Elsevier

The Lancet

Volume 388, Issue 10055, 22–28 October 2016, Pages 1995-2003
The Lancet

Articles
Edoxaban versus enoxaparin–warfarin in patients undergoing cardioversion of atrial fibrillation (ENSURE-AF): a randomised, open-label, phase 3b trial

https://doi.org/10.1016/S0140-6736(16)31474-XGet rights and content

Summary

Background

Edoxaban, an oral factor Xa inhibitor, is non-inferior for prevention of stroke and systemic embolism in patients with atrial fibrillation and is associated with less bleeding than well controlled warfarin therapy. Few safety data about edoxaban in patients undergoing electrical cardioversion are available.

Methods

We did a multicentre, prospective, randomised, open-label, blinded-endpoint evaluation trial in 19 countries with 239 sites comparing edoxaban 60 mg per day with enoxaparin–warfarin in patients undergoing electrical cardioversion of non-valvular atrial fibrillation. The dose of edoxaban was reduced to 30 mg per day if one or more factors (creatinine clearance 15–50 mL/min, low bodyweight [≤60 kg], or concomitant use of P-glycoprotein inhibitors) were present. Block randomisation (block size four)—stratified by cardioversion approach (transoesophageal echocardiography [TEE] or not), anticoagulant experience, selected edoxaban dose, and region—was done through a voice-web system. The primary efficacy endpoint was a composite of stroke, systemic embolic event, myocardial infarction, and cardiovascular mortality, analysed by intention to treat. The primary safety endpoint was major and clinically relevant non-major (CRNM) bleeding in patients who received at least one dose of study drug. Follow-up was 28 days on study drug after cardioversion plus 30 days to assess safety. This trial is registered with ClinicalTrials.gov, number NCT02072434.

Findings

Between March 25, 2014, and Oct 28, 2015, 2199 patients were enrolled and randomly assigned to receive edoxaban (n=1095) or enoxaparin–warfarin (n=1104). The mean age was 64 years (SD 10·54) and mean CHA2DS2-VASc score was 2·6 (SD 1·4). Mean time in therapeutic range on warfarin was 70·8% (SD 27·4). The primary efficacy endpoint occurred in five (<1%) patients in the edoxaban group versus 11 (1%) in the enoxaparin–warfarin group (odds ratio [OR] 0·46, 95% CI 0·12–1·43). The primary safety endpoint occurred in 16 (1%) of 1067 patients given edoxaban versus 11 (1%) of 1082 patients given enoxaparin–warfarin (OR 1·48, 95% CI 0·64–3·55). The results were independent of the TEE-guided strategy and anticoagulation status.

Interpretation

ENSURE-AF is the largest prospective randomised clinical trial of anticoagulation for cardioversion of patients with non-valvular atrial fibrillation. Rates of major and CRNM bleeding and thromboembolism were low in the two treatment groups.

Funding

Daiichi Sankyo provided financial support for the study.

Introduction

Restoration of sinus rhythm in patients with atrial fibrillation is often performed to reduce symptoms. Rhythm control can be achieved by electrical cardioversion, by antiarrhythmic drugs (so-called pharmacological cardioversion), or by both antiarrhythmic drugs and electrical cardioversion.

Cardioversion confers a risk of thromboembolism peri-cardioversion and current guidelines recommend a minimum of 3 weeks of therapeutic anticoagulation (whether oral or parenteral) before elective cardioversion, and continuation of anticoagulation for a minimum of 4 weeks after cardioversion (and longer in patients at risk of atrial fibrillation recurrence or if stroke risk factors are present).1, 2 Cardioversion can be guided by transoesophageal echocardiography (TEE) to exclude left atrial appendage thrombi, which, if present, contraindicate cardioversion.

Vitamin K antagonists (eg, warfarin) have traditionally been used as oral anticoagulation for cardioversion, but vitamin K antagonists have various limitations with substantial interpatient and intrapatient variability, requiring regular monitoring to ensure therapeutic anticoagulation within a target international normalised ratio (INR) range of 2·0–3·0.3 The difficulties of managing vitamin K antagonists led to variable time to achieve the therapeutic range. In some cases the time to achieve the therapeutic range can be prolonged, hence delaying the cardioversion procedure for a minimum of 3 weeks of therapeutic anticoagulation before elective cardioversion. This delay is particularly relevant in medical care settings with no access to TEE, and heparin bridging is sometimes used to optimise anticoagulation pending therapeutic anticoagulation with vitamin K antagonists.

Research in context

Evidence before this study

Post-hoc studies of cardioversion, although limited because of lengthy anticoagulation before the cardioversion, have shown a good safety profile for non-vitamin K antagonist oral anticoagulants (NOACs). One prospective study evaluating rivaroxaban against non-optimised vitamin K antagonist management suggested a good safety profile of this agent. All of these studies are limited by small numbers. Data about the use of NOACs for peri-cardioversion have been provided by post-hoc subgroup analyses from large phase 3 stroke prevention trials and one randomised trial. There are limited data with edoxaban in the setting of electrical cardioversion of non-valvular atrial fibrillation.

Added value of this study

The ENSURE-AF study is the largest randomised clinical trial of anticoagulation for cardioversion in patients with atrial fibrillation, with findings showing very low event rates against exceptionally well controlled warfarin. This trial also provides the largest prospective trial data for an NOAC (ie, edoxaban) in this clinical setting. Edoxaban was also compared against optimised standard care (warfarin with enoxaparin bridging), with the time in the therapeutic range on warfarin being more than 70%, and excellent adherence to edoxaban therapy with more than 99% compliance.

Implications of all the available evidence

The NOACs seem to be safe and user friendly, and provide rapid onset of oral anticoagulation; they provide an alternative to heparin plus warfarin.

More recently, the non-vitamin K antagonist oral anticoagulants (NOACs) have been introduced to overcome some of the limitations associated with vitamin K antagonists.4 Edoxaban had non-inferior efficacy and superior safety (20% reduction in major bleeding and a 53% reduction in intracranial bleeding) compared with well controlled warfarin for the prevention of stroke and systemic embolism in patients with non-valvular atrial fibrillation in the ENGAGE AF-TIMI 48 study.5

Data about NOACs for peri-cardioversion have been provided by post-hoc subgroup analyses from large phase 3 stroke prevention trials6, 7, 8, 9 and one randomised trial (X-VERT).10 These data suggest that NOACs would probably be safe in the peri-cardioversion setting and could be an alternative to vitamin K antagonists. In the ENGAGE AF-TIMI 48 trial, there were relatively few cardioversion procedures, and thus only limited data about edoxaban in the setting of electrical cardioversion of non-valvular atrial fibrillation are available.9 In a phase 3b prospective randomised trial (the ENSURE-AF study), we aimed to assess the efficacy and safety of edoxaban compared with best possible conventional therapy (enoxaparin–warfarin) in patients with atrial fibrillation undergoing cardioversion. For the comparator conventional therapy, we used enoxaparin bridging to reduce time to achieve the therapeutic range and to cover transient suboptimal anticoagulation periods with warfarin alone peri-cardioversion.

Section snippets

Study design and participants

The detailed design of the ENSURE-AF study has been reported previously.11 In brief, we designed a prospective, randomised, open-label, parallel-group study with blinded-endpoint evaluation. Eligible patients were those with non-valvular atrial fibrillation (of duration no shorter than 48 h and no longer than 12 months) and who were planned for electrical cardioversion and anticoagulation therapy. Patients who were receiving or had received previous anticoagulant or antiplatelet therapies (or

Results

Between March 25, 2014, and Oct 28, 2015, 2199 patients were enrolled and randomly assigned in the trial (figure), 1095 to edoxaban and 1104 to enoxaparin–warfarin. 988 (90%) of those assigned to edoxaban and 966 (88%) of those assigned to enoxaparin–warfarin were cardioverted electrically or cardioverted spontaneously. The mean age of participants was 64 years and two-thirds of patients were men (table 1). The mean CHA2DS2-VASc score was 2·6 (SD 1·4) in both treatment groups (table 1).

Discussion

In this prospective randomised trial of patients with atrial fibrillation undergoing cardioversion, edoxaban had similar rates of major bleeding and thromboembolism compared with patients on well managed, optimised enoxaparin–warfarin therapy. The results were similar between conventional or TEE-guided strategies, previous anticoagulation status, and across all other main subgroups.

The evidence base for warfarin was provided by observational cohort studies, and no randomised trial of warfarin

References (14)

There are more references available in the full text version of this article.

Cited by (218)

  • The risk and timing of acute ischemic stroke after electrical cardioversion for atrial fibrillation in Taiwan: A nationwide population-based cohort study

    2022, International Journal of Cardiology
    Citation Excerpt :

    The timing of AIS after ECV for AF may play an important role in the duration of postcardioversion anticoagulation. Previous studies have documented a positive association between the use of ECV and the risk of AIS [11–15] and the timing of thromboembolic events after ECV for AF [16], but limited studies have been conducted in Asian populations in recent years. Therefore, we sought to use the National Health Insurance Research Database (NHIRD) in Taiwan to determine the risk and timing of AIS after ECV for AF.

View all citing articles on Scopus

Co-principal investigators and joint senior authors

For the full list of investigators see appendix pp 10–13

View full text