antithrombotic therapy combining platelet inhibition and anticoagulation.

What's new

Date	Event	Description
15 March 2021	Amended	League tables and Effects of interventions amended so that values were consistent with the Results. Changes did not affect overall findings or conclusions.

Open in a new tab

History

Protocol first published: Issue 1, 2019 Review first published: Issue 12, 2019

Acknowledgements

The authors are grateful for the support of the Cochrane Heart Group editorial team and all the peer reviewers, editors and copy editors for providing precious help.

Appendices

Appendix 1. Search strategy 2019

CENTRAL

#1 MeSH descriptor: [Angioplasty, Balloon, Coronary] this term only

#2 (balloon NEXT angioplast*)

#3 (percutaneous NEAR/3 coronary)

#4 PCI

#5 PTCA

#6 (coronary NEXT angioplast*)

#7 (coronary NEXT stent*)

#8 ((transluminal or trans‐luminal) NEAR/6 coronary)

#9 (coronary NEXT balloon dilation*)

#10 #1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9

#11 ((novel or new) NEXT anticoagulant*)

#12 NOAC*

#13 DOAC*

#14 ((non‐vitamin K or direct) NEXT oral anticoagulant*)

#15 apixaban

#16 dabigatran

#17 rivaroxaban

#18 edoxaban

#19 MeSH descriptor: [Dabigatran] this term only

#20 MeSH descriptor: [Rivaroxaban] this term only

#21 #11 OR #12 OR #13 OR #14 OR #15 OR #16 OR #17 OR #18 OR #19 OR #20

#22 #10 AND #21

MEDLINE Ovid

1 Angioplasty, Balloon, Coronary/

2 (balloon adj3 angioplast*).tw.

3 (percutaneous adj3 coronary).tw.

4 PCI.tw.

5 PTCA.tw.

6 (coronary adj5 angioplast*).tw.

7 (coronary adj5 stent*).tw.

8 ((transluminal or trans‐luminal) adj6 coronary).tw.

9 (coronary adj5 balloon dilation*).tw.

10 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9

11 ((novel or new) adj5 anticoagulant*).tw.

12 NOAC*.tw.

13 DOAC*.tw.

14 ((non‐vitamin K or direct) adj5 oral anticoagulant*).tw.

15 apixaban.tw.

16 dabigatran.tw.

17 rivaroxaban.tw.

18 edoxaban.tw.

19 dabigatran/ or rivaroxaban/

20 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18 or 19

21 10 and 20

22 randomised controlled trial.pt.

23 controlled clinical trial.pt.

24 randomized.ab.

25 placebo.ab.

26 drug therapy.fs.

27 randomly.ab.

28 trial.ab.

29 groups.ab.

30 22 or 23 or 24 or 25 or 26 or 27 or 28 or 29

31 exp animals/ not humans.sh.

32 30 not 31

33 21 and 32

Embase Ovid

1 exp percutaneous coronary intervention/

2 (balloon adj3 angioplast*).tw.

3 (percutaneous adj3 coronary).tw.

4 PCI.tw.

5 PTCA.tw.

6 (coronary adj5 angioplast*).tw.

7 (coronary adj5 stent*).tw.

8 ((transluminal or trans‐luminal) adj6 coronary).tw.

9 (coronary adj5 balloon dilation*).tw.

10 or/1‐9

11 ((novel or new) adj5 anticoagulant*).tw.

12 NOAC*.tw.

13 DOAC*.tw.

14 ((non‐vitamin K or direct) adj5 oral anticoagulant*).tw.

15 apixaban.tw.

16 dabigatran.tw.

17 rivaroxaban.tw.

18 edoxaban.tw.

19 dabigatran/

20 apixaban/

21 rivaroxaban/

22 edoxaban/

23 or/11‐22

24 10 and 20

25 random$.tw.

26 factorial$.tw.

27 crossover$.tw.

28 cross over$.tw.

29 cross‐over$.tw.

30 placebo$.tw.

31 (doubl$ adj blind$).tw.

32 (singl$ adj blind$).tw.

33 assign$.tw.

34 allocat$.tw.

35 volunteer$.tw.

36 crossover procedure/

37 double blind procedure/

38 randomised controlled trial/

39 single blind procedure/

40 25 or 26 or 27 or 28 or 29 or 30 or 31 or 32 or 33 or 34 or 35 or 36 or 37 or 38 or 39

41 (animal/ or nonhuman/) not human/

42 40 not 41

43 24 and 42

CPCI‐S

# 16 #15 AND #14

# 15 TS=(random* or blind* or allocat* or assign* or trial* or placebo* or crossover* or cross‐over*)

# 14 #13 AND #8

# 13 #12 OR #11 OR #10 OR #9

# 12 TS=(apixaban OR dabigatran OR rivaroxaban OR edoxaban)

# 11 TS=((non‐vitamin K or direct) AND oral anticoagulant*)

# 10 TS=(NOAC* OR DOAC*)

# 9 TS=((novel or new) AND anticoagulant*)

# 8 #7 OR #6 OR #5 OR #4 OR #3 OR #2 OR #1

# 7 TS=(coronary AND balloon dilation*)

# 6 TS=((transluminal or trans‐luminal) AND coronary)

# 5 TS=(coronary AND stent*)

# 4 TS=(coronary AND angioplast*)

# 3 TS=(PCI OR PTCA)

# 2 TS=(percutaneous AND coronary)

# 1 TS=(balloon AND angioplast*)

Clinicaltrials.gov

Advanced search: Interventional Studies | NOAC

WHO ICTRP

Advanced search:

Intervention: NOAC

Recruitment status: ALL

Phases are: Phase 2, 3, 4

Data and analyses

Comparison 1. Apixaban vs warfarin.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1.1 Death from cardiovascular causes	2	4930	Risk Ratio (M‐H, Random, 95% CI)	1.06 [0.74, 1.51]
1.2 Myocardial infarction	2	4930	Risk Ratio (M‐H, Random, 95% CI)	0.89 [0.65, 1.20]
1.3 Stroke	2	4930	Risk Ratio (M‐H, Random, 95% CI)	0.68 [0.27, 1.76]
1.4 Major Bleeding	2	4865	Risk Ratio (M‐H, Random, 95% CI)	0.82 [0.55, 1.21]
1.5 Death from any cause	2	4930	Risk Ratio (M‐H, Random, 95% CI)	1.18 [0.74, 1.87]
1.6 Stent thrombosis	1	4614	Risk Ratio (M‐H, Random, 95% CI)	0.78 [0.39, 1.56]
1.7 Non‐Major Bleeding	1	4549	Risk Ratio (M‐H, Random, 95% CI)	0.67 [0.51, 0.88]
1.8 Recurrent hospitalisation (at least one at one year)	1	4614	Risk Ratio (IV, Random, 95% CI)	0.85 [0.77, 0.95]

Open in a new tab

Comparison 2. High dose rivaroxaban vs low dose rivaroxaban.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
2.1 Death from cardiovascular causes	1	1398	Risk Ratio (M‐H, Random, 95% CI)	1.09 [0.53, 2.23]
2.2 Myocardial infarction	1	1398	Risk Ratio (M‐H, Random, 95% CI)	1.13 [0.59, 2.16]
2.3 Stroke	1	1398	Risk Ratio (M‐H, Random, 95% CI)	0.81 [0.32, 2.04]
2.4 Major Bleeding	1	1402	Risk Ratio (M‐H, Random, 95% CI)	1.18 [0.55, 2.54]
2.5 Death from any cause	1	1402	Risk Ratio (M‐H, Random, 95% CI)	0.95 [0.49, 1.87]
2.6 Stent thrombosis	1	1398	Risk Ratio (M‐H, Random, 95% CI)	0.85 [0.26, 2.76]
2.7 Non‐Major Bleeding	1	1402	Risk Ratio (M‐H, Random, 95% CI)	0.93 [0.72, 1.20]
2.8 Recurrent hospitalisation (at least one at one year)	1	1402	Risk Ratio (M‐H, Random, 95% CI)	1.08 [0.92, 1.27]

Open in a new tab

Comparison 3. Rivaroxaban (high dose) vs warfarin.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
3.1 Death from cardiovascular causes	2	1542	Risk Ratio (M‐H, Random, 95% CI)	0.69 [0.14, 3.33]
3.2 Myocardial infarction	2	1542	Risk Ratio (M‐H, Random, 95% CI)	0.94 [0.53, 1.65]
3.3 Stroke	2	1542	Risk Ratio (M‐H, Random, 95% CI)	0.94 [0.38, 2.34]
3.4 Major Bleeding	2	1546	Risk Ratio (M‐H, Random, 95% CI)	0.91 [0.45, 1.85]
3.5 Death from any cause	1	1393	Risk Ratio (M‐H, Random, 95% CI)	1.23 [0.60, 2.54]
3.6 Stent thrombosis	1	1389	Risk Ratio (M‐H, Random, 95% CI)	1.25 [0.34, 4.64]
3.7 Non‐Major Bleeding	1	1393	Risk Ratio (M‐H, Random, 95% CI)	0.66 [0.52, 0.83]
3.8 Recurrent hospitalisation (at least one at one year)	1	1393	Risk Ratio (IV, Random, 95% CI)	0.86 [0.74, 1.00]

Open in a new tab

Comparison 4. Rivaroxaban (low dose) vs warfarin.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
4.1 Death from cardiovascular causes	1	1399	Risk Ratio (M‐H, Random, 95% CI)	1.26 [0.57, 2.75]
4.2 Myocardial infarction	1	1399	Risk Ratio (M‐H, Random, 95% CI)	0.80 [0.43, 1.50]
4.3 Stroke	1	1399	Risk Ratio (M‐H, Random, 95% CI)	1.41 [0.54, 3.68]
4.4 Major Bleeding	1	1403	Risk Ratio (M‐H, Random, 95% CI)	0.59 [0.29, 1.20]
4.5 Death from any cause	1	1403	Risk Ratio (M‐H, Random, 95% CI)	1.29 [0.63, 2.64]
4.6 Stent thrombosis	1	1399	Risk Ratio (M‐H, Random, 95% CI)	1.48 [0.42, 5.22]
4.7 Non‐Major Bleeding	1	1403	Risk Ratio (M‐H, Random, 95% CI)	0.71 [0.57, 0.88]
4.8 Recurrent hospitalisation (at least one at one year)	1	1403	Risk Ratio (M‐H, Random, 95% CI)	0.80 [0.68, 0.92]

Open in a new tab

Comparison 5. Low dose dabigatran vs high dose dabigatran.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
5.1 Myocardial infarction	1	1744	Risk Ratio (M‐H, Random, 95% CI)	1.32 [0.82, 2.12]
5.2 Stroke	1	1744	Risk Ratio (M‐H, Random, 95% CI)	1.47 [0.66, 3.28]
5.3 Major Bleeding	1	1744	Risk Ratio (M‐H, Random, 95% CI)	0.68 [0.33, 1.39]
5.4 Death from any cause	1	1744	Risk Ratio (M‐H, Random, 95% CI)	1.43 [0.92, 2.20]
5.5 Stent thrombosis	1	1744	Risk Ratio (M‐H, Random, 95% CI)	1.67 [0.68, 4.07]
5.6 Non‐Major Bleeding	1	1744	Risk Ratio (M‐H, Random, 95% CI)	0.82 [0.71, 0.96]

Open in a new tab

5.6 — Comparison 5: Low dose dabigatran vs high dose dabigatran, Outcome 6: Non‐Major Bleeding

Comparison 6. Dabigatran (high dose) vs warfarin.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
6.1 Myocardial infarction	1	1527	Risk Ratio (M‐H, Random, 95% CI)	1.18 [0.68, 2.07]
6.2 Stroke	1	1527	Risk Ratio (M‐H, Random, 95% CI)	1.13 [0.44, 2.90]
6.3 Major bleeding	1	1527	Risk Ratio (M‐H, Random, 95% CI)	0.53 [0.29, 0.97]
6.4 Death from any cause	1	1527	Risk Ratio (M‐H, Random, 95% CI)	0.86 [0.53, 1.38]
6.5 Stent thrombosis	1	1527	Risk Ratio (M‐H, Random, 95% CI)	1.00 [0.35, 2.84]
6.6 Non‐Major Bleeding	1	1527	Risk Ratio (M‐H, Random, 95% CI)	0.83 [0.72, 0.96]

Open in a new tab

Comparison 7. Dabigatran (low dose) vs warfarin.

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
7.1 Myocardial infarction	1	1962	Risk Ratio (M‐H, Random, 95% CI)	1.52 [0.96, 2.40]
7.2 Stroke	1	1962	Risk Ratio (M‐H, Random, 95% CI)	1.31 [0.64, 2.68]
7.3 Major Bleeding	1	1962	Risk Ratio (M‐H, Random, 95% CI)	0.38 [0.21, 0.70]
7.4 Death from any cause	1	1962	Risk Ratio (M‐H, Random, 95% CI)	1.15 [0.79, 1.67]
7.5 Stent thrombosis	1	1962	Risk Ratio (M‐H, Random, 95% CI)	1.88 [0.80, 4.40]
7.6 Non‐Major Bleeding	1	1962	Risk Ratio (M‐H, Random, 95% CI)	0.66 [0.58, 0.75]

Open in a new tab

Characteristics of studies

Characteristics of included studies [ordered by study ID]

AUGUSTUS 2019.

*Study characteristics*
Methods	Study design: randomised controlled trial Study grouping: parallel group Follow‐up: 6 months Recruitment period: from September 2015 through April 2018 Treatment duration: 180 days
Participants	Baseline characteristics Apixaban Number randomised: 2306 Number analysed: not mentioned Gender (female %): 29.1% Age in years: mean: 70.4 Ethnicity: White (91.9%); Black (1.3%); Asian (3.1%); Native American (0.4%); other (3.3%) Loss to follow‐up: 6 CHA₂DS₂‐VASc score < 2: not mentioned CHA₂DS₂‐VASc score 2 to 4: not mentioned CHA₂DS₂‐VASc score > 4: not mentioned Indication for NOAC: 100% atrial fibrillation Indication for PCI: ACS and PCI 38.0%; medically managed ACS 23.8%; elective PCI 38.2% Warfarin Number randomised: 2308 Number analysed: not mentioned Gender (female %): 28.9% Age in years: mean: 70.9 Ethnicity: White (91.7%); Black (1.3%); Asian (3.1%); Native American (0.3%); other (3.6%) Loss to follow‐up: 7 CHA₂DS₂‐VASc score < 2: not mentioned CHA₂DS₂‐VASc score 2 to 4: not mentioned CHA₂DS₂‐VASc score > 4: not mentioned Indication for NOAC: 100% atrial fibrillation Indication for PCI: ACS 36.6%; medically managed ACS 23.9%; elective PCI 39.5% Overall Number randomised: 4614 Number analysed: not mentioned Gender (female %): 29.0% Age in years: mean: 70.7 Ethnicity: White (91.8%); Black (1.3%); Asian (3.1%); Native American (0.4%); Other (3.5%). Loss to follow‐up: not mentioned CHA₂DS₂‐VASc score < 2: not mentioned CHA₂DS₂‐VASc score 2 to 4: not mentioned CHA₂DS₂‐VASc score > 4: not mentioned Indication for NOAC: 100% atrial fibrillation Indication for PCI: ACS 37.3%; medically managed ACS 23.9%; elective PCI 38.8% Inclusion criteria: age ≥18 years; previous, persistent, permanent, or paroxysmal atrial fibrillation and planned long‐term use of an oral anticoagulant; recent acute coronary syndrome or PCI; and planned use of a P2Y12 inhibitor for at least 6 months Exclusion criteria: participants who were using anticoagulation for other conditions (e.g. prosthetic valves, venous thromboembolism, and mitral stenosis); severe renal insufficiency; a history of intracranial haemorrhage; recent or planned coronary‐artery bypass graft surgery; coagulopathy or ongoing bleeding; and contraindication to a vitamin K antagonist, apixaban, all P2Y12 inhibitors, or aspirin
Interventions	Intervention characteristics Apixaban Concomitant medications: aspirin or placebo; P2Y12 inhibitor for all participants x 6 months Type of stent:not mentioned Warfarin Concomitant medications: aspirin or placebo; P2Y12 inhibitor for all participants x 6 months Type of stent: not mentioned
Outcomes	Death from cardiovascular causes Outcome type: dichotomous outcome Myocardial infarction Outcome type: dichotomous outcome Stroke Outcome type: dichotomous outcome Major bleeding Outcome type: adverse event Death from any cause Outcome type: dichotomous outcome Stent thrombosis Outcome type: dichotomous outcome Any non‐major TIMI bleeding Outcome type: adverse event Recurrent hospitalisation (at least one at one year) Outcome type: dichotomous outcome Health‐related quality of life (HRQL) Outcome type: continuous outcome
Identification	Sponsorship source: trial sponsored by Bristol‐Myers Squibb and Pfizer Country: 33 countries Setting: 4614 participants from 492 sites in 33 countries Authors name: Professor Renato D. Lopes, MD, MHS, PhD Institution: Division of Cardiology at Duke University Medical Center, Durham, North Carolina, USA Email: renato.lopes@duke.edu Address: Dr. Lopes at the Duke Clinical Research Institute, 200 Morris St., Durham, NC 27701
Notes
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	participants were randomly assigned by means of an interactive voice‐response system. multicenter, two‐by‐two factorial, randomised clinical trial
Allocation concealment (selection bias)	Low risk	Randomization sequence with interactive voice response system
Blinding of participants and personnel (performance bias) All outcomes	High risk	Comparing apixaban with a vitamin K antagonist was open‐label
Blinding of outcome assessment (detection bias) All outcomes	Low risk	All the statistical analyses were performed independently at the Duke Clinical Research Institute
Incomplete outcome data (attrition bias) All outcomes	Low risk	The population for the two secondary outcome analyses included all the participants who underwent randomisation according to the randomised groups.
Selective reporting (reporting bias)	Low risk	Judgement comment: all outcomes were reported according to a priori published protocol
Other bias	Low risk	Industry‐funded study. Source of funding and conflicts of interest declared appropriately. Sponsored by Bristol‐Myers Squibb and Pfizer.

Open in a new tab

Kopin 2018.

*Study characteristics*
Methods	Study design: randomised controlled trial Study grouping: parallel group Follow‐up: 1.8 years Recruitment period: 19 December 2006 to 02 April 2010 Treatment duration: unknown
Participants	Baseline characteristics Apixaban 5 mg Number randomised: 152 Number analysed: 152 Gender (female %): 23.0% Age in years: median (25th to 75th): 71 (64 to 76) Ethnicity: North America 38.8%, Latin America 8.6%, Europe 41.4%, Asia: 11.2% Loss to follow‐up: not mentioned CHA₂DS₂‐VASc score < 2: 4.6% CHA₂DS₂‐VASc score 2 to 4: 20.4% CHA₂DS₂‐VASc score > 4: 75.0% Indication for NOAC: atrial fibrillation Indication for PCI: not mentioned Type of AF: persistent/permanent: 79.5%, paroxysmal 20.5% HAS‐BLED ≤ 1: 35.5% HAS‐BLED 2: 28.3% HAS‐BLED ≥ 3: 36.2% Warfarin Number randomised: 164 Number analysed: 164 Gender (female %): 23.8% Age in years: median (25th to 75th): 71 (64 to 75) Ethnicity: North America 36.6%, Latin America 7.9%, Europe 48.2%, Asia 7.3% Loss to follow‐up: not mentioned CHA₂DS₂‐VASc score < 2: 5.5% CHA₂DS₂‐VASc score 2 to 4: 18.3% CHA₂DS₂‐VASc score > 4: 76.2% Indication for NOAC: atrial fibrillation Indication for PCI: not mentioned Type of AF: persistent/permanent: 79.9%, paroxysmal 20.1% HAS‐BLED ≤ 1: 27.4% HAS‐BLED 2: 47.0% HAS‐BLED ≥ 3: 25.6% Overall Number randomised: 316 Number analysed: 316 Gender (female %): 23.4% Age in years: median (25th to 75th): 71 (64 to 76) Ethnicity: North America 37.7%, Latin America 8.2%, Europe 44.9%, Asia 9.2% Loss to follow‐up: 13 participants were excluded from the efficacy end points analysis because PCI occurred after the efficacy censoring date, not at risk of efficacy endpoints. 40 participants excluded from the safety end points analysis because PCI occurred after the safety censoring date CHA₂DS₂‐VASc score < 2: not mentioned CHA₂DS₂‐VASc score 2 to 4: not mentioned CHA₂DS₂‐VASc score > 4: not mentioned Indication for NOAC: atrial fibrillation Indication for PCI: not mentioned Type of AF: persistent/permanent 79.7%, paroxysmal 20.3% HAS‐BLED ≤ 1: 31.3% HAS‐BLED 2: 38% HAS‐BLED ≥ 3: 30.7% Inclusion criteria: eligible participants had atrial fibrillation or flutter at enrolment, or two or more episodes of atrial fibrillation or flutter, as documented by electrocardiography, at least 2 weeks apart, in the 12 months before enrolment. In addition, at least one of the following risk factors for stroke was required: an age of at least 75 years; previous stroke, transient ischaemic attack, or systemic embolism; symptomatic heart failure within the previous 3 months, or left ventricular ejection fraction of no more than 40%; diabetes mellitus; or hypertension requiring pharmacologic treatment. Exclusion criteria: key exclusion criteria were atrial fibrillation due to a reversible cause; moderate or severe mitral stenosis; conditions other than atrial fibrillation that required anticoagulation (e.g. a prosthetic heart valve); stroke within the previous 7 days; a need for aspirin at a dose of > 165 mg a day, or for both aspirin and clopidogrel; and severe renal insufficiency (serum creatinine level of > 2.5 mg per dL (221 μmol/L), or calculated creatinine clearance of < 25 mL/minute)
Interventions	Intervention characteristics Apixaban 5 mg Concomitant medications: 27% none, 39% DAPT, 24% aspirin alone, 10% thienopyridine Type of stent: overall, 38% of participants received no stent; 32% received drug‐eluting stents, and 30% received bare‐metal stents. Warfarin Concomitant medications: 32% none, 31% DAPT, 22% aspirin alone, 15% thienopyridine Type of stent: overall, 38% of participants received no stent; 32% received drug‐eluting stents; and 30% received bare‐metal stents.
Outcomes	Death from cardiovascular causes Outcome type: dichotomous outcome Reporting: partially reported Myocardial infarction Outcome type: dichotomous outcome Reporting: fully reported Direction: lower is better Stroke Outcome type: dichotomous outcome Reporting: fully reported Unit of measure: person Direction: lower is better Major bleeding Outcome type: adverse event Death from any cause Outcome type: dichotomous outcome Stent thrombosis Outcome type: dichotomous outcome Reporting: not reported Any non‐major TIMI bleeding Outcome type: adverse event Reporting: not reported Recurrent hospitalisation (at least one at one year) Outcome type: dichotomous outcome Reporting: not reported Health‐related quality of life (HRQL) Outcome type: continuous outcome Reporting: not reported
Identification	Sponsorship source: ARISTOTLE study funded by Bristol‐Myers Squibb and Pfizer. Analysis was funded by the Duke Clinical Research Institute. Country: International and multicentre. (1034 clinical sites in 39 countries) Setting: secondary care Authors name: Christopher B Granger Institution: Duke Clinical Research Institute Email: schuyler.jones@dm.duke.edu Address: Division of Cardiology, Duke Clinical Research Institute, Durham, NC
Notes
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomization was performed with the use of a 24 h central computerised and interactive voice‐response system.
Allocation concealment (selection bias)	Low risk	Quote: "Randomization was stratified according to whether patients had received warfarin previously and according to clinical site." Quote from trial design: "Subjects who were on warfarin before randomization discontinued the drug 72 hours before randomization and were not dosed with study drug until the INR was b2.0. Randomization is stratified by investigative site and prior warfarin use status (experienced or naïve)."
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "Warfarin (or matching placebo)" Quote: "Apixaban or matching placebo was administered" Judgement comment: double blinded study; warfarin or apixaban was used with matching placebo.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "The primary and secondary efficacy and safety outcomes were adjudicated on the basis of prespecified criteria by a clinical‐events committee whose members were not aware of study‐group assignments."
Incomplete outcome data (attrition bias) All outcomes	High risk	Quote: "Thirteen patients were excluded from the efficacy end points analysis because PCI occurred after the efficacy censoring date (not at risk of efficacy end points). Forty patients excluded from the safety end points analysis because PCI occurred after the safety censoring date (study drug discontinuation)."
Selective reporting (reporting bias)	High risk	Judgement comment: All outcomes were prespecified in the published protocol. The subset analysis focused on participants who had a PCI.
Other bias	High risk	A post hoc subgroup analysis, and thus were subject to both measured and unmeasured confounding.

Open in a new tab

PIONEER AF‐PCI 2016.

*Study characteristics*
Methods	Study design: randomised controlled trial Study grouping: parallel group Recruitment period: May 2013 to July 2015 Follow‐up: 12 months Treatment duration: 12 months
Participants	Baseline characteristics Rivaroxaban 10 mg to 15 mg once daily Number randomised: 709 Number analysed: 696 Gender (female %): 25.5% Age in years: mean: 70.4 ± 9.1 Ethnicity: white 93.4%, black 1%, Asian 3.5%, American Indian or Alaska native 0.1%, other or unknown 2% Loss to follow‐up: 146 CHA₂DS₂‐VASc score < 2: 77 CHA₂DS₂‐VASc score 2 to 4: 375 CHA₂DS₂‐VASc score > 4: 257 Indication for NOAC: non‐valvular atrial fibrillation: 100% Indication for PCI: 12.3% STEMI, 18.5% NSTEMI, 20.7% unstable angina Rivaroxaban 2.5 mg twice daily Number randomised: 709 Number analysed: 706 Gender (female %): 24.5% Age in years: mean: 70.0 ± 9.1 Ethnicity: white 94.6%, black 0.4%, Asian 3.9%, American Indian or Alaska native 0%, other or unknown 1.0% Loss to follow‐up: 149 CHA₂DS₂‐VASc score < 2: 75 CHA₂DS₂‐VASc score 2 to 4: 368 CHA₂DS₂‐VASc score > 4: 266 Indication for NOAC: non‐valvular atrial fibrillation 100% Indication for PCI: 13.8% STEMI, 18.3% NSTEMI, 21.1% unstable angina Warfarin once daily Number randomised: 706 Number analysed: 697 Gender (female %): 26.6% Age in years: mean: 69.9 ± 8.7 Ethnicity: white 94.1%, black 0.1%, Asian 4.7%, American Indian or Alaska native 0%, other or unknown 1.1% Loss to follow‐up: 205 CHA₂DS₂‐VASc score < 2: 51 CHA₂DS₂‐VASc score 2 to 4: 418 CHA₂DS₂‐VASc score > 4: 237 Indication for NOAC: non‐valvular atrial fibrillation: 100% Indication for PCI: 10.7% STEMI, 17.8% NSTEMI, 23.7% unstable angina Overall Number randomised: 2124 Number analysed: 2099 Gender (female %): no data Age in years: mean: no data Ethnicity: no data Loss to follow‐up: no data CHA₂DS₂‐VASc score < 2: no data CHA₂DS₂‐VASc score 2 to 4: no data CHA₂DS₂‐VASc score > 4: no data Indication for NOAC: 100% non‐valvular atrial fibrillation Indication for PCI: STEMI (Group 1: 86/701 (12.3%), Group 2: 97/703 (13.8%), Group 3: 74/691 (10.7%)); NSTEMI (Group 1: 130/701 (18.5%), Group 2: 129/703 (18.3%), Group 3: 123/691 (17.8%)); unstable angina (Group 1: 145/701 (20.7%), Group 2: 148/703 (21.1%), Group 3: 164/691 (23.7%)) Inclusion criteria: 1. Subjects must ≥18 years of age; 2. Must have undergone a PCI (with stent placement) for primary atherosclerotic disease; 3. Must have documented medical history of atrial fibrillation defined as: electrocardiogram, Holter monitor, pacemaker/defibrillator, or any device that provides a rhythm strip documenting paroxysmal, persistent, or permanent nonvalvular AF within 1 year before screening. OR electrocardiogram, Holter monitor, pacemaker/defibrillator, or any device that provides a rhythm strip documenting paroxysmal, persistent, or permanent nonvalvular AF that was performed more than 1 year before screening if the subject was receiving oral anticoagulation therapy (VKA or a novel oral anticoagulant) for the AF for 3 months immediately before the index PCI; 4. If a woman, before entry she must be postmenopausal, defined as > 45 years of age with amenorrhoea for at least 18 months, or surgically sterile (have had a hysterectomy or bilateral oophorectomy, tubal ligation, or otherwise be incapable of pregnancy), or if heterosexually active and menstrual, practicing a highly effective method of birth control. Women must agree to continue using methods of contraception throughout the study; 5. Women in childbearing age must have a negative human chorionic gonadotropin pregnancy test at screening. Serum pregnancy testing may be performed if required by local regulation. Exclusion criteria: Major exclusion criteria include any condition that contraindicates anticoagulant therapy, or would confer an unacceptable risk of bleeding, such as, but not limited to: 1. Has any condition that in the opinion of the investigator, contraindicates anticoagulant therapy or would have an unacceptable risk of bleeding, such as, but not limited to, the following: active internal bleeding, clinically significant bleeding, bleeding at a non‐compressible site, or bleeding diathesis within 30 days before randomization, platelet count < 90,000/µL at screening or pre‐randomisation, history of intracranial haemorrhage, clinically significant gastrointestinal bleeding within 12 months before randomisation, except for subjects who are taking a VKA at the time of screening, a PT test result that is higher than the upper limit of normal at the time of screening that suggests underlying coagulation disorder, any other condition known to increase the risk of bleeding; 2. Has a history of stroke or TIA; 3. Has cardiogenic shock at the time of randomisation; 4. Has ventricular arrhythmias refractory to treatment at the time of randomisation; 5. Has calculated CrCl < 30 mL/min at screening or pre‐randomisation; 6. Has known significant liver disease (e.g. acute hepatitis, chronic active hepatitis, cirrhosis), or liver function test (LFT) abnormalities at screening (confirmed with repeat testing): alanine transaminase (ALT) > 5 times the upper limit of normal or ALT > 3 times the upper limit of normal plus total bilirubin > 2 times the upper limit of normal; 7. Has anaemia of unknown cause with a haemoglobin level < 10 g/dL (< 6.21 mmol/L) at screening or pre‐randomisation; 8. Has a known clinical history of human immunodeficiency virus (HIV) infection; 9. Has a current substance abuse (drug or alcohol) problem or a history within the previous 6 months; 10. Has any severe condition that would limit life expectancy to less than 12 months; 11. Has had major surgery, biopsy of a parenchymal organ, or serious trauma (including head trauma) within the past 30 days; 12. Has a suspected or documented stent thrombosis during the index procedure, or has a PCI with stent placement for a previously stented lesion (stent within a stent) during the index procedure or within the previous 12 months; 13. Has an incomplete staged PCI procedure (once the completion of the staged procedure has occurred, the final PCI may become the index event and is allowed); 14. Has a CABG planned; 15. Has contraindications to the use of VKAs, ASA, or P2Y12 platelet inhibitors (clopidogrel, prasugrel, or ticagrelor), per prescribing information; 16. Has transient AF caused by a reversible disorder (e.g. thyrotoxicosis, pulmonary embolism, recent surgery); 17. Has condition(s) other than paroxysmal, persistent, or permanent non‐valvular AF requiring long‐term anticoagulation with VKAs during the conduct of the study, including, but not limited to, moderate to severe mitral valve stenosis, mechanical heart valves, deep vein thrombosis, pulmonary embolism, or left ventricular thrombus; 18. Is receiving systemic treatment with strong inhibitors of both cytochrome P450 (CYP) 3A4 and p‐glycoprotein (P‐gp; e.g. the azole‐antimycotic ketoconazole and the HIV protease inhibitor ritonavir). Treatment with the azole‐antimycotic fluconazole is allowed; 19. Has known allergies, hypersensitivity, or intolerance to rivaroxaban or its excipients; 20. Uses disallowed therapies such as: VKAs (other than for subjects randomly assigned to the VKA strategy), heparin, low molecular weight heparin, dabigatran, and FX inhibitors other than rivaroxaban study drug, any dose of ASA after randomisation to Group 1 (rivaroxaban 15 mg once daily treatment strategy), doses of ASA greater than 100 mg per day after randomisation in the rivaroxaban 2.5 mg twice daily and VKA treatment strategies, systemic treatment with drugs that are combined P‐gp and strong CYP3A4 inhibitors (e.g. ketoconazole, itraconazole, lopinavir/ritonavir, ritonavir, indinavir/ritonavir, and conivaptan), and systemic treatment with drugs that are combined P‐gp and strong CYP3A4 inducers (e.g. carbamazepine, phenytoin, rifampin, St. John’s wort); 21. Has an anticipated need for chronic (more than 4 weeks) therapy with nonsteroidal anti‐inflammatory drugs; 22. Has received an investigational drug (including investigational vaccines) or used an invasive investigational medical device within 30 days before the planned first dose of study drug or is currently enrolled in an investigational study; 23. Is a woman who is pregnant or breast‐feeding or planning to become pregnant while enrolled in this study; 24. Has any active malignancy; 25. Has any condition that in the opinion of the investigator, would make participation not in the best interest (e.g. compromise the well‐being) of the subject or that could prevent, limit, or confound the protocol‐specified assessments; 26. Is an employee of the investigator or study site, with direct involvement in the proposed study or other studies under the direction of that investigator or study site, as well as family members of the employees or the investigator
Interventions	Intervention characteristics Rivaroxaban 10 mg to 15 mg once daily Concomitant medications: P2Y12 inhibitor (clopidogrel 93%, prasugrel 2%, ticagrelor 5%) x 12 months Type of stent: drug‐eluting stent 65.4%, bare‐metal stent 32.6% Rivaroxaban 2.5 mg twice daily Concomitant medications: DAPT(Aspirin + P2Y12 inhibitor (clopidogrel 93%, prasugrel 2%, ticagrelor 5%)) x (1, 6, or 12) months Type of stent: drug‐eluting stent 66.8%, bare‐metal stent 31.2% Warfarin once daily Concomitant medications: DAPT (Aspirin + P2Y12 inhibitor (clopidogrel 93%, prasugrel 2%, ticagrelor 5%)) x (1, 6, or 12) months Type of stent: drug‐eluting stent 66.5%, bare‐metal stent 31.8%
Outcomes	Death from cardiovascular causes Outcome type: dichotomous outcome Reporting: fully reported Unit of measure: person Direction: lower is better Data value: endpoint Myocardial infarction Outcome type: dichotomous outcome Reporting: fully reported Unit of measure: person Direction: lower is better Data value: endpoint Stroke Outcome type: dichotomous outcome Reporting: fully reported Unit of measure: person Direction: lower is better Data value: endpoint Major bleeding Outcome type: adverse event Reporting: fully reported Unit of measure: person Direction: lower is better Data value: endpoint Death from any cause Outcome type: dichotomous outcome Reporting: fully reported Direction: lower is better Data value: endpoint Stent thrombosis Outcome type: dichotomous outcome Reporting: fully reported Unit of measure: person Direction: lower is better Data value: endpoint Any non‐major TIMI bleeding Outcome type: adverse event Reporting: fully reported Unit of measure: person Direction: lower is better Data value: endpoint Recurrent hospitalisation (at least one at one year) Outcome type: dichotomous outcome Reporting: fully reported Unit of measure: person Direction: lower is better Data value: endpoint Health‐related quality of life (HRQL) Outcome type: continuous outcome Reporting: not reported Direction: higher is better Data value: endpoint
Identification	Sponsorship source: supported by Janssen Scientific Affairs and Bayer Pharmaceuticals Country: 426 sites in 26 countries Setting: International, multicenter Authors name: Michael Gibson Institution: Beth Israel Deaconess Medical Center Email: mgibson@perfuse.org Address: Cardiovascular Division, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA
Notes
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "Subjects are randomly assigned a treatment strategy by an interactive voice/Web response system on day 1." Quote: "Randomization must occur after the INR is 2.5 or below within 72 hours after sheath removal and may be performed while the subject is in the hospital or at the study site if after hospital discharge."
Allocation concealment (selection bias)	Low risk	Quote: "prespecify the intended duration of DAPT (1, 6, or 12 months) and the intended use of an alternate P2Y12 inhibitor (prasugrel or ticagrelor) instead of clopidogrel before randomization." Quote: "data. Randomization and treatment protocol, randomization in equal proportion to 1 of 3 treatment strategies is stratified by the intended duration of DAPT (1, 6, or 12 months)" Quote: "by randomly permuted blocks"
Blinding of participants and personnel (performance bias) All outcomes	High risk	Judgement comment: open label design; different treatment length for DAPT; compliance assessed by pill counts
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "study team will remain blinded to treatment information until database lock." Judgement comment: study team was blinded to treatment information until database lock.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "Two analysis sets are to be used: the ITT analysis set and the safety analysis set. All primary analyses in the trial will be based on the safety analysis set, which consists of all randomised subjects who receive at least 1 dose of study drug. The ITT analysis set includes all randomised" Judgement comment: all dropouts were explained, assigned to their groups, and accounted for; ITT analysis and safety analysis were performed to include all randomised participants
Selective reporting (reporting bias)	Low risk	Judgement comment: all outcomes were reported according to a priori published protocol
Other bias	Low risk	Judgement comment: no sources of other bias identified; source of funding and conflicts of interests were clearly declared.

Open in a new tab

RE‐DUAL PCI 2017.

*Study characteristics*
Methods	Study design: randomised controlled trial Study grouping: parallel group Follow‐up: 14 months Recruitment period: 21 July 2014 to 31 October 2016 Treatment duration: 12.3 months
Participants	Baseline characteristics Dabigatran 110 mg Number randomised: 981 Number analysed: 972 Gender (female %): 25.8 Age in years: mean: 71.5 ± 8.9 Ethnicity: not mentioned Loss to follow‐up: 20 CHA₂DS₂‐VASc score < 2: not mentioned CHA₂DS₂‐VASc score 2 to 4: not mentioned CHA₂DS₂‐VASc score > 4: not mentioned Indication for NOAC: nonvalvular atrial fibrillation Indication for PCI: 51.9% ACS, 44.1% stable angina or positive stress test, 15.9% staged procedure Type of AF: 17.7% persistent, 32.6% permanent, 49.6% paroxysmal Dabigatran 150 mg Number randomised: 763 Number analysed: 758 Gender (female %): 22.4 Age in years: mean: 68.6 ± 7.7 Ethnicity: not mentioned Loss to follow‐up: 4 CHA₂DS₂‐VASc score < 2: not mentioned CHA₂DS₂‐VASc score 2 to 4: not mentioned CHA₂DS₂‐VASc score > 4: not mentioned Indication for NOAC: nonvalvular atrial fibrillation Indication for PCI: 51.2% ACS, 41.9% stable angina or positive stress test, 18.1% staged procedure, 8.5% other Type of AF: 17.3% persistent, 32.8% permanent, 49.8% paroxysmal Warfarin corresponding to dabigatran 110 mg Number randomised: 981 Number analysed: 948 Gender (female %): 23.5 Age in years: mean: 71.7 ± 8.9 Ethnicity: not mentioned Loss to follow‐up: 38 CHA₂DS₂‐VASc score < 2: not mentioned CHA₂DS₂‐VASc score 2 to 4: not mentioned CHA₂DS₂‐VASc score > 4: not mentioned Indication for NOAC: nonvalvular atrial fibrillation Indication for PCI: 48.4% ACS, 43.7% stable angina or positive stress test, 17.1% staged procedure, 6.3% other Type of AF: 18.2% persistent, 32.4% permanent, 49.4% paroxysmal Warfarin corresponding to dabigatran 150 mg Number randomised: 981 Number analysed: 764 Gender (female %): 22.3 Age in years: mean: 68.8 ± 7.7 Ethnicity: not mentioned Loss to follow‐up: not mentioned CHA₂DS₂‐VASc score < 2: not mentioned CHA₂DS₂‐VASc score 2 to 4: not mentioned CHA₂DS₂‐VASc score > 4: not mentioned Indication for NOAC: nonvalvular atrial fibrillation Indication for PCI: 48.3% ACS, 44.4% stable angina or positive stress test, 17.5% staged procedure, 6.5% other Type of AF: 19.5% persistent, 31.2% permanent, 49.3% paroxysmal Overall Number randomised: 2725 Number analysed: 110 mg dual‐therapy group: 972; 150 mg dual‐therapy group: 758; triple therapy with warfarin: 948 Gender: 110 mg dual‐therapy group: 74.2% male, 25.8% women; 150 mg dual‐therapy group: 77.6% male, 22.4% women; triple therapy group with warfarin: 76.5% male, 23.5% women Age in years: mean: 110 mg dual‐therapy group: 71.5 ± 8.9; 150 mg dual‐therapy group: 68.6 ± 7.7; triple therapy with warfarin): 71.7 ± 8.9 Ethnicity: not mentioned Loss to follow‐up: 0.2% CHA₂DS₂‐VASc score < 2: not mentioned CHA₂DS₂‐VASc score 2 to 4: not mentioned CHA₂DS₂‐VASc score > 4: not mentioned Indication for NOAC: nonvalvular atrial fibrillation Indication for PCI: STEMI: 91 (9.3%) in dabigatran 110 mg dual‐therapy group, and 80 (8.2%) in corresponding triple therapy group; 74 (9.7%) in dabigatran 150 mg dual‐therapy group, and 62 (8.1%) in corresponding triple therapy group. NSTEMI: 203 (20.7%) in dabigatran 110 mg dual‐therapy group, and 206 (21.0%) in corresponding triple therapy group; 179 (23.5%) in dabigatran 150 mg dual‐therapy group, and 151 (19.8%) in corresponding triple therapy group. Unstable angina pectoris: 195 (19.9%) in dabigatran 110 mg dual‐therapy group, and 166 (16.9%) in corresponding triple therapy group. 126 (16.5%) in dabigatran 150 mg dual‐therapy group, and 138 (18.1%) in corresponding triple therapy group. Type of AF: not mentioned Inclusion criteria: Age ≥ 18 years, non‐valvular atrial fibrillation, PCI within the previous 120 hours for ACS or stable or coronary disease Exclusion criteria: mechanical or biological heart valve prosthesis • cardiogenic shock during current hospitalisation • use of fibrinolytic agents within 24 hr of randomisation that will put the individual at high risk of bleeding (in the opinion of the investigator) • stroke within 1 month prior to screening • major surgery within 1 month prior to screening • organ transplant, or on the waiting list for organ transplant • history of intraocular, spinal, retroperitoneal, or traumatic intra‐articular bleeding, unless the causative factor has been permanently resolved • GI haemorrhage within 1 month before screening, unless the causative factor has been permanently resolved • a major bleeding episode including life‐threatening bleeding within 1 month before screening • haemorrhagic disorder or bleeding diathesis • anaemia or thrombocytopenia • severe renal impairment (eCrCl < 30 mL/min) • active liver disease (ALT, AST, AP > 3× ULN or known active hepatitis A, B, or C) • recent malignancy or radiation therapy (≤ 6 months), unless life expectancy is > 36 months • continued treatment with systemic ketoconazole, itraconazole, posaconazole, cyclosporine, tacrolimus, dronedarone, rifampicin, phenytoin, carbamazepine, or St John’s wort • continued treatment with NSAIDs • known allergy to dabigatran or warfarin, or excipients of a study drug • patients who should not be treated with an OAC • contraindication to clopidogrel, ticagrelor, or ASA • premenopausal women who are pregnant, breast‐feeding, not surgically sterile, or not practicing 2 acceptable methods of birth control • participation in another trial with an investigational drug or device within the past 30 days • patients unable or unwilling to comply with the protocol, or with life expectancy shorter than the duration of the study
Interventions	Intervention characteristics Dabigatran 110 mg Concomitant medications: clopidogrel or ticagrelor Type of stent: drug‐eluting 82.1%, bare‐metal 15.1%, drug‐eluting and bare‐metal 1.9%, other 0.8% Dabigatran 150 mg Concomitant medications: clopidogrel or ticagrelor Type of stent: drug‐eluting 81.5%, bare‐metal 16.1%, drug‐eluting and bare‐metal 1.3%, other 1% Warfarin corresponding to dabigatran 110 mg Concomitant medications: aspirin (≤ 100 mg daily), and either clopidogrel or ticagrelor. Aspirin was discontinued after 1 month in participants in whom a bare‐metal stent was implanted, and after 3 months in participants in whom a drug‐eluting stent was implanted. Type of stent: drug‐eluting 84.6%, bare‐metal 13.6%, drug‐eluting and bare‐metal 1.2%, other 0.5% Warfarin corresponding to dabigatran 150 mg Concomitant medications: aspirin (≤ 100 mg daily), and either clopidogrel or ticagrelor. Aspirin was discontinued after 1 month in participants in whom a bare‐metal stent was implanted, and after 3 months in participants in whom a drug‐eluting stent was implanted. Type of stent: drug‐eluting 84.1%, bare‐metal 14.1%, drug‐eluting and bare‐metal 1.2%, other 0.7%
Outcomes	Death from cardiovascular causes Outcome type: dichotomous outcome Reporting: not reported Direction: lower is better Myocardial infarction Outcome type: dichotomous outcome Reporting: fully reported Direction: lower is better Stroke Outcome type: dichotomous outcome Reporting: fully reported Direction: lower is better Major bleeding Outcome type: adverse event Reporting: fully reported Direction: lower is better Death from any cause Outcome type: dichotomous outcome Reporting: fully reported Direction: lower is better Stent thrombosis Outcome type: dichotomous outcome Reporting: fully reported Direction: lower is better Any non‐major TIMI bleeding Outcome type: adverse event Reporting: fully reported Direction: lower is better Recurrent hospitalisation (at least one at one year) Outcome type: dichotomous outcome Reporting: not reported Direction: lower is better Health‐related quality of life (HRQL) Outcome type: continuous outcome Reporting: not reported Direction: higher is better
Identification	Sponsorship source: supported by Boehringer Ingelheim Country: international, multicenter (414 sites in 41 countries) Setting: secondary care Authors name: Dr. Christopher Cannon Institution: Baim Institute for Clinical Research Email: christopher.cannon@baiminstitute.org Address: 930 Commonwealth Ave., Boston, MA, 02215
Notes
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote from protocol: "The randomisation list will be generated using a validated system and a pseudorandom number generator so that the resulting treatment will be both reproducible and nonpredictable."
Allocation concealment (selection bias)	Low risk	Judgement comment: randomisation was performed with the use of permuted blocks Quote: "Randomization was performed with the use of permuted blocks, with stratification according to age group (non elderly or elderly (< 80 or ≥ 80 years of age; < 70 or ≥ 70 years of age in Japan)), and region (United States, Japan, or other countries)."
Blinding of participants and personnel (performance bias) All outcomes	High risk	Judgement comment: open label design. different treatment regimens between intervention and control
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "all primary and secondary end‐point events were adjudicated by an independent committee whose members were unaware of the treatment assignments." Quote: "All clinical end‐point events were adjudicated by an independent committee whose members were unaware of the treatment assignments."
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "The primary analysis, which was performed on an intention‐to‐treat basis, included all patients who underwent randomization, regardless of whether they received treatment. A sensitivity analysis, which was performed on an on‐treatment basis, included all patients who had received at least one dose of the trial anticoagulant; data on events that occurred more than 7 days after the trial anticoagulant was permanently discontinued were censored."
Selective reporting (reporting bias)	Low risk	Quote: "The authors vouch for the accuracy and completeness of the data and analyses and the adherence of the trial to the protocol."
Other bias	Low risk	Judgement comment: industry funded study. source of funding and conflicts of interests documented. no other sources of bias identified.

Open in a new tab

Sherwood 2016.

*Study characteristics*
Methods	Study design: randomised controlled trial Study grouping: parallel group Follow‐up: 806 days of follow‐up Recruitment period: 18 December to 17 June 2009 Treatment duration: unknown
Participants	Baseline characteristics Rivaroxaban 15 mg to 20 mg once daily Number randomised: 61 Number analysed: 61 Gender (female %): not mentioned Age in years: mean: not mentioned Ethnicity: not mentioned Loss to follow‐up: not mentioned CHA₂DS₂‐VASc score < 2: not mentioned CHA₂DS₂‐VASc score 2 to 4: not mentioned CHA₂DS₂‐VASc score > 4: not mentioned Indication for NOAC: atrial fibrillation Indication for PCI: not mentioned Type of AF: not mentioned Warfarin once daily Number randomised: 92 Number analysed: 92 Gender (female %): not mentioned Age in years: mean: not mentioned Ethnicity: not mentioned Loss to follow‐up: not mentioned CHA₂DS₂‐VASc score < 2: not mentioned CHA₂DS₂‐VASc score 2 to 4: not mentioned CHA₂DS₂‐VASc score > 4: not mentioned Indication for NOAC: not mentioned Indication for PCI: not mentioned Type of AF: not mentioned Overall Number randomised: 153 Number analysed: 153 Gender (female %): 17.6% Age in years: mean: 73 (67, 79) Ethnicity: not mentioned Loss to follow‐up: not mentioned CHA₂DS₂‐VASc score < 2: not mentioned CHA₂DS₂‐VASc score 2 to 4: not mentioned CHA₂DS₂‐VASc score > 4: not mentioned Indication for NOAC: atrial fibrillation Indication for PCI: not mentioned Type of AF: persistent 75.8%, paroxysmal 22.2%, newly diagnosed 2% Inclusion criteria: men or women aged ≥18 years with non‐valvular atrial fibrillation • atrial fibrillation must be documented by ECG evidence (e.g. 12‐lead ECG, rhythm strip, Holter, pacemaker interrogation) within 30 days before randomisation; subjects must have medical evidence of atrial fibrillation within 1 year before and at least one day before the qualifying ECG evidence – this could be obtained from a notation in the subject's record (e.g. medical chart, hospital discharge summary) • subjects with newly diagnosed atrial fibrillation are eligible provided that – there is evidence that the atrial fibrillation is non‐valvular, cardioversion is not planned, there is ECG evidence on 2 occasions 24 hours apart demonstrating atrial fibrillation • history of prior ischaemic stroke, TIA, or non‐CNS systemic embolism believed to be cardio‐embolic in origin, or has 2 or more of the following risk factors: heart failure, left ventricular ejection fraction ≤ 35%, or both; hypertension (defined as use of antihypertensive medications within 6 months before the screening visit, or persistent systolic blood pressure above 140 mmHg, or diastolic blood pressure above 90 mmHg); age ≥ 75 years; diabetes mellitus (defined as a history of type 1 or type 2 diabetes mellitus, or use of antidiabetic medications within 6 months before screening visit); • female subjects must be postmenopausal (for at least 2 years), surgically sterile, abstinent, or if sexually active, be practicing an effective method of birth control (e.g. prescription oral contraceptives, contraceptive injections, intrauterine device, double‐barrier method, contraceptive patch, male partner sterilisation) before entry and throughout the study; and for those of childbearing potential, have a negative serum β‐hCG pregnancy test at screening; • subjects must have signed an informed consent document indicating that they understand the purpose of and procedures required for the study and are willing to participate in the study • in order to participate in the optional pharmacogenomic component, subjects must have signed the informed consent for DNA research document indicating willingness to participate in the pharmacogenomics component of the study (where local regulations permit) Exclusion criteria: cardiac‐related conditions – haemodynamically significant mitral valve stenosis; prosthetic heart valve (annuloplasty with or without prosthetic ring, commissurotomy, valvuloplasty, or both are permitted); planned cardioversion (electrical or pharmacological); transient atrial fibrillation caused by a reversible disorder (e.g. thyrotoxicosis, PE, recent surgery, MI); known presence of atrial myxoma or left ventricular thrombus; active endocarditis. Haemorrhage risk‐related criteria – active internal bleeding; history of, or condition associated with, increased bleeding risk including, but not limited to: major surgical procedure or trauma within 30 days before the randomisation visit; clinically significant gastrointestinal bleeding within 6 months before the randomisation visit; history of intracranial, intraocular, spinal, or atraumatic intra‐articular bleeding; chronic haemorrhagic disorder; known intracranial neoplasm, arteriovenous malformation, or aneurysm; planned invasive procedure with potential for uncontrolled bleeding, including major surgery; platelet count < 90,000/μL at the screening visit; sustained uncontrolled hypertension: systolic blood pressure ≥180 mmHg or diastolic blood pressure ≥ 100 mmHg. Concomitant conditions and therapies – severe, disabling stroke (modified Rankin score of 4 to 5) within 3 months, or any stroke within 14 days before the randomisation visit; transient ischaemic attack within 3 days before the randomisation visit; indication for anticoagulant therapy for a condition other than atrial fibrillation (e.g. VTE); treatment with aspirin > 100 mg daily, aspirin in combination with thienopyridines within 5 days before randomisation; intravenous antiplatelets within 5 days before randomisation; fibrinolytics within 10 days before randomisation (note: aspirin ≤100 mg monotherapy is allowed and thienopyridine monotherapy is allowed); anticipated need for chronic treatment with a non‐steroidal anti‐inflammatory drug; systemic treatment with a strong inhibitor of cytochrome P450 3A4, such as ketoconazole or protease inhibitors, within 4 days before randomisation, or planned treatment during the time period of the study; treatment with a strong inducer of cytochrome P450 3A4, such as rifampin or rifampicin, within 4 days before randomisation, or planned treatment during the time period of the study; anaemia (haemoglobin < 10 g/dL) at the screening visit; pregnancy or breast‐feeding; any other contraindication to warfarin; known HIV infection at time of screening; calculated CLCR < 30 mL/min at the screening visit ; known significant liver disease (e.g. acute clinical hepatitis, chronic active hepatitis, cirrhosis), or ALT > 3x the ULN Study Participation and Follow‐up‐Related criteria; serious concomitant illness associated with a life expectancy of less than 2 years; drug addiction or alcohol abuse within 3 years before the randomisation visit; have received an experimental drug or used an experimental medical device within 30 days before the planned start of treatment; previous randomisation in the present study or other study of rivaroxaban; known allergy or hypersensitivity to any component of rivaroxaban, warfarin, or placebo excipients (includes lactose, microcrystalline cellulose, magnesium stearate, hypromellose, macrogol, croscarmellose sodium, sodium lauryl sulfate, titanium oxide or ferric oxide red, titanium dioxide or ferric oxide red, anhydrouslactose, pregelatinized starch, FD & C Red #6 barium lake, FD & C Yellow #10 aluminium lake, FD & C Blue #1 aluminium lake, FD & C Yellow #6 aluminium lake, cornstarch, lactose monohydrate); inability or unwillingness to comply with study‐related procedures; employees of the investigator or study centre, with direct involvement in the proposed study or other studies under the direction of that investigator or study centre, as well as family members of the employees or the investigator.
Interventions	Intervention characteristics Concomitant medications: DAPT (aspirin and clopidogrel) was used in 37% of overall participants, single antiplatelet therapy (aspirin or clopidogrel) in 34%, and 15% of participants received no antiplatelet therapy Rivaroxaban 15 mg to 20 mg once daily Type of stent: 34% drug‐eluting stent, 41% bare‐metal stent Warfarin once daily Type of stent: 35% drug‐eluting stent, 47% bare‐metal stent
Outcomes	Death from cardiovascular causes Outcome type: dichotomous outcome Reporting: partially reported Direction: lower is better Myocardial infarction Outcome type: dichotomous outcome Reporting: partially reported Direction: lower is better Stroke Outcome type: dichotomous outcome Reporting: partially reported Direction: lower is better Major bleeding Outcome type: adverse event Reporting: partially reported Direction: lower is better Death from any cause Outcome type: dichotomous outcome Reporting: not reported Stent thrombosis Outcome type: dichotomous outcome Reporting: not reported Any non‐major TIMI bleeding Outcome type: adverse event Reporting: not reported Recurrent hospitalisation (at least one at one year) Outcome type: dichotomous outcome Reporting: not reported Health‐related quality of life (HRQL) Outcome type: continuous outcome Reporting: not reported
Identification	Sponsorship source: Supported by Johnson & Johnson Pharmaceutical Research & Development (Raritan, New Jersey) and Bayer HealthCare AG (Leverkusen, Germany). Country: International multicentre trial (1178 participating sites in 45 countries) Setting: Inpatient Authors name: Matthew W. Sherwood Institution: Duke Clinical Research Institute Email: matthew.sherwood@dm.duke.edu Address: Duke Clinical Research Institute, Durham, North Carolina
Notes
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "multi‐center, randomised, double‐blind, double‐dummy, event‐driven trial" Quote: "Randomization was performed with the use of a central 24‐hour, computerized, automated voice‐response system."
Allocation concealment (selection bias)	Low risk	Judgement comment: randomisation was performed with the use of a central 24‐hour, computerised, automated voice‐response system.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quote: "Despite the double‐blind, double‐dummy design of the ROCKET AF trial, there was a significant difference in rates of PCI between treatment arms. Patients treated with rivaroxaban were significantly less likely to undergo PCI compared with those on warfarin." Quote: "A point‐of‐care device was used to generate encrypted values that were sent to an independent study monitor, who provided sites with either real INR values (for patients in the warfarin group in order to adjust the dose) or sham values (for patients in the rivaroxaban group receiving placebo warfarin) during the course of the trial. Sham INR results were generated by means of a validated algorithm reflecting the distribution of values in warfarin‐treated patients with characteristics similar to those in the study population." Quote: "Events occurring at the end of the study were probably related to increased difficulty in achieving the transition from blinded trial therapy to the open‐label use of a vitamin K antagonist when the patient had previously been assigned to the rivaroxaban group, since presumably many patients who had previously been assigned to the warfarin group would have already had a therapeutic INR." Quote: "Patients in each group also received a placebo tablet in order to maintain blinding."
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "An independent clinical end‐point committee applied protocol definitions to adjudicate all suspected cases of stroke, systemic embolism, myocardial infarction, death, and bleeding events that contributed to the prespecified end points." Judgement comment: all reported outcomes are objective events
Incomplete outcome data (attrition bias) All outcomes	Low risk	Quote: "performed post hoc analyses of events in the intention‐to‐treat population" Quote: "The primary analysis was prespecified to be performed in the per‐protocol population, which included all patients who received at least one dose of a study drug, did not have a major protocol violation, and were followed for events while receiving a study drug or within 2 days after discontinuation"
Selective reporting (reporting bias)	High risk	Quote: "Pertinent national regulatory authorities and ethics committees at participating centers approved the protocol, which is available with the full text of this article at NEJM.org." Judgement comment: all outcomes prespecified in the protocol were reported
Other bias	High risk	A post hoc subgroup analysis, and thus were subject to both measured and unmeasured confounding. Industry‐funded study. Source of funding and conflicts of interest declared appropriately. Quote: "The study was supported by Johnson & Johnson Pharmaceutical Research and Development and Bayer HealthCare. The Duke Clinical Research Institute coordinated the trial, managed the database, and performed the primary analyses independently of the sponsors." The Duke Clinical Research Institute coordinated the trial, managed the database, and performed the primary analyses independently of the sponsors.

Open in a new tab

NOAC: non‐vitamin K antagonist oral anticoagulants

PCI: percutaneous coronary intervention

ACS: acute coronary syndrom

CCS: chronic coronary syndrom

HAS‐BLED: a scoring system developed to assess 1‐year risk of major bleeding in patients taking anticoagulants with atrial fibrillation

DAPT: dual antiplatelet therapy

TIA: transient ischemic attack

VKA: vitamin K antagonist

Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Ako 2019	Wrong intervention
Carrie 2017	Wrong intervention
Casamira 2017	Wrong study design
Cavender 2015	Wrong population
Connolly 2018	Wrong population
Cornel 2012	Wrong population
Faggioni 2016	Wrong study design
Hess 2015	Wrong population
Hohnloser 2018	Wrong outcomes
Hoshi 2017	Wrong comparator
Korjian 2018	Wrong outcomes
Lahsaei 2016	Wrong study design
Lip 2018	Wrong outcomes
Matsumura Nakano 2018	Wrong comparator
Musumeci 2014	Wrong study design
NCT01962545	Wrong comparator
NCT03023020	Wrong outcomes
Pinto 2017	Wrong outcomes
Potter 2018	Wrong study design
Povsic 2016	Wrong population
Secemsky 2016	Wrong study design
Sindet Pedersen 2018	Wrong study design
Sra 2016	Wrong study design
tenBerg 2002	Wrong study design
Verheugt 2014	Wrong indication
Vranckx 2013	Wrong comparator
Xanthopoulou 2018	Wrong study design
Xiaoyan 2014	Wrong study design
Yasuda 2018	Wrong comparator
Yoshida 2018	Wrong study design

Open in a new tab

Characteristics of ongoing studies [ordered by study ID]

Gao 2015.

Study name	RT‐AF: rivaroxaban in patients with atrial fibrillation and coronary artery disease undergoing percutaneous coronary intervention
Methods	Allocation: randomised Intervention model: parallel assignment Masking: none (open‐label)
Participants	420 participants Inclusion criteria: history or new onset paroxysmal, persistent, or permanent non‐valvular AF; a long‐term indication for oral anticoagulation treatment (score of CHADS2‐VASc ≥ 2); a severe coronary lesion (at least 75% stenosis on angiography or fractional flow reserve lower than 0.80) with indication for PCI; and age 18 to 80 years. The indication for PCI will be based on established guidelines Exclusion criteria: major exclusion criteria include history of intracranial bleeding; cardiogenic shock; contraindication to using aspirin, clopidogrel, warfarin or rivaroxaban; peptic ulcer in the previous 6 months; thrombocytopenia (platelet concentration lower than 50/L to 109/L); major bleeding (according to the International Society of Thrombosis and Hemostasis (ISTH) definition) in the past 12 months; and pregnancy.
Interventions	rivaroxaban vs warfarin
Outcomes	Major or clinically relevant non‐major bleeding. Time frame: 12 months Composite outcome of death, myocardial infarction, stent thrombosis, and ischaemic stroke. Time frame: 12 months
Starting date	August 2013
Contact information	not mentioned
Notes	clinicaltrials.gov/ct2/show/NCT02334254

Open in a new tab

NCT02789917.

Study name	APPROACH‐ACS‐AF (APPROACH): APixaban vs. PhenpRocoumon in patients with ACS and AF
Methods	Allocation: randomised Intervention model: parallel assignment Masking: none (open‐label)
Participants	400 participants
Interventions	Apixaban 5 mg/dL and phrenprocoumon Concommitent treatment: clopidogrel bisulfate 75 mg
Outcomes	The combined end point of moderate or major bleeding complications during the initial hospitalisation and follow‐up (Bleeding Academic Research Consortium (BARC) type ≥ 2 bleeding (Time frame: up to 6 months after randomisation)
Starting date	June 2016
Contact information	Contact: Reza Wakili, MD +49 89 4400 ext 73036; reza.wakili@med.uni‐muenchen.de Contact: Lisa Riesinger, MD +49 89 4400 ext 0; Lisa.Riesinger@med.uni‐muenchen.de
Notes	NCT02789917

Open in a new tab

NCT03536611.

Study name	A randomised study comparing dabigatran etexilate versus warfarin in Chinese patients with nonvalvular atrial fibrillation who undergo percutaneous coronary intervention with stenting (DES)
Methods	Allocation: randomised Intervention model: parallel assignment Masking: single (outcomes assessor)
Participants	1120 participants Inclusion criteria: Aged ≥18 years; Patients with non‐secondary (e.g. pericarditis, hyperthyroidism, recent surgery, etc.) nonvalvular atrial fibrillation requiring long‐term anticoagulant treatment; Patients who have PCI indications and coronary heart disease that was successfully treated with drug‐eluting stenting (DES); Patients who sign the informed consent form. Exclusion criteria: Patients with mechanical or biological heart valve prosthesis; Patients proposed to undergo left atrial appendage occlusion or atrial fibrillation radiofrequency ablation Cardiogenic shock during current hospitalisation; Patients who have used fibrinolytic agents within 24 hours of randomisation that in the opinion of the Investigator, will put the patient at high risk of bleeding; Stroke within 1 month prior to screening visit; Patients who in the opinion of the Investigator, have had major surgery within the month prior to screening; Patient has received an organ transplant or is on a waiting list for an organ transplant; History of intraocular, spinal, retroperitoneal, or a traumatic intra‐articular bleeding unless the causative factor has been permanently eliminated or repaired (e.g. by surgery); Gastrointestinal (GI) haemorrhage within one month prior to screening, unless, in the opinion of the Investigator, the cause has been permanently eliminated (e.g. by surgery); Major bleeding episode (reduction in the haemoglobin level of at least 2 g/dL, transfusion of at least two units of blood, or symptomatic bleeding in a critical area or organ) including life‐threatening bleeding episode (symptomatic intracranial bleeding, bleeding with a decrease in the haemoglobin level of at least 5 g/dL, or bleeding requiring transfusion of at least 4 units of blood or inotropic agents, or necessitating surgery) in one month prior to screening visit; Haemorrhagic disorder or bleeding diathesis (e.g. von Willebrand disease, haemophilia A or B, or other hereditary bleeding disorder, history of spontaneous intra‐articular bleeding, history of prolonged bleeding after surgery or intervention); Anaemia (haemoglobin < 10 g/dL) or thrombocytopenia, including heparin‐induced thrombocytopenia (platelet count < 100 × 109/L) at screening (Visit 1); Severe renal impairment (estimated CrCl calculated by Cockcroft‐Gault equation) < 30 mL/min at screening; Active liver disease as indicated by at least one of the following: prior and persistent alanine aminotransferase (ALT) or aspartate transaminase (AST), or alkaline phosphatase (AP) > 3 x upper limit of normal (ULN); known active hepatitis C, known active hepatitis B, known active hepatitis A; Recent malignancy or radiation therapy (≤ 6 months), unless, in the opinion of the investigator, the estimated life expectancy is longer than 36 months; Need for continued treatment with systemic ketoconazole, itraconazole, posaconazole, cyclosporine, tacrolimus, dronedarone, rifampicin, phenytoin, carbamazepine, or any cytotoxic or myelosuppressive therapy; Patients, who in the investigator's opinion, need continuous treatment with non‐steroidal anti‐inflammatory drugs (NSAIDs); Patients with a known allergy to dabigatran etexilate or to the excipients used for the capsule of the drug; Patients with a known allergy to warfarin tablets or to the excipients; Patients, who in the investigator's opinion, should not be treated with OAC; Patients with a contraindication to clopidogrel, or ASA; Pre‐menopausal women (last menstruation ≤ 1 year prior to screening), who experienced any of the following conditions: are pregnant or breast feeding; are pregnant surgically sterile; are of child bearing potential and not practising two acceptable methods of birth control, or do not plan to continue practising an acceptable method of birth control throughout the trial (acceptable methods of birth control are oral or parenteral (patch, injection, implant) hormonal contraception, which has been used continuously for at least one month prior to the first dose of study medication, intrauterine device or intrauterine system, double‐barrier method of contraception (condom and occlusive cap or condom and spermicidal agent), male sterilisation and complete sexual abstinence (if acceptable by local authorities)); periodic abstinence is not an acceptable method of contraception
Interventions	Dabigatran etexilate versus warfarin
Outcomes	Primary end point: time to the first occurrence of BARC‐defined (grades 2 to 5) clinically relevant bleeding Key secondary end point: time to the first occurrence of net clinical adverse events. A net clinical adverse event is composed of major cardiovascular and cerebrovascular adverse events (all death, recurrent myocardial infarction, ischaemia‐induced revascularisation of the target vessel, or stroke or systemic embolism), or BARC‐defined (grades 2 to 5) clinically relevant bleeding Secondary end points:1) major cardiovascular and cerebrovascular adverse events; 2) major bleeding or clinically relevant non‐major bleeding (ISTH definition); 3) major bleeding (ISTH definition); 4) any bleeding event (BARC‐defined grades 1 to 5); 5) clinically relevant bleeding (BARC‐defined grades 2 to 5); 6) bleeding (BARC‐defined grades 3 to 5)
Starting date	24 May 2018
Contact information	Ming Liang, PhD
Notes	clinicaltrials.gov/ct2/show/NCT03536611

Open in a new tab

Vranckx 2018.

Study name	ENTRUST‐AF PCI trial: edoxaban treatment versus vitamin K antagonist in patients with atrial fibrillation undergoing percutaneous coronary intervention
Methods	Allocation: randomised Intervention model: parallel assignment Masking: none (open‐label) A multinational, multicenter, randomised, open‐label, phase‐3b study with blinded evaluation of end points by an independent Clinical Event Committee (PROBE design). An independent Data and Safety Monitoring Board (DSMB) is responsible for monitoring safety during the study.
Participants	approximately 150 study sites in Europe and Asia. 1500 patients (750 per antithrombotic regimen)
Interventions	Edoxaban 60 mg (or 30 mg dose adjusted) and warfarin for 12 months Concommitent treatment: clopidogrel bisulfate 75 mg
Outcomes	Number of participants with major or clinically relevant non‐major bleeding, defined by International Society on Thrombosis and Haemostasis (ISTH). Time frame: day 1 to 12 months Number of participants with composite of cardiovascular (CV) death, stroke, systemic embolic events (SEE), myocardial infarction (MI), and definite stent thrombosis. Time frame: day 1 to 12 months Number of participants with major bleeding defined by ISTH. Time frame: day 1 to 12 months
Starting date	24 February 2017
Contact information	Not mentioned
Notes	Vranckx P, Lewalter T, Valgimigli M, Tijssen JG, Reimitz PE, Eckardt L, et al. Evaluation of the safety and efficacy of an edoxaban‐based antithrombotic regimen in patients with atrial fibrillation following successful percutaneous coronary intervention (PCI) with stent placement: Rationale and design of the ENTRUST‐AF PCI trial. American Heart Journal 2018;196:105‐12.

Open in a new tab

Differences between protocol and review

A number of analyses were planned in the protocol but not conducted in the review:

We included the following co‐interventions even if they were part of the randomised treatment: single or dual antiplatelet therapy.
The 'recurrent hospitalisation' outcome was assessed at the longest point of follow‐up for each trial.
Investigation of heterogeneity through subgroup analyses was planned, but abandoned due to the limited number of included studies.
Creation of net heat plots for a visualisation of inconsistencies between direct and indirect evidence was planned, but abandoned due to the limited number of included studies.
Assessment of the risk of publication bias was planned using funnel plots with at least 10 trials to explore possible small‐study biases for the primary outcomes. This was not possible due to the limited number of included studies.
Studies were excluded based on the outcomes reported, if they were substudies of included studies but did not contain any information helpful to our review, or addressed outcomes not specified in our review.
Analysis of health‐related quality of life was planned, but not reported in the included studies.

Contributions of authors

S Al Said conceived, designed, co‐ordinated, and drafted the protocol and review. He screened the studies for inclusion, contributed to data extraction and analysis, and acts as guarantor of the review.

S Alabed designed and drafted the protocol and review. He screened the studies for inclusion, and contributed to data extraction and analysis.

K Kaier and A Tan designed the network meta‐analysis, and provided statistical advice and input.

JJ Meerpohl and C Bode and D Duerschmied provided clinical expertise and general advice, and revised the protocol and review.

All authors approved the final version of the review.

Sources of support

Internal sources

No sources of support supplied

External sources

This project was supported by the National Institute for Health Research (NIHR), via Cochrane Infrastructure funding to the Heart Group and an NIHR Incentive Award (128423). The views and opinions expressed therein are those of the authors and do not necessarily reflect those of the Systematic Reviews Programme, NIHR, National Health Service (NHS), or the Department of Health and Social Care, UK
National Institute for Health Research, UK

S Alabed currently holds an NIHR Academic Clinical Fellowship (ACF)
This project was supported by the Complex Reviews Support Unit, funded by the National Institute for Health Research (project number 14/178/29), UK

Declarations of interest

S Al Said: none known

S Alabed: none known

K Kaier: none known

A Tan: none known

C Bode: received research grants from Bayer, GlaxoSmithKline, and Merck; speaker’s honoraria from Bayer, Bristol‐Myers Squibb/Pfizer, Daiichi Sankyo, Boehringer Ingelheim; and consulting fees from Bayer.

JJ Meerpohl: none known

D Duerschmied: received speaker's honoraria from Bayer Healthcare, Pfizer and Daiichi Sankyo, travel support for national and international cardiology meetings from Bayer Healthcare and Daiichi Sankyo, and support for the organisation of local scientific meetings from Bayer Healthcare, Pfizer, and Daiichi Sankyo.

Edited (no change to conclusions)

References

References to studies included in this review

AUGUSTUS 2019 {published data only}

Lopes RD, Heizer G, Aronson R, Vora AN, Massaro T, Mehran R, et al. Antithrombotic therapy after acute coronary syndrome or PCI in atrial fibrillation. New England Journal of Medicine 2019;380(16):1509-24. [DOI: 10.1056/NEJMoa1817083] [DOI] [PubMed] [Google Scholar]
Lopes RD, Vora AN, Liaw D, Granger CB, Darius H, Goodman SG, et al. An open-Label, 2 x 2 factorial, randomized controlled trial to evaluate the safety of apixaban vs. vitamin K antagonist and aspirin vs. placebo in patients with atrial fibrillation and acute coronary syndrome and/or percutaneous coronary intervention: rationale and design of the AUGUSTUS trial. American Heart Journal 2018;200:17-23. [DOI: 10.1016/j.ahj.2018.03.001] [DOI] [PubMed] [Google Scholar]
NCT02415400. A study of apixaban in patients with atrial fibrillation, not caused by a heart valve problem, who are at risk for thrombosis (blood clots) due to having had a recent coronary event, such as a heart attack or a procedure to open the vessels of the heart. clinicaltrials.gov/show/nct02415400 (first posted 14 April 2015).

Kopin 2018 {published data only}

Granger CB, Alexander JH, McMurray JJ, Lopes RD, Hylek EM, Hanna M, et al. Apixaban versus warfarin in patients with atrial fibrillation. New England Journal of Medicine 2011;365(11):981-92. [DOI: 10.1056/NEJMoa1107039] [DOI] [PubMed] [Google Scholar]
Kopin D, Jones WS, Sherwood MW, Wojdyla DM, Wallentin L, Lewis BS, et al. Percutaneous coronary intervention and antiplatelet therapy in patients with atrial fibrillation receiving apixaban or warfarin: insights from the ARISTOTLE trial. American Heart Journal 2018;197:133-41. [DOI: 10.1016/j.ahj.2017.11.005] [DOI] [PubMed] [Google Scholar]
Kopin D, Sherwood MW, Jones WS, Wojdyla DM, Lewis BS, Verheugt F, et al. Percutaneous coronary intervention and antiplatelet therapy on apixaban or warfarin: insights from the ARISTOTLE trial. Circulation 2016;134(Suppl 1):A16695. [DOI: 10.1161/circ.134.suppl_1.16695] [Abstract 16695] [DOI] [PubMed] [Google Scholar]
Lopes RD, Alexander JH, Al-Khatib SM, Ansell J, Diaz R, Easton JD, et al. Apixaban for reduction in stroke and other thromboembolic events in atrial fibrillation (ARISTOTLE) trial: design and rationale. American Heart Journal 2010;159:331-9. [DOI] [PubMed] [Google Scholar]

PIONEER AF‐PCI 2016 {published data only}

Chi G, Korjian S, Daaboul Y, Nafee T, Mehran R, Bode C, et al. Effect of procedure and lesion characteristics on bleeding and ischemic outcomes among atrial fibrillation patients undergoing percutaneous coronary intervention: results from the PIONEER AF-PCI trial. European Heart Journal 2017;38(Supplement 1):1290. [DOI: 10.1093/eurheartj/ehx502.P6112] [Abstract P6112] [DOI] [Google Scholar]
Chi G, Yee MK, Kalayci A, Kerneis M, AlKhalfan F, Mehran R, et al. Total bleeding with rivaroxaban versus warfarin in patients with atrial fibrillation receiving antiplatelet therapy after percutaneous coronary intervention. Journal of Thrombosis and Thrombolysis 2018;46(3):346-50. [DOI: 10.1007/s11239-018-1703-5] [DOI] [PubMed] [Google Scholar]
Gibson CM, Arbetter DF, Jain P, Mehran R, Bode C, Halperin J, et al. Rivaroxaban strategies improve the number of days patients remain out of the hospital and event free: a PIONEER substudy. European Heart Journal 2017;38(Supplement 1):768. [DOI: 10.1093/eurheartj/ehx502.P3590] [Abstract P3590] [DOI] [Google Scholar]
Gibson CM, Mehran R, Bode C, Halperin J, Verheugt F, Wildgoose P, et al. An open-label, randomized, controlled, multicenter study exploring two treatment strategies of rivaroxaban and a dose-adjusted oral vitamin K antagonist treatment strategy in subjects with atrial fibrillation who undergo percutaneous coronary intervention (PIONEER AF-PCI). American Heart Journal 2015;169(4):472-8.e5. [DOI: 10.1016/j.ahj.2014.12.006] [DOI] [PubMed] [Google Scholar]
Gibson CM, Mehran R, Bode C, Halperin J, Verheugt FW, Wildgoose P, et al. An open-label, randomized, controlled, multicenter study exploring two treatment strategies of rivaroxaban and a dose-adjusted oral vitamin K antagonist treatment strategy in subjects with atrial fibrillation who undergo percutaneous coronary intervention PIONEER AF-PCI. Circulation 2016;134(25):E705. [DOI] [PubMed] [Google Scholar]
Gibson CM, Mehran R, Bode C, Halperin J, Verheugt FW, Wildgoose P, et al. Prevention of bleeding in patients with atrial fibrillation undergoing PCI. New England Journal of Medicine 2016;375(25):2423-34. [DOI: 10.1056/NEJMoa1611594] [DOI] [PubMed] [Google Scholar]
Gibson CM, Pinto DS, Chi G, Arbetter D, Yee M, Mehran R, et al. Recurrent hospitalization among patients with atrial fibrillation undergoing intracoronary stenting treated with 2 treatment strategies of rivaroxaban or a dose-adjusted oral vitamin K antagonist treatment strategy. Circulation 2017;135(4):323-33. [DOI: 10.1161/CIRCULATIONAHA.116.025783] [DOI] [PMC free article] [PubMed] [Google Scholar]
Kerneis M, Gibson CM, Chi G, Mehran R, AlKhalfan F, Talib U, et al. Effect of procedure and coronary lesion characteristics on clinical outcomes among atrial fibrillation patients undergoing percutaneous coronary intervention: insights from the PIONEER AF-PCI trial. JACC. Cardiovascular Interventions 2018;11(7):626-34. [DOI: 10.1016/j.jcin.2017.11.009] [DOI] [PubMed] [Google Scholar]
Marszalek J, Mehrsefat S, Chi G. Effect of low-dose versus very-low-dose rivaroxaban on stroke and major bleeding among atrial fibrillation patients undergoing percutaneous coronary intervention: a bivariate analysis on the net clinical benefit. Arteriosclerosis, Thrombosis, and Vascular Biology 2017;37:A299. [Google Scholar]
Yee MK, Chi G, Jain P, Mehran R, Bode C, Halperin J, et al. Rivaroxaban treatment strategies reduce not only the first bleeding event, but all (including multiple) bleeding events: a PIONEER substudy. European Heart Journal 2017;38(Supplement 1):963-4. [DOI: 10.1093/eurheartj/ehx502.P4565] [DOI] [Google Scholar]

RE‐DUAL PCI 2017 {published data only}

Cannon CP, Bhatt DL, Oldgren J, Lip GY, Ellis SG, Kimura T, et al. Dual antithrombotic therapy with dabigatran after PCI in atrial fibrillation. New England Journal of Medicine 2017;377(16):1513-24. [DOI: 10.1056/NEJMoa1708454] [DOI] [PubMed] [Google Scholar]
Cannon CP, Gropper S, Bhatt DL, Ellis SG, Kimura T, Lip GY, et al. Design and rationale of the RE-DUAL PCI trial: a prospective, randomized, phase 3b study comparing the safety and efficacy of dual antithrombotic therapy with dabigatran etexilate versus warfarin triple therapy in patients with nonvalvular atrial fibrillation who have undergone percutaneous coronary intervention with stenting. Clinical Cardiolology 2016;39(10):555-64. [DOI: 10.1002/clc.22572] [DOI] [PMC free article] [PubMed] [Google Scholar]
Oldgren J, Steg PG, Hohnloser SH, Lip GY, Ellis SG, Kimura T, et al. Subgroup analysis from the RE-DUAL PCI trial: dual antithrombotic therapy with dabigatran in patients with atrial fibrillation undergoing percutaneous coronary intervention. Circulation 2017;136(24):e457. [DOI: 10.1161/CIR.0000000000000546] [Abstract 11775] [DOI] [Google Scholar]
Steg P, Bhatt D, Ellis S, Hohnloser S, Kleine E, Nordaby M, et al. Bleeding events and their consequences on cardiovascular events, readmission and net clinical benefit. Journal of the American College of Cardiology 2018;71(11 Suppl):31695-4. [DOI: 10.1016/S0735-1097(18)31695-4] [DOI] [Google Scholar]
Ten Berg J, De Veer A, Oldgren J, Steg P, Zateyshchikov DA, Janský P, et al. Is it safe to switch OAC in patients with atrial fibrillation undergoing percutaneous coronary intervention? results from a RE-DUAL PCI subanalysis. Journal of the American College of Cardiology 2018;71(11 Suppl):A1206. [DOI: 10.1016/S0735-1097(18)31747-9] [DOI] [Google Scholar]

Sherwood 2016 {published data only}

Becker R, Berkowitz SD, Breithardt G, Califf RM, Fox K, Hacke W. Rivaroxaban-once daily, oral, direct factor Xa inhibition compared with vitamin K antagonism for prevention of stroke and Embolism Trial in Atrial Fibrillation: rationale and design of the ROCKET AF study. American Heart Journal 2010;159:340-7. [DOI] [PubMed] [Google Scholar]
Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. New England Journal of Medicine 2011;365(10):883-91. [DOI: 10.1056/NEJMoa1009638] [DOI] [PubMed] [Google Scholar]
Sherwood MW, Cyr DD, Jones WS, Becker RC, Berkowitz SD, Washam JB, et al. Use of dual antiplatelet therapy and patient outcomes in those undergoing percutaneous coronary intervention: the ROCKET AF trial. JACC. Cardiovascular Interventions 2016;9(16):1694-702. [DOI: 10.1016/j.jcin.2016.05.039] [DOI] [PMC free article] [PubMed] [Google Scholar]

References to studies excluded from this review

Ako 2019 {published data only}

Ako J, Okumura K, Nakao K, Kozuma K, Morino Y, Okazaki K, et al. Dual anti-thrombotic therapy with dabigatran after percutaneous coronary intervention in atrial fibrillation - Japanese and East-Asian subgroup analysis of the RE-DUAL PCI trial. Circulation Journal 2019;83(2):327-33. [DOI: 10.1253/circj.CJ-18-0874] [DOI] [PubMed] [Google Scholar]

Carrie 2017 {published data only}

Carrie D, Menown I, Oldroyd K, Copt S, Talwar S, Maillard L, et al. Safety and efficacy of polymer-free biolimus A9-coated versus bare-metal stents in orally anticoagulated patients: 2-year results of the LEADERS FREE oral anticoagulation substudy. JACC: Cardiovascular Interventions 2017;10(16):1633-42. [DOI: 10.1016/j.jcin.2017.05.033] [DOI] [PubMed] [Google Scholar]

Casamira 2017 {published data only}

Casamira N, Sambola A, Garcia T, Montoro JB, Veiga G, Garcia DB, et al. Use and outcomes of triple antithrombotic therapy with vitamin K antagonists versus non-vitamin K antagonists in patients with atrial fibrillation undergoing percutaneous coronary intervention. European Heart Journal 2017;38(Suppl 1):881. [DOI: 10.1093/eurheartj/ehx502.P4298] [DOI] [Google Scholar]

Cavender 2015 {published data only}

Cavender MA, Gibson CM, Braunwald E, Wiviott SD, Murphy SA, Toda KE, et al. The effect of rivaroxaban on myocardial infarction in the ATLAS ACS 2 - TIMI 51 trial. European Heart Journal 2015;4(5):468-74. [DOI: 10.1177/2048872614554109] [DOI] [PubMed] [Google Scholar]

Connolly 2018 {published data only}

Connolly SJ, Eikelboom JW, Bosch J, Dagenais G, Dyal L, Lanas F, et al. Rivaroxaban with or without aspirin in patients with stable coronary artery disease: an international, randomised, double-blind, placebo-controlled trial. Lancet 2018;391(10117):205-18. [DOI: 10.1016/S0140-6736(17)32458-3] [DOI] [PubMed] [Google Scholar]

Cornel 2012 {published data only}

Cornel J, James S, Lopes RD, Mohan P, Neely M, Amerena J, et al. Apixaban after acute coronary syndrome in patients with heart failure: insights from the APPRAISE-2 trial. European Heart Journal 2012;1:335-6. [DOI: 10.1093/eurheartj/ehs281] [DOI] [Google Scholar]

Faggioni 2016 {published data only}

Faggioni M, Chandrasekhar J, Aquino M, Vogel B, Baber U, Sharma S, et al. Real world use of antithrombotic regimens after percutaneous coronary intervention in patients with atrial fibrillation. Journal of the American College of Cardiology 2016;68(18 Suppl 1):B91. [Google Scholar]

Hess 2015 {published data only}

Hess CN, James S, Lopes RD, Wojdyla DM, Neely ML, Liaw D, et al. Apixaban plus mono versus dual antiplatelet therapy in acute coronary syndromes: insights from the APPRAISE-2 trial. Journal of the American College of Cardiology 2015;66(7):777-87. [DOI: 10.1016/j.jacc.2015.06.027] [DOI] [PubMed] [Google Scholar]

Hohnloser 2018 {published data only}

Hohnloser S, Steg P, Oldgren J, Nickenig G, Kiss RG, Ongen Z, et al. Does renal function influence the benefit of dual antithrombotic therapy with dabigatran in patients with atrial fibrillation undergoing percutaneous coronary intervention? results from RE-DUAL PCI. Journal of the American College of Cardiology 2018;18(11 Suppl 1):30855-6. [DOI: 10.1016/S0735-1097(18)30855-6] [DOI] [Google Scholar]

Hoshi 2017 {published data only}

Hoshi T, Sato A, Nogami A, Gosho M, Aonuma K. Rationale and design of the SAFE-A study: SAFety and Effectiveness trial of Apixaban use in association with dual antiplatelet therapy in patients with atrial fibrillation undergoing percutaneous coronary intervention. Journal of Cardiology 2017;69(4):648-51. [DOI: 10.1016/j.jjcc.2016.06.007] [DOI] [PubMed] [Google Scholar]

Korjian 2018 {published data only}

Korjian S, Daaboul Y, Laliberte F, Zhao Q, Mehran R, Bode C, et al. Cost implications of anticoagulation strategies after percutaneous coronary intervention among patients with atrial fibrillation (a PIONEER-AF PCI analysis). American Journal of Cardiology 2018;06:06. [DOI: 10.1016/j.amjcard.2018.10.033] [DOI] [PubMed] [Google Scholar]

Lahsaei 2016 {published data only}

Lahsaei S, Govil A, Wong J, Brewster J, Beyer A, Shaw R. Outcomes of single versus dual antiplatelet therapy with target specific oral anticoagulant agents in patients with atrial fibrillation undergoing coronary artery intervention. Journal of the American College of Cardiology 2016;1:890. [Google Scholar]

Lip 2018 {published data only}

Lip G, Mauri L, Montalescot G, Ozkor M, Vardas P, Steg PG, et al. Relation of stroke and bleeding risk profiles to efficacy and safety of dabigatran in atrial fibrillation after PCI: an ancillary analysis from the RE-DUAL PCI trial. Journal of the American College of Cardiology 2018;18(11 Suppl 1):30830-1. [DOI: 10.1016/S0735-1097%2818%2930830-1] [DOI] [Google Scholar]

Matsumura Nakano 2018 {published data only}

Matsumura-Nakano Y, Shizuta S, Komasa A, Morimoto T, Masuda H, Shiomi H, et al. An open-label randomized trial comparing oral anticoagulation with and without single antiplatelet therapy in patients with atrial fibrillation and stable coronary artery disease beyond one year after coronary stent implantation: the OAC-ALONE study. Circulation 2018;24:24. [DOI: 10.1161/CIRCULATIONAHA.118.036768] [DOI] [PubMed] [Google Scholar]

Musumeci 2014 {published data only}

Musumeci G, Rossini R. Primary angioplasty in patients with non valvular atrial fibrillation treated with NOACs. Thrombosis Research 2014;2:S73. [Google Scholar]

NCT01962545 {published data only}

NCT01962545. Optimizing antithrombotic care in patients with atriaL fibrillation and coronary stent (OAC-ALONE) study. clinicaltrials.gov/ct2/show/NCT01962545 (first posted 14 October 2013).

NCT03023020 {published data only}

NCT03023020. Management of high bleeding risk patients post bioresorbable polymer coated stent implantation with an abbreviated versus prolonged DAPT regimen (MASTER DAPT). clinicaltrials.gov/ct2/show/NCT03023020 (first posted 18 January 2017).

Pinto 2017 {published data only}

Pinto DS, Tamez H, Korjian S, Daaboul Y, Mehran R, Bode C, et al. Rivaroxaban treatment strategies reduce costs associated with rehospitalizations due to bleeding and cardiovascular events: results from the PIONEER AF-PCI trial. European Heart Journal 2017;38(Suppl 1):965. [DOI: 10.1093/eurheartj/ehx502.P4569] [DOI] [Google Scholar]

Potter 2018 {published data only}

Potter BJ, Ando G, Cimmino G, Ladeiras-Lopes R, Frikah Z, Chen XY, et al. Time trends in antithrombotic management of patients with atrial fibrillation treated with coronary stents: results from TALENT-AF (the internAtionaL stENT - Atrial Fibrillation study) multicenter registry. Clinical Cardioliology 2018;41(4):470-5. [DOI: 10.1002/clc.22898] [DOI] [PMC free article] [PubMed] [Google Scholar]

Povsic 2016 {published data only}

Povsic TJ, Roe MT, Ohman EM, Steg PG, James S, Plotnikov A, et al. A randomized trial to compare the safety of rivaroxaban vs aspirin in addition to either clopidogrel or ticagrelor in acute coronary syndrome: the design of the GEMINI-ACS-1 phase II study. American Heart Journal 2016;174:120-8. [DOI: 10.1016/j.ahj.2016.01.004] [DOI] [PubMed] [Google Scholar]

Secemsky 2016 {published data only}

Secemsky EA, Butala NM, Kartoun U, Mahmood S, Wasfy JH, Kennedy KF, et al. Use of chronic oral anticoagulation and associated outcomes among patients undergoing percutaneous coronary intervention. Journal of the American Heart Association 2016;5(10):17. [DOI] [PMC free article] [PubMed] [Google Scholar]

Sindet Pedersen 2018 {published data only}

Sindet-Pedersen C, Staerk L, Lamberts M, Gerds TA, Berger JS, Nissen Bonde A, et al. Use of oral anticoagulants in combination with antiplatelet(s) in atrial fibrillation. Heart 2018;104(11):912-20. [DOI: 10.1136/heartjnl-2017-311976] [DOI] [PubMed] [Google Scholar]

Sra 2016 {published data only}

Sra S, Tan MK, Mehta SR, Fisher HN, Dery JP, Welsh RC, et al. Ischemic and bleeding events in patients with myocardial infarction undergoing percutaneous coronary intervention who require oral anticoagulation: insights from the Canadian observational AntiPlatelet sTudy. American Heart Journal 2016;180:82-9. [DOI: 10.1016/j.ahj.2016.07.015] [DOI] [PubMed] [Google Scholar]

tenBerg 2002 {published data only}

ten Berg JM, Gerritsen WB, Haas FJ, Kelder JC, Verheugt FW, Thijs Plokker HW. Pretreatment with oral anticoagulants decreases platelet activation in patients before and after percutaneous coronary intervention. Thrombosis and Haemostasis 2002;88(6):924-30. [PubMed] [Google Scholar]

Verheugt 2014 {published data only}

Verheugt FW, Vranckx P, Lindeboom W, Kim SY, Kirsch B, Van Eickels M, Leebeek FW. Rivaroxaban in elective percutaneous coronary intervention to treat stable coronary artery disease. European Heart Journal 2014;35:992. [DOI: 10.1093/eurheartj/ehu325] [DOI] [PMC free article] [PubMed] [Google Scholar]

Vranckx 2013 {published data only}

Vranckx P, Verheugt FW, Maat MP, Ulmans VA, Regar E, Smits P, et al. A randomised study of dabigatran in elective percutaneous coronary intervention in stable coronary artery disease patients. EuroIntervention 2013;8(9):1052-60. [DOI: 10.4244/EIJV8I9A162] [DOI] [PubMed] [Google Scholar]

Xanthopoulou 2018 {published data only}

Xanthopoulou I, Dragona VM, Davlouros P, Tsioufis C, Iliodromitis E, Alexopoulos D. Contemporary antithrombotic treatment in patients with non-valvular atrial fibrillation undergoing percutaneous coronary intervention: rationale and design of the Greek AntiPlatElet Atrial Fibrillation (GRAPE-AF) registry. Cardiovascular Drugs and Therapy 2018;32(2):191-6. [DOI: 10.1007/s10557-018-6789-7] [DOI] [PubMed] [Google Scholar]

Xiaoyan 2014 {published data only}

Xiaoyan L, Ying L, Hong ZE, Ping Z. Safety and effectiveness of rivaroxaban in patients with acute coronary syndrome undergoing percutaneous coronary intervention. Journal of the American College of Cardiology 2014;64(16 Suppl 1):C150. [Google Scholar]

Yasuda 2018 {published data only}

Yasuda S, Kaikita K, Ogawa H, Akao M, Ako J, Matoba T, et al. Atrial fibrillation and ischemic events with rivaroxaban in patients with stable coronary artery disease (AFIRE): protocol for a multicenter, prospective, randomized, open-label, parallel group study. International Journal of Cardiology 2018;265:108-12. [DOI: 10.1016/j.ijcard.2018.04.131] [DOI] [PubMed] [Google Scholar]

Yoshida 2018 {published data only}

Yoshida R, Morishima I, Takagi K, Morita Y, Tsuboi H, Murohara T. Comparison between long-term clinical outcomes of vitamin K antagonist and direct oral anticoagulants in patients with atrial fibrillation undergoing percutaneous coronary intervention. Circulation Journal 2018;82(8):2016-24. [DOI: 10.1253/circj.CJ-17-1171] [DOI] [PubMed] [Google Scholar]

References to ongoing studies

Gao 2015 {published data only}

Gao F, Shen H, Wang ZJ, Yang SW, Liu XL, Zhou YJ. Rationale and design of the RT-AF study: combination of rivaroxaban and ticagrelor in patients with atrial fibrillation and coronary artery disease undergoing percutaneous coronary intervention. Contemporary Clinical Trials 2015;43:129-32. [DOI: 10.1016/j.cct.2015.05.012] [DOI] [PubMed] [Google Scholar]

NCT02789917 {published data only}

NCT02789917. Edoxaban treatment versus vitamin K antagonist in patients with atrial fibrillation undergoing percutaneous coronary intervention (ENTRUST-AF-PCI). clinicaltrials.gov/ct2/show/NCT02866175 (first posted 15 August 2016).

NCT03536611 {published data only}

NCT03536611. Dabigatran versus warfarin with NVAF who undergo PCI. clinicaltrials.gov/ct2/show/NCT03536611 (first posted 24 May 2018).

Vranckx 2018 {published data only}

NCT02866175. Edoxaban treatment versus vitamin K antagonist in patients with atrial fibrillation undergoing percutaneous coronary intervention. clinicaltrials.gov/show/NCT02866175 (first posted 15 August 2016).
Vranckx P, Lewalter T, Valgimigli M, Tijssen JG, Reimitz PE, Eckardt L, et al. Evaluation of the safety and efficacy of an edoxaban-based antithrombotic regimen in patients with atrial fibrillation following successful percutaneous coronary intervention (PCI) with stent placement: rationale and design of the ENTRUST-AF PCI trial. American Heart Journal 2018;196:105-12. [DOI: 10.1016/j.ahj.2017.10.009] [DOI] [PubMed] [Google Scholar]

Additional references

Ahrens 2012

Ahrens I, Peter K, Lip GY, Bode C. Development and clinical applications of novel oral anticoagulants. Part II. Drugs under clinical investigation. Discovery Medicine 2012;13(73):445-50. [PubMed] [Google Scholar]

Alexander 2009

Alexander JH, Becker RC, Bhatt DL, Cools F, Crea F, Dellborg M, et al. Apixaban, an oral, direct, selective factor Xa inhibitor, in combination with antiplatelet therapy after acute coronary syndrome: results of the Apixaban for Prevention of Acute Ischemic and Safety Events (APPRAISE) trial. Circulation 2009;119(22):2877-85. [DOI] [PubMed] [Google Scholar]

Alexander 2011

Alexander JH, Lopes RD, James S, Kilaru R, He Y, Mohan P, et al. Apixaban with antiplatelet therapy after acute coronary syndrome. New England Journal of Medicine 2011;365(8):699-708. [DOI] [PubMed] [Google Scholar]

Andreotti 2006

Andreotti F, Testa L, Biondi-Zoccai GG, Crea F. Aspirin plus warfarin compared to aspirin alone after acute coronary syndromes: an updated and comprehensive meta-analysis of 25,307 patients. European Heart Journal 2006;27(5):519-26. [DOI] [PubMed] [Google Scholar]

Atar 2014

Atar D, Bode C, Stuerzenbecher A, Verheugt FW. Anticoagulants for secondary prevention after acute myocardial infarction: lessons from the past decade. Fundamental Clinical Pharmacology 2014;28(4):353–63. [DOI] [PMC free article] [PubMed] [Google Scholar]

Bauer 2013

Bauer KA. Pros and cons of new oral anticoagulants. Hematology 2013;2013(1):464-70. [DOI] [PubMed] [Google Scholar]

Bayer Pharma AG 2016

Bayer Pharma AG. Xarelto® (rivaroxaban) summary of product characteristics. www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Product_Information/human/000944/WC500057108.pdf (accessed 30 March 2018).

Benjamin 2018

Benjamin EJ, Virani SS, Callaway CW, Chamberlain AM, Chang AR, Cheng S, et al. Heart disease and stroke statistics - 2018 update: a report from the American Heart Association. Circulation 2018;137(12):e67-492. [DOI] [PubMed] [Google Scholar]

Brignardello‐Petersen 2017

Brignardello-Petersen R, Bonner A, Alexander PE, Siemieniuk RA, Furukawa TA, Rochwerg B, et al. Advances in the GRADE approach to rate the certainty in estimates from a network meta-analysis. Journal of Clinical Epidemiology 2018;93:36-44. [DOI] [PubMed] [Google Scholar]

Brignardello‐Petersen 2018a

Brignardello-Petersen R, Murad MH, Walter SD, McLeod S, Carrasco-Labra A, Rochwerg B, et al. GRADE approach to rate the certainty from a network meta-analysis: avoiding spurious judgments of imprecision in sparse networks. Journal of Clinical Epidemiology 2018;18:30497-9. [DOI] [PubMed] [Google Scholar]

Brignardello‐Petersen 2018b

Brignardello-Petersen R, Bonner A, Alexander PE, Siemieniuk RA, Furukawa TA, Rochwerg B, et al. Advances in the GRADE approach to rate the certainty in estimates from a network meta-analysis. Journal of Clinical Epidemiology 2018;93:36-44. [DOI] [PubMed] [Google Scholar]

Brignardello‐Petersen 2019

Brignardello-Petersen R, Mustafa RA, Siemieniuk RA, Murad MH, Agoritsas T, Izcovich A, et al. GRADE approach to rate the certainty from a network meta-analysis: addressing incoherence. Journal of Clinical Epidemiology 2019;108:77-85. [DOI] [PubMed] [Google Scholar]

Brunetti 2018

Brunetti ND, Tarantino N, De Gennaro L, Correale M, Santoro F, Di Biase M. Direct oral anticoagulants versus standard triple therapy in atrial fibrillation and PCI: meta-analysis. Open Heart 2018;5:000785. [DOI] [PMC free article] [PubMed] [Google Scholar]

Camm 2010

Camm AJ, Kirchhof P, Lip GY, Schotten U, Savelieva I, Ernst S, et al. Guidelines for the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Europace 2010;12(10):1360-420. [DOI] [PubMed] [Google Scholar]

Chaimani 2019

Chaimani A, Caldwell DM, Li T, Higgins JP, Salanti G. Chapter 11: Undertaking network meta-analyses. In: Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for Systematic Reviews of Interventions version 6.0 (updated July 2019). Cochrane, 2019. Available from www.training.cochrane.org/handbook.

Cho 2015

Cho JR, Angiolillo DJ. Percutaneous coronary intervention and atrial fibrillation: the triple therapy dilemma. Journal of Thrombosis and Thrombolysis 2015;39(2):203-8. [DOI] [PubMed] [Google Scholar]

Connolly 2006

Connolly S, Pogue J, Hart R, Pfeffer M, Hohnloser S, Chrolavicius S, et al. Clopidogrel plus aspirin versus oral anticoagulation for atrial fibrillation in the Atrial fibrillation Clopidogrel Trial with Irbesartan for prevention of Vascular Events (ACTIVE W): a randomised controlled trial. Lancet 2006;367(9526):1903-12. [DOI] [PubMed] [Google Scholar]

Connolly 2009

Connolly SJ, Ezekowitz MD, Yusuf S, Eikelboom J, Oldgren J, Parekh A, et al. Dabigatran versus warfarin in patients with atrial fibrillation. New England Journal of Medicine 2009;361(12):1139-51. [DOI] [PubMed] [Google Scholar]

Cuker 2019

Cuker A, Burnett A, Triller D, Crowther M, Ansell J, Van Cott EM, et al. Reversal of direct oral anticoagulants: guidance from the Anticoagulation Forum. American Journal of Hematology 2019;94:697. [DOI] [PubMed] [Google Scholar]

Deeks 2017

Deeks JJ, Higgins JP, Altman DG, editor(s), on behalf of the Cochrane Statistical Methods Group. Chapter 9: Analysing data and undertaking meta-analyses. In: Higgins JP, Churchill R, Chandler J, Cumpston MS, editor(s). Cochrane Handbook for Systematic Reviews of Interventions version 5.2.0 (updated June 2017). The Cochrane Collaboration, 2017. Available at www.training.cochrane.org/handbook.

Dewilde 2013

Dewilde WJ, Oirbans T, Verheugt FW, Kelder JC, De Smet B, Herrman JP, et al. Use of clopidogrel with or without aspirin in patients taking oral anticoagulant therapy and undergoing percutaneous coronary intervention: an open-label, randomised, controlled trial. Lancet 2013;381(9872):1107-15. [DOI] [PubMed] [Google Scholar]

European Medicines Agency 2013

European Medicines Agency. Assessment report: Xarelto (Procedure No. EMEA/H/C/000944/X/00017) 2013. www.ema.europa.eu/docs/en GB/document library/EPAR - Assessment Report - Variation/human/000944/WC500144718.pdf (accessed 10 July 2013).

Fiedler 2015

Fiedler KA, Maeng M, Mehilli J, Schulz-Schüpke S, Byrne RA, Sibbing D, et al. Duration of triple therapy in patients requiring oral anticoagulation after drug-eluting stent implantation: the ISAR-TRIPLE trial. Journal of the American College Cardiology 2015;65(16):1619-29. [DOI] [PubMed] [Google Scholar]

Finegold 2013

Finegold JA, Asaria P, Francis DP. Mortality from ischaemic heart disease by country, region, and age: statistics from World Health Organization and United Nations. International Journal of Cardiology 2013;168(2):934–45. [DOI] [PMC free article] [PubMed] [Google Scholar]

Gibson 2011

Gibson CM, Mega JL, Burton P, Goto S, Verheugt F, Bode C, et al. Rationale and design of the anti-Xa therapy to lower cardiovascular events in addition to standard therapy in subjects with acute coronary syndrome-thrombolysis in myocardial infarction 51 (ATLAS-ACS 2 TIMI 51) trial: a randomized, double-blind, placebo-controlled study to evaluate the efficacy and safety of rivaroxaban in subjects with acute coronary syndrome. American Heart Journal 2011;161(5):815-21. [DOI] [PubMed] [Google Scholar]

Gibson 2016a

Gibson CM, Mehran R, Bode C, Halperin J, Verheugt FW, Wildgoose P, et al. Prevention of bleeding in patients with atrial fibrillation undergoing PCI. New England Journal of Medicine 2016;375(25):2423-34. [DOI] [PubMed] [Google Scholar]

Giugliano 2013

Giugliano RP, Ruff CT, Braunwald E, Murphy SA, Wiviott SD, Halperin JL, et al. Edoxaban versus warfarin in patients with atrial fibrillation. New England Journal of Medicine 2013;369(22):2093-104. [DOI] [PubMed] [Google Scholar]

Glund 2015

Glund S, Moschetti V, Norris S, Stangier J, Schmohl M, Ryn J, et al. A randomised study in healthy volunteers to investigate the safety, tolerability and pharmacokinetics of idarucizumab, a specific antidote to dabigatran. Thrombosis and Haemostasis 2015;113(5):943-51. [DOI] [PubMed] [Google Scholar]

GRADEpro GDT [Computer program]

McMaster University (developed by Evidence Prime) GRADEpro GDT. Version accessed 14 January 2019. Hamilton (ON): McMaster University (developed by Evidence Prime), 2015. Available at gradepro.org.

Granger 2011

Granger CB, Alexander JH, McMurray JJ, Lopes RD, Hylek EM, Hanna M, et al. Apixaban versus warfarin in patients with atrial fibrillation. New England Journal of Medicine 2011;365(11):981-92. [DOI] [PubMed] [Google Scholar]

Higgins 2017

Higgins JP, Churchill R, Chandler J, Cumpston MS, editor(s). Cochrane Handbook for Systematic Reviews of Interventions version 5.2.0 (updated June 2017). The Cochrane Collaboration, 2017. Available at www.training.cochrane.org/handbook.

Ibanez 2018

Ibanez B, James S, Agewall S, Antunes MJ, Bucciarelli-Ducci C, Bueno H, et al. 2017 ESC guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation: the Task Force for the management of acute myocardial infarction in patients presenting with ST-segment elevation of the European Society of Cardiology (ESC). European Heart Journal 2018;39(2):119-77. [DOI] [PubMed] [Google Scholar]

Jackson 2012

Jackson D, White IR and Riley RD. Quantifying the impact of between-study heterogeneity in multivariate meta-analyses. Statistics in Medicine 2012;31:3805–20. [DOI] [PMC free article] [PubMed] [Google Scholar]

Jansen 2008

Jansen JP, Crawford B, Bergman G, Stam W. Bayesian meta-analysis of multiple treatment comparisons: an introduction to mixed treatment comparisons. Value in Health 2008;11(5):956-64. [DOI] [PubMed] [Google Scholar]

January 2014

January CT, Wann LS, Alpert JS, Calkins H, Cigarroa JE, Cleveland JC, et al. 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines and the Heart Rhythm Society. Circulation 2014;130(23):2071-104. [DOI] [PubMed] [Google Scholar]

Kearon 2016

Kearon C, Akl EA, Ornelas J, Blaivas A, Jimenez D, Bounameaux H, et al. Antithrombotic therapy for VTE disease: chest guideline and expert panel report. Chest 2016;149(2):315-52. [DOI] [PubMed] [Google Scholar]

Khera 2016

Khera S, Kolte D, Bhatt DL. Chapter 16: Percutaneous coronary intervention. In: Aronow WS, McClung JA, editors(s). Translational Research in Coronary Artery Disease. Boston: Academic Press, 2016:179-94. [Google Scholar]

Kirchhof 2016

Kirchhof P, Benussi S, Kotecha D, Ahlsson A, Atar D, Casadei B, et al. 2016 ESC guidelines for the management of atrial fibrillation developed in collaboration with EACTS. European Journal of Cardio-thoracic Surgery 2016;50(5):e1-88. [DOI] [PubMed] [Google Scholar]

Knuuti 2019

Knuuti J, Wijns W, Saraste A, Capodanno D, Barbato E, Funck-Brentano C, et al. ESC Scientific Document Group, 2019 ESC Guidelines for the diagnosis and management of chronic coronary syndromes: the Task Force for the diagnosis and management of chronic coronary syndromes of the European Society of Cardiology (ESC). European Heart Journal 2019;10:1093. [Google Scholar]

Kohli 2014

Kohli P, Waters DD. Looking for coronary disease in patients with atrial fibrillation. Canadian Journal of Cardiology 2014;30(8):861-3. [DOI] [PubMed] [Google Scholar]

Konstantinides 2014

Konstantinides SV, Torbicki A, Agnelli G, Danchin N, Fitzmaurice D, Galie N, et al. 2014 ESC guidelines on the diagnosis and management of acute pulmonary embolism. European Heart Journal 2014;35(43):3033-69. [DOI] [PubMed] [Google Scholar]

Krahn 2013

Krahn U, Binder H, König J. A graphical tool for locating inconsistency in network meta-analyses. BMC Medical Research Methodology 2013;13:35. [DOI] [PMC free article] [PubMed] [Google Scholar]

Lamberts 2013

Lamberts M, Gislason GH, Olesen JB, Kristensen SL, Schjerning Olsen AM, Mikkelsen A, et al. Oral anticoagulation and antiplatelets in atrial fibrillation patients after myocardial infarction and coronary intervention. Journal of the American College of Cardiology 2013;62(11):981-9. [DOI] [PubMed] [Google Scholar]

Lee 2016

Lee LH. DOACs advances and limitations in real world. Thrombosis Journal 2016;14(Suppl 1):17. [DOI] [PMC free article] [PubMed] [Google Scholar]

Lefebvre 2011

Lefebvre C, Manheimer E, Glanville J. Chapter 6: Searching for studies. In: Higgins JP, Green S, editor(s). Cochrane Handbook for Systematic Reviews of Interventions. Version 5.1.0 (updated March 2011). The Cochrane Collaboration, 2011. Available from handbook.cochrane.org.

Leon 1998

Leon MB, Baim DS, Popma JJ, Gordon PC, Cutlip DE, Ho KK, et al. A clinical trial comparing three antithrombotic-drug regimens after coronary-artery stenting. New England Journal of Medicine 1998;339(23):1665-71. [DOI] [PubMed] [Google Scholar]

Levine 2016

Levine GN, Bates ER, Bittl JA, Brindis RG, Fihn SD, Fleisher LA, et al. ACC/AHA guideline focused update on duration of dual antiplatelet therapy in patients with coronary artery disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Journal of the American College of Cardiology 2016;68:1082–115. [DOI] [PubMed] [Google Scholar]

Liberati 2009

Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Medicine 2009;6(7):e1000100. [DOI] [PMC free article] [PubMed] [Google Scholar]

Lip 2014

Lip GY, Windecker S, Huber K, Kirchhof P, Marin F, Ten Berg JM, et al. Management of antithrombotic therapy in atrial fibrillation patients presenting with acute coronary syndrome and/or undergoing percutaneous coronary or valve interventions: a joint consensus document of the European Society of Cardiology Working Group on Thrombosis, European Heart Rhythm Association (EHRA), European Association of Percutaneous Cardiovascular Interventions (EAPCI) and European Association of Acute Cardiac Care (ACCA) endorsed by the Heart Rhythm Society (HRS) and Asia-Pacific Heart Rhythm Society (APHRS). European Heart Journal 2014;35(45):3155-79. [DOI] [PubMed]

Lopes 2019a

Lopes RD, Hong H, Harskamp RE. Safety and efficacy of antithrombotic strategies in patients with atrial fibrillation undergoing percutaneous coronary intervention: a network meta-analysis of randomized controlled trials. JAMA Cardiology 2019;10:1001. [DOI] [PMC free article] [PubMed] [Google Scholar]

Lu 2004

Lu G, Ades AE. Combination of direct and indirect evidence in mixed treatment comparisons. Statistics in Medicine 2004;23(20):3105-24. [DOI] [PubMed] [Google Scholar]

Mackman 2008

Mackman N. Triggers, targets and treatments for thrombosis. Nature 2008;451(7181):914-8. [DOI] [PMC free article] [PubMed] [Google Scholar]

Mauri 2014

Mauri L, Kereiakes DJ, Yeh RW, Driscoll-Shempp P, Cutlip DE, Steg G, et al. Twelve or 30 months of dual antiplatelet therapy after drug-eluting stents. New England Journal of Medicine 2014;371(23):2155–66. [DOI] [PMC free article] [PubMed]

McDermott 2017

McDermott KW, Freeman WJ, Elixhauser A. Overview of operating room procedures during inpatient stays in U.S. hospitals, 2014. Rockville (MD): Agency for Healthcare Research and Quality (AHRQ), Healthcare Cost and Utilization Project (HCUP); 2017 December. Statistical Brief #233. [PubMed]

McGrath 1999

McGrath PD, Malenka DJ, Wennberg DE, Shubrooks SJ, Bradley WA, Robb JF, et al. Changing outcomes in percutaneous coronary interventions: a study of 34,752 procedures in Northern New England, 1990 to 1997. Journal of the American College of Cardiology 1999;34(3):674-80. [DOI] [PubMed] [Google Scholar]

Mega 2012

Mega JL, Braunwald E, Wiviott SD, Bassand JP, Bhatt D, Bode C, et al. Rivaroxaban in patients with a recent acute coronary syndrome. New England Journal of Medicine 2012;366(1):9-19. [DOI] [PubMed] [Google Scholar]

Mehran 2011

Mehran R, Rao SV, Bhatt DL, Gibson CM, Caixeta A, Eikelboom J, et al. Standardized bleeding definitions for cardiovascular clinical trials: a consensus report from the Bleeding Academic Research Consortium. Circulation 2011;123:2736-47. [DOI] [PubMed] [Google Scholar]

Mekaj 2015

Mekaj YH, Mekaj AY, Duci SB, Miftari EI. New oral anticoagulants: their advantages and disadvantages compared with vitamin K antagonists in the prevention and treatment of patients with thromboembolic events. Therapeutics and Clinical Risk Management 2015;11:967-77. [DOI] [PMC free article] [PubMed] [Google Scholar]

Merlini 1994

Merlini PA, Bauer KA, Oltrona L, Ardissino D, Cattaneo M, Belli C, et al. Persistent activation of coagulation mechanism in unstable angina and myocardial infarction. Circulation 1994;90(1):61-8. [DOI] [PubMed] [Google Scholar]

Miller 2012

Miller CS, Grandi SM, Shimony A, Filion KB, Eisenberg MJ. Meta-analysis of efficacy and safety of new oral anticoagulants (dabigatran, rivaroxaban, apixaban) versus warfarin in patients with atrial fibrillation. American Journal of Cardiology 2012;110(3):453-60. [DOI] [PubMed] [Google Scholar]

Mills 2013

Mills EJ, Thorlund K, Ioannidis JP. Demystifying trial networks and network meta-analysis. BMJ 2013;346:f2914. [DOI] [PubMed] [Google Scholar]

Moon 2017

Moon JY, Nagaraju D, Franchi F, Rollini F, Angiolillo DJ. The role of oral anticoagulant therapy in patients with acute coronary syndrome. Therapeutic Advances in Hematology 2017;8(12):353-66. [DOI] [PMC free article] [PubMed] [Google Scholar]

Neelankavil 2015

Neelankavil J, Rau CD, Wang Y. The genetic basis of coronary artery diseases and atrial fibrillation: a search for disease mechanisms and therapeutic targets. Journal of Cardiothoracic and Vascular Anesthesia 2015;29(5):1328-32. [DOI] [PMC free article] [PubMed] [Google Scholar]

Oldgren 2011

Oldgren J, Budaj A, Granger CB, Khder Y, Roberts J, Siegbahn A, et al. Dabigatran vs. placebo in patients with acute coronary syndromes on dual antiplatelet therapy: a randomized, double-blind, phase II trial. European Heart Journal 2011;32(22):2781-9. [DOI] [PubMed] [Google Scholar]

Paikin 2010

Paikin JS, Wright DS, Crowther MA, Mehta SR, Eikelboom JW. Triple antithrombotic therapy in patients with atrial fibrillation and coronary artery stents. Circulation 2010;121(18):2067-70. [DOI] [PubMed] [Google Scholar]

Patel 2011

Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. New England Journal of Medicine 2011;365(10):883-91. [DOI] [PubMed] [Google Scholar]

Pérez‐Gómez 2004

Pérez-Gómez F, Alegría E, Berjón J, Iriarte JA, Zumalde J, Salvador A, et al. Comparative effects of antiplatelet, anticoagulant, or combined therapy in patients with valvular and nonvalvular atrial fibrillation: a randomized multicenter study. Journal of the American College of Cardiology 2004;44(8):1557-66. [DOI] [PubMed] [Google Scholar]

Peterson 2000

Peterson ED, Lansky AJ, Anstrom KJ, Muhlbaier LH, Popma JJ, Satler LF, et al. Evolving trends in interventional device use and outcomes: results from the National Cardiovascular Network database. American Heart Journal 2000;139(2):198-207. [DOI] [PubMed] [Google Scholar]

Pollack 2015

Pollack CV, Reilly PA, Eikelboom J, Glund S, Verhamme P, Bernstein RA, et al. Idarucizumab for dabigatran reversal. New England Journal of Medicine 2015;373(6):511-20. [DOI] [PubMed] [Google Scholar]

Puhan 2014

Puhan MA, Schünemann HJ, Murad MH, Li T, Brignardello-Petersen R, Singh JA, et al. Approach for rating the quality of treatment effect estimates from network meta-analysis. BMJ 2014;349:g5630. [DOI] [PubMed] [Google Scholar]

RevMan 2014 [Computer program]

Nordic Cochrane Centre, The Cochrane Collaboration Review Manager 5 (RevMan 5). Version 5.3. Copenhagen: Nordic Cochrane Centre, The Cochrane Collaboration, 2014.

Roffi 2016

Roffi M, Patrono C, Collet JP, Mueller C, Valgimigli M, Andreotti F, et al. 2015 ESC guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: Task Force for the Management of Acute Coronary Syndromes in Patients Presenting Without Persistent ST-Segment Elevation of the European Society of Cardiology (ESC). European Heart Journal 2016;37(3):267-315. [DOI] [PubMed] [Google Scholar]

Rothberg 2005

Rothberg MB, Celestin C, Fiore LD, Lawler E, Cook JR. Warfarin plus aspirin after myocardial infarction or the acute coronary syndrome: meta-analysis with estimates of risk and benefit. Annals of Internal Medicine 2005;143(4):241-50. [DOI] [PubMed] [Google Scholar]

Rubboli 2007

Rubboli A, Colletta M, Herzfeld J, Sangiorgio P, Di Pasquale G. Periprocedural and medium-term antithrombotic strategies in patients with an indication for long-term anticoagulation undergoing coronary angiography and intervention. Coronary Artery Disease 2007;18(3):193-9. [DOI] [PubMed] [Google Scholar]

Rücker 2015

Rücker G, Schwarzer G. Ranking treatments in frequentist network meta-analysis works without resampling methods. BMC Medical Research Methodology 2015;15:58. [DOI] [PMC free article] [PubMed]

Rücker 2017

Rücker G, Schwarzer G, Krahn U, König J. Network meta-analysis using frequentist methods - package netmeta. cran.r–project.org/web/packages/netmeta/ (accessed prior to 12 January 2019).

Ruff 2014

Ruff CT, Giugliano RP, Braunwald E, Hoffman EB, Deenadayalu N, Ezekowitz MD, 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(9921):955-62. [DOI] [PubMed] [Google Scholar]

Schünemann 2017

Schünemann HJ, Oxman AD, Vist GE, Higgins JP, Deeks JJ, Glasziou P, et al. Chapter 12: Interpreting results and drawing conclusions. In: Higgins JP, Churchill R, Chandler J, Cumpston MS, editor(s). Cochrane Handbook for Systematic Reviews of Interventions version 5.2.0 (updated June 2017). The Cochrane Collaboration, 2017. Available at www.training.cochrane.org/handbook.

Steg 2011

Steg GP, Mehta SR, Jukema JW, Lip GY, Gibson CM, Kovar F, et al. RUBY-1: a randomized, double-blind, placebo-controlled trial of the safety and tolerability of the novel oral factor Xa inhibitor darexaban (YM150) following acute coronary syndrome. European Heart Journal 2011;32(20):2541-54. [DOI] [PMC free article] [PubMed] [Google Scholar]

Valgimigli 2017

Valgimigli M, Bueno H, Byrne RA, Collet JP, Costa F, Jeppsson A, et al. 2017 ESC focused update on dual antiplatelet therapy in coronary artery disease developed in collaboration with EACTS: the Task Force for dual antiplatelet therapy in coronary artery disease of the European Society of Cardiology (ESC) and of the European Association for Cardio-Thoracic Surgery (EACTS). European Heart Journal 2017;39(3):213-60. [DOI] [PubMed] [Google Scholar]

Wallentin 2003

Wallentin L, Wilcox RG, Weaver WD, Emanuelsson H, Goodvin A, Nystroem P, et al. Oral ximelagatran for secondary prophylaxis after myocardial infarction: the ESTEEM randomised controlled trial. Lancet 2003;362(9386):789-97. [DOI] [PubMed] [Google Scholar]

Wallentin 2009

Wallentin L, Becker RC, Budaj A, Cannon CP, Emanuelsson H, Held C, et al. Ticagrelor versus clopidogrel in patients with acute coronary syndromes. New England Journal of Medicine 2009;361:1045–57. [DOI] [PubMed] [Google Scholar]

Wang 2008

Wang TY, Robinson LA, Ou FS, Roe MT, Ohman EM, Gibler WB, et al. Discharge antithrombotic strategies among patients with acute coronary syndrome previously on warfarin anticoagulation: physician practice in the CRUSADE registry. American Heart Journal 2008;155(2):361-8. [DOI] [PubMed] [Google Scholar]

WHO 2015

World Health Organization. Cardiovascular diseases. www.who.int/cardiovascular_diseases/en/ (accessed 30 March 2018).

Windecker 2014

Windecker S, Kolh P, Alfonso F, Collet JP, Cremer J, Falk V, et al. 2014 ESC/EACTS guidelines on myocardial revascularization: the Task Force on Myocardial Revascularization of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS). Developed with the special contribution of the European Association of Percutaneous Cardiovascular Interventions (EAPCI). European Heart Journal 2014;35(37):2541-619. [DOI] [PubMed] [Google Scholar]

Wiviott 2007

Wiviott SD, Braunwald E, McCabe CH, Montalescot G, Ruzyllo W, Gottlieb S, et al. Prasugrel versus clopidogrel in patients with acute coronary syndromes. New England Journal of Medicine 2007;357:2001-15. [DOI] [PubMed] [Google Scholar]

Yusuf 2001

Yusuf S, Zhao F, Mehta SR, Chrolavicius S, Tognoni G, Fox KK. Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST-segment elevation. New England Journal of Medicine 2001;345:494–502. [DOI] [PubMed] [Google Scholar]

Zirlik 2017

Zirlik A, Bode C. Vitamin K antagonists: relative strengths and weaknesses vs. direct oral anticoagulants for stroke prevention in patients with atrial fibrillation. Journal of Thrombosis and Thrombolysis 2017;43(3):365-79. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0001] Lopes RD, Heizer G, Aronson R, Vora AN, Massaro T, Mehran R, et al. Antithrombotic therapy after acute coronary syndrome or PCI in atrial fibrillation. New England Journal of Medicine 2019;380(16):1509-24. [DOI: 10.1056/NEJMoa1817083] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0002] Lopes RD, Vora AN, Liaw D, Granger CB, Darius H, Goodman SG, et al. An open-Label, 2 x 2 factorial, randomized controlled trial to evaluate the safety of apixaban vs. vitamin K antagonist and aspirin vs. placebo in patients with atrial fibrillation and acute coronary syndrome and/or percutaneous coronary intervention: rationale and design of the AUGUSTUS trial. American Heart Journal 2018;200:17-23. [DOI: 10.1016/j.ahj.2018.03.001] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0003] NCT02415400. A study of apixaban in patients with atrial fibrillation, not caused by a heart valve problem, who are at risk for thrombosis (blood clots) due to having had a recent coronary event, such as a heart attack or a procedure to open the vessels of the heart. clinicaltrials.gov/show/nct02415400 (first posted 14 April 2015).

[CD013252-bib-0004] Granger CB, Alexander JH, McMurray JJ, Lopes RD, Hylek EM, Hanna M, et al. Apixaban versus warfarin in patients with atrial fibrillation. New England Journal of Medicine 2011;365(11):981-92. [DOI: 10.1056/NEJMoa1107039] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0005] Kopin D, Jones WS, Sherwood MW, Wojdyla DM, Wallentin L, Lewis BS, et al. Percutaneous coronary intervention and antiplatelet therapy in patients with atrial fibrillation receiving apixaban or warfarin: insights from the ARISTOTLE trial. American Heart Journal 2018;197:133-41. [DOI: 10.1016/j.ahj.2017.11.005] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0006] Kopin D, Sherwood MW, Jones WS, Wojdyla DM, Lewis BS, Verheugt F, et al. Percutaneous coronary intervention and antiplatelet therapy on apixaban or warfarin: insights from the ARISTOTLE trial. Circulation 2016;134(Suppl 1):A16695. [DOI: 10.1161/circ.134.suppl_1.16695] [Abstract 16695] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0007] Lopes RD, Alexander JH, Al-Khatib SM, Ansell J, Diaz R, Easton JD, et al. Apixaban for reduction in stroke and other thromboembolic events in atrial fibrillation (ARISTOTLE) trial: design and rationale. American Heart Journal 2010;159:331-9. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0008] Chi G, Korjian S, Daaboul Y, Nafee T, Mehran R, Bode C, et al. Effect of procedure and lesion characteristics on bleeding and ischemic outcomes among atrial fibrillation patients undergoing percutaneous coronary intervention: results from the PIONEER AF-PCI trial. European Heart Journal 2017;38(Supplement 1):1290. [DOI: 10.1093/eurheartj/ehx502.P6112] [Abstract P6112] [DOI] [Google Scholar]

[CD013252-bib-0009] Chi G, Yee MK, Kalayci A, Kerneis M, AlKhalfan F, Mehran R, et al. Total bleeding with rivaroxaban versus warfarin in patients with atrial fibrillation receiving antiplatelet therapy after percutaneous coronary intervention. Journal of Thrombosis and Thrombolysis 2018;46(3):346-50. [DOI: 10.1007/s11239-018-1703-5] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0010] Gibson CM, Arbetter DF, Jain P, Mehran R, Bode C, Halperin J, et al. Rivaroxaban strategies improve the number of days patients remain out of the hospital and event free: a PIONEER substudy. European Heart Journal 2017;38(Supplement 1):768. [DOI: 10.1093/eurheartj/ehx502.P3590] [Abstract P3590] [DOI] [Google Scholar]

[CD013252-bib-0011] Gibson CM, Mehran R, Bode C, Halperin J, Verheugt F, Wildgoose P, et al. An open-label, randomized, controlled, multicenter study exploring two treatment strategies of rivaroxaban and a dose-adjusted oral vitamin K antagonist treatment strategy in subjects with atrial fibrillation who undergo percutaneous coronary intervention (PIONEER AF-PCI). American Heart Journal 2015;169(4):472-8.e5. [DOI: 10.1016/j.ahj.2014.12.006] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0012] Gibson CM, Mehran R, Bode C, Halperin J, Verheugt FW, Wildgoose P, et al. An open-label, randomized, controlled, multicenter study exploring two treatment strategies of rivaroxaban and a dose-adjusted oral vitamin K antagonist treatment strategy in subjects with atrial fibrillation who undergo percutaneous coronary intervention PIONEER AF-PCI. Circulation 2016;134(25):E705. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0013] Gibson CM, Mehran R, Bode C, Halperin J, Verheugt FW, Wildgoose P, et al. Prevention of bleeding in patients with atrial fibrillation undergoing PCI. New England Journal of Medicine 2016;375(25):2423-34. [DOI: 10.1056/NEJMoa1611594] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0014] Gibson CM, Pinto DS, Chi G, Arbetter D, Yee M, Mehran R, et al. Recurrent hospitalization among patients with atrial fibrillation undergoing intracoronary stenting treated with 2 treatment strategies of rivaroxaban or a dose-adjusted oral vitamin K antagonist treatment strategy. Circulation 2017;135(4):323-33. [DOI: 10.1161/CIRCULATIONAHA.116.025783] [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0015] Kerneis M, Gibson CM, Chi G, Mehran R, AlKhalfan F, Talib U, et al. Effect of procedure and coronary lesion characteristics on clinical outcomes among atrial fibrillation patients undergoing percutaneous coronary intervention: insights from the PIONEER AF-PCI trial. JACC. Cardiovascular Interventions 2018;11(7):626-34. [DOI: 10.1016/j.jcin.2017.11.009] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0016] Marszalek J, Mehrsefat S, Chi G. Effect of low-dose versus very-low-dose rivaroxaban on stroke and major bleeding among atrial fibrillation patients undergoing percutaneous coronary intervention: a bivariate analysis on the net clinical benefit. Arteriosclerosis, Thrombosis, and Vascular Biology 2017;37:A299. [Google Scholar]

[CD013252-bib-0017] Yee MK, Chi G, Jain P, Mehran R, Bode C, Halperin J, et al. Rivaroxaban treatment strategies reduce not only the first bleeding event, but all (including multiple) bleeding events: a PIONEER substudy. European Heart Journal 2017;38(Supplement 1):963-4. [DOI: 10.1093/eurheartj/ehx502.P4565] [DOI] [Google Scholar]

[CD013252-bib-0018] Cannon CP, Bhatt DL, Oldgren J, Lip GY, Ellis SG, Kimura T, et al. Dual antithrombotic therapy with dabigatran after PCI in atrial fibrillation. New England Journal of Medicine 2017;377(16):1513-24. [DOI: 10.1056/NEJMoa1708454] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0019] Cannon CP, Gropper S, Bhatt DL, Ellis SG, Kimura T, Lip GY, et al. Design and rationale of the RE-DUAL PCI trial: a prospective, randomized, phase 3b study comparing the safety and efficacy of dual antithrombotic therapy with dabigatran etexilate versus warfarin triple therapy in patients with nonvalvular atrial fibrillation who have undergone percutaneous coronary intervention with stenting. Clinical Cardiolology 2016;39(10):555-64. [DOI: 10.1002/clc.22572] [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0020] Oldgren J, Steg PG, Hohnloser SH, Lip GY, Ellis SG, Kimura T, et al. Subgroup analysis from the RE-DUAL PCI trial: dual antithrombotic therapy with dabigatran in patients with atrial fibrillation undergoing percutaneous coronary intervention. Circulation 2017;136(24):e457. [DOI: 10.1161/CIR.0000000000000546] [Abstract 11775] [DOI] [Google Scholar]

[CD013252-bib-0021] Steg P, Bhatt D, Ellis S, Hohnloser S, Kleine E, Nordaby M, et al. Bleeding events and their consequences on cardiovascular events, readmission and net clinical benefit. Journal of the American College of Cardiology 2018;71(11 Suppl):31695-4. [DOI: 10.1016/S0735-1097(18)31695-4] [DOI] [Google Scholar]

[CD013252-bib-0022] Ten Berg J, De Veer A, Oldgren J, Steg P, Zateyshchikov DA, Janský P, et al. Is it safe to switch OAC in patients with atrial fibrillation undergoing percutaneous coronary intervention? results from a RE-DUAL PCI subanalysis. Journal of the American College of Cardiology 2018;71(11 Suppl):A1206. [DOI: 10.1016/S0735-1097(18)31747-9] [DOI] [Google Scholar]

[CD013252-bib-0023] Becker R, Berkowitz SD, Breithardt G, Califf RM, Fox K, Hacke W. Rivaroxaban-once daily, oral, direct factor Xa inhibition compared with vitamin K antagonism for prevention of stroke and Embolism Trial in Atrial Fibrillation: rationale and design of the ROCKET AF study. American Heart Journal 2010;159:340-7. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0024] Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. New England Journal of Medicine 2011;365(10):883-91. [DOI: 10.1056/NEJMoa1009638] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0025] Sherwood MW, Cyr DD, Jones WS, Becker RC, Berkowitz SD, Washam JB, et al. Use of dual antiplatelet therapy and patient outcomes in those undergoing percutaneous coronary intervention: the ROCKET AF trial. JACC. Cardiovascular Interventions 2016;9(16):1694-702. [DOI: 10.1016/j.jcin.2016.05.039] [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0026] Ako J, Okumura K, Nakao K, Kozuma K, Morino Y, Okazaki K, et al. Dual anti-thrombotic therapy with dabigatran after percutaneous coronary intervention in atrial fibrillation - Japanese and East-Asian subgroup analysis of the RE-DUAL PCI trial. Circulation Journal 2019;83(2):327-33. [DOI: 10.1253/circj.CJ-18-0874] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0027] Carrie D, Menown I, Oldroyd K, Copt S, Talwar S, Maillard L, et al. Safety and efficacy of polymer-free biolimus A9-coated versus bare-metal stents in orally anticoagulated patients: 2-year results of the LEADERS FREE oral anticoagulation substudy. JACC: Cardiovascular Interventions 2017;10(16):1633-42. [DOI: 10.1016/j.jcin.2017.05.033] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0028] Casamira N, Sambola A, Garcia T, Montoro JB, Veiga G, Garcia DB, et al. Use and outcomes of triple antithrombotic therapy with vitamin K antagonists versus non-vitamin K antagonists in patients with atrial fibrillation undergoing percutaneous coronary intervention. European Heart Journal 2017;38(Suppl 1):881. [DOI: 10.1093/eurheartj/ehx502.P4298] [DOI] [Google Scholar]

[CD013252-bib-0029] Cavender MA, Gibson CM, Braunwald E, Wiviott SD, Murphy SA, Toda KE, et al. The effect of rivaroxaban on myocardial infarction in the ATLAS ACS 2 - TIMI 51 trial. European Heart Journal 2015;4(5):468-74. [DOI: 10.1177/2048872614554109] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0030] Connolly SJ, Eikelboom JW, Bosch J, Dagenais G, Dyal L, Lanas F, et al. Rivaroxaban with or without aspirin in patients with stable coronary artery disease: an international, randomised, double-blind, placebo-controlled trial. Lancet 2018;391(10117):205-18. [DOI: 10.1016/S0140-6736(17)32458-3] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0031] Cornel J, James S, Lopes RD, Mohan P, Neely M, Amerena J, et al. Apixaban after acute coronary syndrome in patients with heart failure: insights from the APPRAISE-2 trial. European Heart Journal 2012;1:335-6. [DOI: 10.1093/eurheartj/ehs281] [DOI] [Google Scholar]

[CD013252-bib-0032] Faggioni M, Chandrasekhar J, Aquino M, Vogel B, Baber U, Sharma S, et al. Real world use of antithrombotic regimens after percutaneous coronary intervention in patients with atrial fibrillation. Journal of the American College of Cardiology 2016;68(18 Suppl 1):B91. [Google Scholar]

[CD013252-bib-0033] Hess CN, James S, Lopes RD, Wojdyla DM, Neely ML, Liaw D, et al. Apixaban plus mono versus dual antiplatelet therapy in acute coronary syndromes: insights from the APPRAISE-2 trial. Journal of the American College of Cardiology 2015;66(7):777-87. [DOI: 10.1016/j.jacc.2015.06.027] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0034] Hohnloser S, Steg P, Oldgren J, Nickenig G, Kiss RG, Ongen Z, et al. Does renal function influence the benefit of dual antithrombotic therapy with dabigatran in patients with atrial fibrillation undergoing percutaneous coronary intervention? results from RE-DUAL PCI. Journal of the American College of Cardiology 2018;18(11 Suppl 1):30855-6. [DOI: 10.1016/S0735-1097(18)30855-6] [DOI] [Google Scholar]

[CD013252-bib-0035] Hoshi T, Sato A, Nogami A, Gosho M, Aonuma K. Rationale and design of the SAFE-A study: SAFety and Effectiveness trial of Apixaban use in association with dual antiplatelet therapy in patients with atrial fibrillation undergoing percutaneous coronary intervention. Journal of Cardiology 2017;69(4):648-51. [DOI: 10.1016/j.jjcc.2016.06.007] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0036] Korjian S, Daaboul Y, Laliberte F, Zhao Q, Mehran R, Bode C, et al. Cost implications of anticoagulation strategies after percutaneous coronary intervention among patients with atrial fibrillation (a PIONEER-AF PCI analysis). American Journal of Cardiology 2018;06:06. [DOI: 10.1016/j.amjcard.2018.10.033] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0037] Lahsaei S, Govil A, Wong J, Brewster J, Beyer A, Shaw R. Outcomes of single versus dual antiplatelet therapy with target specific oral anticoagulant agents in patients with atrial fibrillation undergoing coronary artery intervention. Journal of the American College of Cardiology 2016;1:890. [Google Scholar]

[CD013252-bib-0038] Lip G, Mauri L, Montalescot G, Ozkor M, Vardas P, Steg PG, et al. Relation of stroke and bleeding risk profiles to efficacy and safety of dabigatran in atrial fibrillation after PCI: an ancillary analysis from the RE-DUAL PCI trial. Journal of the American College of Cardiology 2018;18(11 Suppl 1):30830-1. [DOI: 10.1016/S0735-1097%2818%2930830-1] [DOI] [Google Scholar]

[CD013252-bib-0039] Matsumura-Nakano Y, Shizuta S, Komasa A, Morimoto T, Masuda H, Shiomi H, et al. An open-label randomized trial comparing oral anticoagulation with and without single antiplatelet therapy in patients with atrial fibrillation and stable coronary artery disease beyond one year after coronary stent implantation: the OAC-ALONE study. Circulation 2018;24:24. [DOI: 10.1161/CIRCULATIONAHA.118.036768] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0040] Musumeci G, Rossini R. Primary angioplasty in patients with non valvular atrial fibrillation treated with NOACs. Thrombosis Research 2014;2:S73. [Google Scholar]

[CD013252-bib-0041] NCT01962545. Optimizing antithrombotic care in patients with atriaL fibrillation and coronary stent (OAC-ALONE) study. clinicaltrials.gov/ct2/show/NCT01962545 (first posted 14 October 2013).

[CD013252-bib-0042] NCT03023020. Management of high bleeding risk patients post bioresorbable polymer coated stent implantation with an abbreviated versus prolonged DAPT regimen (MASTER DAPT). clinicaltrials.gov/ct2/show/NCT03023020 (first posted 18 January 2017).

[CD013252-bib-0043] Pinto DS, Tamez H, Korjian S, Daaboul Y, Mehran R, Bode C, et al. Rivaroxaban treatment strategies reduce costs associated with rehospitalizations due to bleeding and cardiovascular events: results from the PIONEER AF-PCI trial. European Heart Journal 2017;38(Suppl 1):965. [DOI: 10.1093/eurheartj/ehx502.P4569] [DOI] [Google Scholar]

[CD013252-bib-0044] Potter BJ, Ando G, Cimmino G, Ladeiras-Lopes R, Frikah Z, Chen XY, et al. Time trends in antithrombotic management of patients with atrial fibrillation treated with coronary stents: results from TALENT-AF (the internAtionaL stENT - Atrial Fibrillation study) multicenter registry. Clinical Cardioliology 2018;41(4):470-5. [DOI: 10.1002/clc.22898] [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0045] Povsic TJ, Roe MT, Ohman EM, Steg PG, James S, Plotnikov A, et al. A randomized trial to compare the safety of rivaroxaban vs aspirin in addition to either clopidogrel or ticagrelor in acute coronary syndrome: the design of the GEMINI-ACS-1 phase II study. American Heart Journal 2016;174:120-8. [DOI: 10.1016/j.ahj.2016.01.004] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0046] Secemsky EA, Butala NM, Kartoun U, Mahmood S, Wasfy JH, Kennedy KF, et al. Use of chronic oral anticoagulation and associated outcomes among patients undergoing percutaneous coronary intervention. Journal of the American Heart Association 2016;5(10):17. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0047] Sindet-Pedersen C, Staerk L, Lamberts M, Gerds TA, Berger JS, Nissen Bonde A, et al. Use of oral anticoagulants in combination with antiplatelet(s) in atrial fibrillation. Heart 2018;104(11):912-20. [DOI: 10.1136/heartjnl-2017-311976] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0048] Sra S, Tan MK, Mehta SR, Fisher HN, Dery JP, Welsh RC, et al. Ischemic and bleeding events in patients with myocardial infarction undergoing percutaneous coronary intervention who require oral anticoagulation: insights from the Canadian observational AntiPlatelet sTudy. American Heart Journal 2016;180:82-9. [DOI: 10.1016/j.ahj.2016.07.015] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0049] ten Berg JM, Gerritsen WB, Haas FJ, Kelder JC, Verheugt FW, Thijs Plokker HW. Pretreatment with oral anticoagulants decreases platelet activation in patients before and after percutaneous coronary intervention. Thrombosis and Haemostasis 2002;88(6):924-30. [PubMed] [Google Scholar]

[CD013252-bib-0050] Verheugt FW, Vranckx P, Lindeboom W, Kim SY, Kirsch B, Van Eickels M, Leebeek FW. Rivaroxaban in elective percutaneous coronary intervention to treat stable coronary artery disease. European Heart Journal 2014;35:992. [DOI: 10.1093/eurheartj/ehu325] [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0051] Vranckx P, Verheugt FW, Maat MP, Ulmans VA, Regar E, Smits P, et al. A randomised study of dabigatran in elective percutaneous coronary intervention in stable coronary artery disease patients. EuroIntervention 2013;8(9):1052-60. [DOI: 10.4244/EIJV8I9A162] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0052] Xanthopoulou I, Dragona VM, Davlouros P, Tsioufis C, Iliodromitis E, Alexopoulos D. Contemporary antithrombotic treatment in patients with non-valvular atrial fibrillation undergoing percutaneous coronary intervention: rationale and design of the Greek AntiPlatElet Atrial Fibrillation (GRAPE-AF) registry. Cardiovascular Drugs and Therapy 2018;32(2):191-6. [DOI: 10.1007/s10557-018-6789-7] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0053] Xiaoyan L, Ying L, Hong ZE, Ping Z. Safety and effectiveness of rivaroxaban in patients with acute coronary syndrome undergoing percutaneous coronary intervention. Journal of the American College of Cardiology 2014;64(16 Suppl 1):C150. [Google Scholar]

[CD013252-bib-0054] Yasuda S, Kaikita K, Ogawa H, Akao M, Ako J, Matoba T, et al. Atrial fibrillation and ischemic events with rivaroxaban in patients with stable coronary artery disease (AFIRE): protocol for a multicenter, prospective, randomized, open-label, parallel group study. International Journal of Cardiology 2018;265:108-12. [DOI: 10.1016/j.ijcard.2018.04.131] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0055] Yoshida R, Morishima I, Takagi K, Morita Y, Tsuboi H, Murohara T. Comparison between long-term clinical outcomes of vitamin K antagonist and direct oral anticoagulants in patients with atrial fibrillation undergoing percutaneous coronary intervention. Circulation Journal 2018;82(8):2016-24. [DOI: 10.1253/circj.CJ-17-1171] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0056] Gao F, Shen H, Wang ZJ, Yang SW, Liu XL, Zhou YJ. Rationale and design of the RT-AF study: combination of rivaroxaban and ticagrelor in patients with atrial fibrillation and coronary artery disease undergoing percutaneous coronary intervention. Contemporary Clinical Trials 2015;43:129-32. [DOI: 10.1016/j.cct.2015.05.012] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0057] NCT02789917. Edoxaban treatment versus vitamin K antagonist in patients with atrial fibrillation undergoing percutaneous coronary intervention (ENTRUST-AF-PCI). clinicaltrials.gov/ct2/show/NCT02866175 (first posted 15 August 2016).

[CD013252-bib-0058] NCT03536611. Dabigatran versus warfarin with NVAF who undergo PCI. clinicaltrials.gov/ct2/show/NCT03536611 (first posted 24 May 2018).

[CD013252-bib-0059] NCT02866175. Edoxaban treatment versus vitamin K antagonist in patients with atrial fibrillation undergoing percutaneous coronary intervention. clinicaltrials.gov/show/NCT02866175 (first posted 15 August 2016).

[CD013252-bib-0060] Vranckx P, Lewalter T, Valgimigli M, Tijssen JG, Reimitz PE, Eckardt L, et al. Evaluation of the safety and efficacy of an edoxaban-based antithrombotic regimen in patients with atrial fibrillation following successful percutaneous coronary intervention (PCI) with stent placement: rationale and design of the ENTRUST-AF PCI trial. American Heart Journal 2018;196:105-12. [DOI: 10.1016/j.ahj.2017.10.009] [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0061] Ahrens I, Peter K, Lip GY, Bode C. Development and clinical applications of novel oral anticoagulants. Part II. Drugs under clinical investigation. Discovery Medicine 2012;13(73):445-50. [PubMed] [Google Scholar]

[CD013252-bib-0062] Alexander JH, Becker RC, Bhatt DL, Cools F, Crea F, Dellborg M, et al. Apixaban, an oral, direct, selective factor Xa inhibitor, in combination with antiplatelet therapy after acute coronary syndrome: results of the Apixaban for Prevention of Acute Ischemic and Safety Events (APPRAISE) trial. Circulation 2009;119(22):2877-85. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0063] Alexander JH, Lopes RD, James S, Kilaru R, He Y, Mohan P, et al. Apixaban with antiplatelet therapy after acute coronary syndrome. New England Journal of Medicine 2011;365(8):699-708. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0064] Andreotti F, Testa L, Biondi-Zoccai GG, Crea F. Aspirin plus warfarin compared to aspirin alone after acute coronary syndromes: an updated and comprehensive meta-analysis of 25,307 patients. European Heart Journal 2006;27(5):519-26. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0065] Atar D, Bode C, Stuerzenbecher A, Verheugt FW. Anticoagulants for secondary prevention after acute myocardial infarction: lessons from the past decade. Fundamental Clinical Pharmacology 2014;28(4):353–63. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0066] Bauer KA. Pros and cons of new oral anticoagulants. Hematology 2013;2013(1):464-70. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0067] Bayer Pharma AG. Xarelto® (rivaroxaban) summary of product characteristics. www.ema.europa.eu/docs/en_GB/document_library/EPAR_-_Product_Information/human/000944/WC500057108.pdf (accessed 30 March 2018).

[CD013252-bib-0068] Benjamin EJ, Virani SS, Callaway CW, Chamberlain AM, Chang AR, Cheng S, et al. Heart disease and stroke statistics - 2018 update: a report from the American Heart Association. Circulation 2018;137(12):e67-492. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0069] Brignardello-Petersen R, Bonner A, Alexander PE, Siemieniuk RA, Furukawa TA, Rochwerg B, et al. Advances in the GRADE approach to rate the certainty in estimates from a network meta-analysis. Journal of Clinical Epidemiology 2018;93:36-44. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0070] Brignardello-Petersen R, Murad MH, Walter SD, McLeod S, Carrasco-Labra A, Rochwerg B, et al. GRADE approach to rate the certainty from a network meta-analysis: avoiding spurious judgments of imprecision in sparse networks. Journal of Clinical Epidemiology 2018;18:30497-9. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0071] Brignardello-Petersen R, Bonner A, Alexander PE, Siemieniuk RA, Furukawa TA, Rochwerg B, et al. Advances in the GRADE approach to rate the certainty in estimates from a network meta-analysis. Journal of Clinical Epidemiology 2018;93:36-44. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0072] Brignardello-Petersen R, Mustafa RA, Siemieniuk RA, Murad MH, Agoritsas T, Izcovich A, et al. GRADE approach to rate the certainty from a network meta-analysis: addressing incoherence. Journal of Clinical Epidemiology 2019;108:77-85. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0073] Brunetti ND, Tarantino N, De Gennaro L, Correale M, Santoro F, Di Biase M. Direct oral anticoagulants versus standard triple therapy in atrial fibrillation and PCI: meta-analysis. Open Heart 2018;5:000785. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0074] Camm AJ, Kirchhof P, Lip GY, Schotten U, Savelieva I, Ernst S, et al. Guidelines for the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Europace 2010;12(10):1360-420. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0075] Chaimani A, Caldwell DM, Li T, Higgins JP, Salanti G. Chapter 11: Undertaking network meta-analyses. In: Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for Systematic Reviews of Interventions version 6.0 (updated July 2019). Cochrane, 2019. Available from www.training.cochrane.org/handbook.

[CD013252-bib-0076] Cho JR, Angiolillo DJ. Percutaneous coronary intervention and atrial fibrillation: the triple therapy dilemma. Journal of Thrombosis and Thrombolysis 2015;39(2):203-8. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0077] Connolly S, Pogue J, Hart R, Pfeffer M, Hohnloser S, Chrolavicius S, et al. Clopidogrel plus aspirin versus oral anticoagulation for atrial fibrillation in the Atrial fibrillation Clopidogrel Trial with Irbesartan for prevention of Vascular Events (ACTIVE W): a randomised controlled trial. Lancet 2006;367(9526):1903-12. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0078] Connolly SJ, Ezekowitz MD, Yusuf S, Eikelboom J, Oldgren J, Parekh A, et al. Dabigatran versus warfarin in patients with atrial fibrillation. New England Journal of Medicine 2009;361(12):1139-51. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0079] Cuker A, Burnett A, Triller D, Crowther M, Ansell J, Van Cott EM, et al. Reversal of direct oral anticoagulants: guidance from the Anticoagulation Forum. American Journal of Hematology 2019;94:697. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0080] Deeks JJ, Higgins JP, Altman DG, editor(s), on behalf of the Cochrane Statistical Methods Group. Chapter 9: Analysing data and undertaking meta-analyses. In: Higgins JP, Churchill R, Chandler J, Cumpston MS, editor(s). Cochrane Handbook for Systematic Reviews of Interventions version 5.2.0 (updated June 2017). The Cochrane Collaboration, 2017. Available at www.training.cochrane.org/handbook.

[CD013252-bib-0081] Dewilde WJ, Oirbans T, Verheugt FW, Kelder JC, De Smet B, Herrman JP, et al. Use of clopidogrel with or without aspirin in patients taking oral anticoagulant therapy and undergoing percutaneous coronary intervention: an open-label, randomised, controlled trial. Lancet 2013;381(9872):1107-15. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0082] European Medicines Agency. Assessment report: Xarelto (Procedure No. EMEA/H/C/000944/X/00017) 2013. www.ema.europa.eu/docs/en GB/document library/EPAR - Assessment Report - Variation/human/000944/WC500144718.pdf (accessed 10 July 2013).

[CD013252-bib-0083] Fiedler KA, Maeng M, Mehilli J, Schulz-Schüpke S, Byrne RA, Sibbing D, et al. Duration of triple therapy in patients requiring oral anticoagulation after drug-eluting stent implantation: the ISAR-TRIPLE trial. Journal of the American College Cardiology 2015;65(16):1619-29. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0084] Finegold JA, Asaria P, Francis DP. Mortality from ischaemic heart disease by country, region, and age: statistics from World Health Organization and United Nations. International Journal of Cardiology 2013;168(2):934–45. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0085] Gibson CM, Mega JL, Burton P, Goto S, Verheugt F, Bode C, et al. Rationale and design of the anti-Xa therapy to lower cardiovascular events in addition to standard therapy in subjects with acute coronary syndrome-thrombolysis in myocardial infarction 51 (ATLAS-ACS 2 TIMI 51) trial: a randomized, double-blind, placebo-controlled study to evaluate the efficacy and safety of rivaroxaban in subjects with acute coronary syndrome. American Heart Journal 2011;161(5):815-21. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0086] Gibson CM, Mehran R, Bode C, Halperin J, Verheugt FW, Wildgoose P, et al. Prevention of bleeding in patients with atrial fibrillation undergoing PCI. New England Journal of Medicine 2016;375(25):2423-34. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0087] Giugliano RP, Ruff CT, Braunwald E, Murphy SA, Wiviott SD, Halperin JL, et al. Edoxaban versus warfarin in patients with atrial fibrillation. New England Journal of Medicine 2013;369(22):2093-104. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0088] Glund S, Moschetti V, Norris S, Stangier J, Schmohl M, Ryn J, et al. A randomised study in healthy volunteers to investigate the safety, tolerability and pharmacokinetics of idarucizumab, a specific antidote to dabigatran. Thrombosis and Haemostasis 2015;113(5):943-51. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0089] McMaster University (developed by Evidence Prime) GRADEpro GDT. Version accessed 14 January 2019. Hamilton (ON): McMaster University (developed by Evidence Prime), 2015. Available at gradepro.org.

[CD013252-bib-0090] Granger CB, Alexander JH, McMurray JJ, Lopes RD, Hylek EM, Hanna M, et al. Apixaban versus warfarin in patients with atrial fibrillation. New England Journal of Medicine 2011;365(11):981-92. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0091] Higgins JP, Churchill R, Chandler J, Cumpston MS, editor(s). Cochrane Handbook for Systematic Reviews of Interventions version 5.2.0 (updated June 2017). The Cochrane Collaboration, 2017. Available at www.training.cochrane.org/handbook.

[CD013252-bib-0092] Ibanez B, James S, Agewall S, Antunes MJ, Bucciarelli-Ducci C, Bueno H, et al. 2017 ESC guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation: the Task Force for the management of acute myocardial infarction in patients presenting with ST-segment elevation of the European Society of Cardiology (ESC). European Heart Journal 2018;39(2):119-77. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0093] Jackson D, White IR and Riley RD. Quantifying the impact of between-study heterogeneity in multivariate meta-analyses. Statistics in Medicine 2012;31:3805–20. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0094] Jansen JP, Crawford B, Bergman G, Stam W. Bayesian meta-analysis of multiple treatment comparisons: an introduction to mixed treatment comparisons. Value in Health 2008;11(5):956-64. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0095] January CT, Wann LS, Alpert JS, Calkins H, Cigarroa JE, Cleveland JC, et al. 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines and the Heart Rhythm Society. Circulation 2014;130(23):2071-104. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0096] Kearon C, Akl EA, Ornelas J, Blaivas A, Jimenez D, Bounameaux H, et al. Antithrombotic therapy for VTE disease: chest guideline and expert panel report. Chest 2016;149(2):315-52. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0097] Khera S, Kolte D, Bhatt DL. Chapter 16: Percutaneous coronary intervention. In: Aronow WS, McClung JA, editors(s). Translational Research in Coronary Artery Disease. Boston: Academic Press, 2016:179-94. [Google Scholar]

[CD013252-bib-0098] Kirchhof P, Benussi S, Kotecha D, Ahlsson A, Atar D, Casadei B, et al. 2016 ESC guidelines for the management of atrial fibrillation developed in collaboration with EACTS. European Journal of Cardio-thoracic Surgery 2016;50(5):e1-88. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0099] Knuuti J, Wijns W, Saraste A, Capodanno D, Barbato E, Funck-Brentano C, et al. ESC Scientific Document Group, 2019 ESC Guidelines for the diagnosis and management of chronic coronary syndromes: the Task Force for the diagnosis and management of chronic coronary syndromes of the European Society of Cardiology (ESC). European Heart Journal 2019;10:1093. [Google Scholar]

[CD013252-bib-0100] Kohli P, Waters DD. Looking for coronary disease in patients with atrial fibrillation. Canadian Journal of Cardiology 2014;30(8):861-3. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0101] Konstantinides SV, Torbicki A, Agnelli G, Danchin N, Fitzmaurice D, Galie N, et al. 2014 ESC guidelines on the diagnosis and management of acute pulmonary embolism. European Heart Journal 2014;35(43):3033-69. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0102] Krahn U, Binder H, König J. A graphical tool for locating inconsistency in network meta-analyses. BMC Medical Research Methodology 2013;13:35. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0103] Lamberts M, Gislason GH, Olesen JB, Kristensen SL, Schjerning Olsen AM, Mikkelsen A, et al. Oral anticoagulation and antiplatelets in atrial fibrillation patients after myocardial infarction and coronary intervention. Journal of the American College of Cardiology 2013;62(11):981-9. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0104] Lee LH. DOACs advances and limitations in real world. Thrombosis Journal 2016;14(Suppl 1):17. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0105] Lefebvre C, Manheimer E, Glanville J. Chapter 6: Searching for studies. In: Higgins JP, Green S, editor(s). Cochrane Handbook for Systematic Reviews of Interventions. Version 5.1.0 (updated March 2011). The Cochrane Collaboration, 2011. Available from handbook.cochrane.org.

[CD013252-bib-0106] Leon MB, Baim DS, Popma JJ, Gordon PC, Cutlip DE, Ho KK, et al. A clinical trial comparing three antithrombotic-drug regimens after coronary-artery stenting. New England Journal of Medicine 1998;339(23):1665-71. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0107] Levine GN, Bates ER, Bittl JA, Brindis RG, Fihn SD, Fleisher LA, et al. ACC/AHA guideline focused update on duration of dual antiplatelet therapy in patients with coronary artery disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Journal of the American College of Cardiology 2016;68:1082–115. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0108] Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. PLoS Medicine 2009;6(7):e1000100. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0109] Lip GY, Windecker S, Huber K, Kirchhof P, Marin F, Ten Berg JM, et al. Management of antithrombotic therapy in atrial fibrillation patients presenting with acute coronary syndrome and/or undergoing percutaneous coronary or valve interventions: a joint consensus document of the European Society of Cardiology Working Group on Thrombosis, European Heart Rhythm Association (EHRA), European Association of Percutaneous Cardiovascular Interventions (EAPCI) and European Association of Acute Cardiac Care (ACCA) endorsed by the Heart Rhythm Society (HRS) and Asia-Pacific Heart Rhythm Society (APHRS). European Heart Journal 2014;35(45):3155-79. [DOI] [PubMed]

[CD013252-bib-0110] Lopes RD, Hong H, Harskamp RE. Safety and efficacy of antithrombotic strategies in patients with atrial fibrillation undergoing percutaneous coronary intervention: a network meta-analysis of randomized controlled trials. JAMA Cardiology 2019;10:1001. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0111] Lu G, Ades AE. Combination of direct and indirect evidence in mixed treatment comparisons. Statistics in Medicine 2004;23(20):3105-24. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0112] Mackman N. Triggers, targets and treatments for thrombosis. Nature 2008;451(7181):914-8. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0113] Mauri L, Kereiakes DJ, Yeh RW, Driscoll-Shempp P, Cutlip DE, Steg G, et al. Twelve or 30 months of dual antiplatelet therapy after drug-eluting stents. New England Journal of Medicine 2014;371(23):2155–66. [DOI] [PMC free article] [PubMed]

[CD013252-bib-0114] McDermott KW, Freeman WJ, Elixhauser A. Overview of operating room procedures during inpatient stays in U.S. hospitals, 2014. Rockville (MD): Agency for Healthcare Research and Quality (AHRQ), Healthcare Cost and Utilization Project (HCUP); 2017 December. Statistical Brief #233. [PubMed]

[CD013252-bib-0115] McGrath PD, Malenka DJ, Wennberg DE, Shubrooks SJ, Bradley WA, Robb JF, et al. Changing outcomes in percutaneous coronary interventions: a study of 34,752 procedures in Northern New England, 1990 to 1997. Journal of the American College of Cardiology 1999;34(3):674-80. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0116] Mega JL, Braunwald E, Wiviott SD, Bassand JP, Bhatt D, Bode C, et al. Rivaroxaban in patients with a recent acute coronary syndrome. New England Journal of Medicine 2012;366(1):9-19. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0117] Mehran R, Rao SV, Bhatt DL, Gibson CM, Caixeta A, Eikelboom J, et al. Standardized bleeding definitions for cardiovascular clinical trials: a consensus report from the Bleeding Academic Research Consortium. Circulation 2011;123:2736-47. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0118] Mekaj YH, Mekaj AY, Duci SB, Miftari EI. New oral anticoagulants: their advantages and disadvantages compared with vitamin K antagonists in the prevention and treatment of patients with thromboembolic events. Therapeutics and Clinical Risk Management 2015;11:967-77. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0119] Merlini PA, Bauer KA, Oltrona L, Ardissino D, Cattaneo M, Belli C, et al. Persistent activation of coagulation mechanism in unstable angina and myocardial infarction. Circulation 1994;90(1):61-8. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0120] Miller CS, Grandi SM, Shimony A, Filion KB, Eisenberg MJ. Meta-analysis of efficacy and safety of new oral anticoagulants (dabigatran, rivaroxaban, apixaban) versus warfarin in patients with atrial fibrillation. American Journal of Cardiology 2012;110(3):453-60. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0121] Mills EJ, Thorlund K, Ioannidis JP. Demystifying trial networks and network meta-analysis. BMJ 2013;346:f2914. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0122] Moon JY, Nagaraju D, Franchi F, Rollini F, Angiolillo DJ. The role of oral anticoagulant therapy in patients with acute coronary syndrome. Therapeutic Advances in Hematology 2017;8(12):353-66. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0123] Neelankavil J, Rau CD, Wang Y. The genetic basis of coronary artery diseases and atrial fibrillation: a search for disease mechanisms and therapeutic targets. Journal of Cardiothoracic and Vascular Anesthesia 2015;29(5):1328-32. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0124] Oldgren J, Budaj A, Granger CB, Khder Y, Roberts J, Siegbahn A, et al. Dabigatran vs. placebo in patients with acute coronary syndromes on dual antiplatelet therapy: a randomized, double-blind, phase II trial. European Heart Journal 2011;32(22):2781-9. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0125] Paikin JS, Wright DS, Crowther MA, Mehta SR, Eikelboom JW. Triple antithrombotic therapy in patients with atrial fibrillation and coronary artery stents. Circulation 2010;121(18):2067-70. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0126] Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. New England Journal of Medicine 2011;365(10):883-91. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0127] Pérez-Gómez F, Alegría E, Berjón J, Iriarte JA, Zumalde J, Salvador A, et al. Comparative effects of antiplatelet, anticoagulant, or combined therapy in patients with valvular and nonvalvular atrial fibrillation: a randomized multicenter study. Journal of the American College of Cardiology 2004;44(8):1557-66. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0128] Peterson ED, Lansky AJ, Anstrom KJ, Muhlbaier LH, Popma JJ, Satler LF, et al. Evolving trends in interventional device use and outcomes: results from the National Cardiovascular Network database. American Heart Journal 2000;139(2):198-207. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0129] Pollack CV, Reilly PA, Eikelboom J, Glund S, Verhamme P, Bernstein RA, et al. Idarucizumab for dabigatran reversal. New England Journal of Medicine 2015;373(6):511-20. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0130] Puhan MA, Schünemann HJ, Murad MH, Li T, Brignardello-Petersen R, Singh JA, et al. Approach for rating the quality of treatment effect estimates from network meta-analysis. BMJ 2014;349:g5630. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0131] Nordic Cochrane Centre, The Cochrane Collaboration Review Manager 5 (RevMan 5). Version 5.3. Copenhagen: Nordic Cochrane Centre, The Cochrane Collaboration, 2014.

[CD013252-bib-0132] Roffi M, Patrono C, Collet JP, Mueller C, Valgimigli M, Andreotti F, et al. 2015 ESC guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: Task Force for the Management of Acute Coronary Syndromes in Patients Presenting Without Persistent ST-Segment Elevation of the European Society of Cardiology (ESC). European Heart Journal 2016;37(3):267-315. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0133] Rothberg MB, Celestin C, Fiore LD, Lawler E, Cook JR. Warfarin plus aspirin after myocardial infarction or the acute coronary syndrome: meta-analysis with estimates of risk and benefit. Annals of Internal Medicine 2005;143(4):241-50. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0134] Rubboli A, Colletta M, Herzfeld J, Sangiorgio P, Di Pasquale G. Periprocedural and medium-term antithrombotic strategies in patients with an indication for long-term anticoagulation undergoing coronary angiography and intervention. Coronary Artery Disease 2007;18(3):193-9. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0135] Rücker G, Schwarzer G. Ranking treatments in frequentist network meta-analysis works without resampling methods. BMC Medical Research Methodology 2015;15:58. [DOI] [PMC free article] [PubMed]

[CD013252-bib-0136] Rücker G, Schwarzer G, Krahn U, König J. Network meta-analysis using frequentist methods - package netmeta. cran.r–project.org/web/packages/netmeta/ (accessed prior to 12 January 2019).

[CD013252-bib-0137] Ruff CT, Giugliano RP, Braunwald E, Hoffman EB, Deenadayalu N, Ezekowitz MD, 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(9921):955-62. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0138] Schünemann HJ, Oxman AD, Vist GE, Higgins JP, Deeks JJ, Glasziou P, et al. Chapter 12: Interpreting results and drawing conclusions. In: Higgins JP, Churchill R, Chandler J, Cumpston MS, editor(s). Cochrane Handbook for Systematic Reviews of Interventions version 5.2.0 (updated June 2017). The Cochrane Collaboration, 2017. Available at www.training.cochrane.org/handbook.

[CD013252-bib-0139] Steg GP, Mehta SR, Jukema JW, Lip GY, Gibson CM, Kovar F, et al. RUBY-1: a randomized, double-blind, placebo-controlled trial of the safety and tolerability of the novel oral factor Xa inhibitor darexaban (YM150) following acute coronary syndrome. European Heart Journal 2011;32(20):2541-54. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CD013252-bib-0140] Valgimigli M, Bueno H, Byrne RA, Collet JP, Costa F, Jeppsson A, et al. 2017 ESC focused update on dual antiplatelet therapy in coronary artery disease developed in collaboration with EACTS: the Task Force for dual antiplatelet therapy in coronary artery disease of the European Society of Cardiology (ESC) and of the European Association for Cardio-Thoracic Surgery (EACTS). European Heart Journal 2017;39(3):213-60. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0141] Wallentin L, Wilcox RG, Weaver WD, Emanuelsson H, Goodvin A, Nystroem P, et al. Oral ximelagatran for secondary prophylaxis after myocardial infarction: the ESTEEM randomised controlled trial. Lancet 2003;362(9386):789-97. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0142] Wallentin L, Becker RC, Budaj A, Cannon CP, Emanuelsson H, Held C, et al. Ticagrelor versus clopidogrel in patients with acute coronary syndromes. New England Journal of Medicine 2009;361:1045–57. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0143] Wang TY, Robinson LA, Ou FS, Roe MT, Ohman EM, Gibler WB, et al. Discharge antithrombotic strategies among patients with acute coronary syndrome previously on warfarin anticoagulation: physician practice in the CRUSADE registry. American Heart Journal 2008;155(2):361-8. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0144] World Health Organization. Cardiovascular diseases. www.who.int/cardiovascular_diseases/en/ (accessed 30 March 2018).

[CD013252-bib-0145] Windecker S, Kolh P, Alfonso F, Collet JP, Cremer J, Falk V, et al. 2014 ESC/EACTS guidelines on myocardial revascularization: the Task Force on Myocardial Revascularization of the European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (EACTS). Developed with the special contribution of the European Association of Percutaneous Cardiovascular Interventions (EAPCI). European Heart Journal 2014;35(37):2541-619. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0146] Wiviott SD, Braunwald E, McCabe CH, Montalescot G, Ruzyllo W, Gottlieb S, et al. Prasugrel versus clopidogrel in patients with acute coronary syndromes. New England Journal of Medicine 2007;357:2001-15. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0147] Yusuf S, Zhao F, Mehta SR, Chrolavicius S, Tognoni G, Fox KK. Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST-segment elevation. New England Journal of Medicine 2001;345:494–502. [DOI] [PubMed] [Google Scholar]

[CD013252-bib-0148] Zirlik A, Bode C. Vitamin K antagonists: relative strengths and weaknesses vs. direct oral anticoagulants for stroke prevention in patients with atrial fibrillation. Journal of Thrombosis and Thrombolysis 2017;43(3):365-79. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Non‐vitamin K antagonist oral anticoagulants (NOACs) post‐percutaneous coronary intervention: a network meta‐analysis

Samer Al Said

Samer Alabed

Klaus Kaier

Audrey R Tan

Christoph Bode

Joerg J Meerpohl

Daniel Duerschmied

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

Summary of findings

Background

Description of the condition

Description of the intervention

How the intervention might work

Why it is important to do this review

1.

2.

3.

4.

5.

6.

7.

8.

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Types of interventions

Types of outcome measures

Primary outcomes

Secondary outcomes

Search methods for identification of studies

Electronic searches

Searching other resources

Data collection and analysis

Selection of studies

Data extraction and management

Data on potential effect modifiers

Assessment of risk of bias in included studies

Assessment of bias in conducting the systematic review

Measures of treatment effect

Unit of analysis issues

Dealing with missing data

Assessment of heterogeneity

Assessment of intransitivity across treatment comparisons

Assessment of reporting biases

Data synthesis

Methods for direct treatment comparisons

Methods for indirect and network comparisons

Subgroup analysis and investigation of heterogeneity

Sensitivity analysis

Summary of findings and assessment of the certainty of the evidence

'Summary of findings' table

League table

Results

Description of studies

Results of the search

9.

Included studies

Study designs

Sample sizes

Participants

Interventions

Outcomes

Excluded studies

Ongoing trials

Risk of bias in included studies

10.

11.

Allocation