OUP user menu

★ fast track ★

Rivaroxaban vs. vitamin K antagonists for cardioversion in atrial fibrillation

Riccardo Cappato, Michael D. Ezekowitz, Allan L. Klein, A. John Camm, Chang-Sheng Ma, Jean-Yves Le Heuzey, Mario Talajic, Maurício Scanavacca, Panos E. Vardas, Paulus Kirchhof, Melanie Hemmrich, Vivian Lanius, Isabelle Ling Meng, Peter Wildgoose, Martin van Eickels, Stefan H. Hohnloser
DOI: http://dx.doi.org/10.1093/eurheartj/ehu367 First published online: 2 September 2014

Abstract

Aims X-VeRT is the first prospective randomized trial of a novel oral anticoagulant in patients with atrial fibrillation undergoing elective cardioversion.

Methods and results We assigned 1504 patients to rivaroxaban (20 mg once daily, 15 mg if creatinine clearance was between 30 and 49 mL/min) or dose-adjusted vitamin K antagonists (VKAs) in a 2:1 ratio. Investigators selected either an early (target period of 1–5 days after randomization) or delayed (3–8 weeks) cardioversion strategy. The primary efficacy outcome was the composite of stroke, transient ischaemic attack, peripheral embolism, myocardial infarction, and cardiovascular death. The primary safety outcome was major bleeding. The primary efficacy outcome occurred in 5 (two strokes) of 978 patients (0.51%) in the rivaroxaban group and in 5 (two strokes) of 492 patients (1.02%) in the VKA group [risk ratio 0.50; 95% confidence interval (CI) 0.15–1.73]. In the rivaroxaban group, four patients experienced primary efficacy events following early cardioversion (0.71%) and one following delayed cardioversion (0.24%). In the VKA group, three patients had primary efficacy events following early cardioversion (1.08%) and two following delayed cardioversion (0.93%). Rivaroxaban was associated with a significantly shorter time to cardioversion compared with VKAs (P < 0.001). Major bleeding occurred in six patients (0.6%) in the rivaroxaban group and four patients (0.8%) in the VKA group (risk ratio 0.76; 95% CI 0.21–2.67).

Conclusion Oral rivaroxaban appears to be an effective and safe alternative to VKAs and may allow prompt cardioversion.

Name of the trial registry Clinicaltrials.gov; Trial registration number: NCT01674647.

  • Cardioversion
  • Oral anticoagulant
  • Stroke
  • Thromboembolism

Introduction

Atrial fibrillation (AF) is the most frequently encountered sustained cardiac arrhythmia, with a prevalence of about 1% in the general population.1 In symptomatic patients, pharmacological or electrical cardioversion can be used to rapidly restore sinus rhythm.1 However, there is a peri-procedural risk of thromboembolic events associated with cardioversion, with stroke rates between 5 and 7% in non-anticoagulated patients.24 Vitamin K antagonist (VKA) therapy, although never validated in controlled clinical trials, reduces the peri-procedural incidence of thromboembolic events to between 0.5 and 1.6%.4,5 Current European Society of Cardiology and American Heart Association/American College of Cardiology/Heart Rhythm Society guidelines recommend at least 3 weeks of effective anticoagulation before cardioversion, followed by at least 4 weeks of anticoagulation after the procedure.1,6 The use of transesophageal echocardiography to rule out left atrial (LA) thrombus plus heparin and VKA treatment immediately before, during, and for at least 4 more weeks after cardioversion is effective to expedite cardioversion.6

Novel oral anticoagulants are alternatives to VKAs for long-term stroke prevention in patients with non-valvular AF.710 In addition, recent post hoc analyses have found dabigatran, rivaroxaban, and apixaban to be as safe and effective as VKA treatment in the setting of cardioversion when the pre-cardioversion anticoagulation time period is long.1114 This study was designed to explore prospectively the efficacy and safety of once-daily rivaroxaban compared with dose-adjusted VKA treatment (with or without heparin), in anticoagulation-naïve or -experienced patients undergoing elective cardioversion.

Methods

X-VeRT (eXplore the efficacy and safety of once-daily oral riVaroxaban for the prevention of caRdiovascular events in patients with non-valvular aTrial fibrillation scheduled for cardioversion) was a multinational, randomized, open-label, parallel-group phase IIIb study of patients with haemodynamically stable non-valvular AF of >48 h or of unknown duration. Details of the study protocol have been published previously.15 Briefly, patients scheduled for cardioversion were randomly assigned to rivaroxaban or VKA therapy in a 2:1 ratio. The decision regarding early cardioversion (a goal of between 1 and 5 days of rivaroxaban or usual VKA therapy before the procedure) or delayed cardioversion (rivaroxaban or VKA for 3–8 weeks prior to the procedure) was made by the local investigator. Randomization to rivaroxaban or VKA treatment was performed using an Interactive Voice and Web Response System.

Patients and treatment regimens

Patients aged 18 years or older scheduled for elective electrical or pharmacological cardioversion were eligible for the trial. Patients could be naïve to oral anticoagulation or could have been previously anticoagulated with a VKA or novel oral anticoagulant. The main exclusion criteria were haemodynamically significant mitral valve stenosis, prosthetic heart valves, known LA thrombi, severe disabling stroke within the previous 3 months, and any stroke or transient ischaemic attack up to 2 weeks or 3 days, respectively, prior to randomization. The study protocol was approved by local ethics committees at participating centres and patients provided written informed consent to participate.

Patients randomized to rivaroxaban received a once-daily dose of 20 mg orally (or 15 mg once daily in patients with creatinine clearance of 30–49 mL/min). Patients randomized to the VKA arm received warfarin or another VKA at the investigator's discretion, based on local standard of care. The target international normalized ratio (INR) was 2.5 (range 2.0–3.0). Investigators had the option to use a parenteral anticoagulant drug in addition to VKA therapy especially prior to cardioversion until the target INR was obtained.

Cardioversion strategies and follow-up

Because of the potential adverse consequences of AF, it is desirable to conduct cardioversion as soon as possible. According to guidelines, an early cardioversion strategy can be followed either when transesophageal echocardiography rules out an LA thrombus or if ≥3 weeks’ pre-treatment with therapeutic oral anticoagulation is proven.1,6 In the early cardioversion strategy group in X-VeRT, rivaroxaban or a VKA was given with a goal of between 1 and 5 days before planned cardioversion and continued for 6 weeks post-cardioversion.15 In patients randomized to rivaroxaban, medication was started at least 4 h before cardioversion. Patients with a LA thrombus detected during the study did not undergo cardioversion. In these patients, study treatment was stopped and patients were treated according to local standard of care and followed for 30 days. In the delayed cardioversion strategy group, patients were treated with either a VKA or rivaroxaban for at least 3 weeks and up to a maximum of 8 weeks before cardioversion. Oral anticoagulation with a VKA was considered adequate if the INR was maintained in the range 2.0–3.0 for at least three consecutive weeks prior to cardioversion. Oral anticoagulation with rivaroxaban was considered adequate if the pill count was ≥80% for three consecutive weeks prior to cardioversion. Rivaroxaban or the VKA was continued for 6 weeks after cardioversion.

After study termination, patients assigned to rivaroxaban could transition to non-study rivaroxaban (or another novel oral anticoagulant) or to a VKA (if INR ≥2.0) or to VKA plus parenteral anticoagulants. In patients who were treated with a VKA, transition to a novel oral anticoagulant occurred when the INR was ≤3 for rivaroxaban or ≤2.0 for dabigatran or apixaban.

Concurrent medications and procedures

The use of strong inhibitors of both cytochrome P450 3A4 and P-glycoprotein, a VKA or factor Xa inhibitors other than study medication, factor IIa inhibitors, low-molecular-weight heparin or unfractionated heparin unless for short-term bridging of VKA therapy, chronic acetylsalicylic acid therapy >100 mg daily, or dual antiplatelet therapy were not permitted in the study. Strong inducers of cytochrome P450 3A4 could be administered with caution. Patients receiving concomitant treatments that affect haemostasis were to be monitored carefully during the study. Patients at risk of ulcerative gastrointestinal disease or bleeding could receive prophylactic treatment with proton-pump inhibitors. When an invasive or surgical intervention was required, rivaroxaban was discontinued 24 h before the intervention and restarted afterwards as soon as possible.15 Interruption or restarting of VKA therapy was according to usual practice.

Outcome assessment

Clinical events were adjudicated by an independent, blinded clinical events committee. The primary efficacy outcome was the composite of all adjudicated events classified as stroke or transient ischaemic attack, peripheral embolism, myocardial infarction, and cardiovascular death. Secondary efficacy outcomes included adjudicated all-cause mortality, a composite of stroke, transient ischaemic attack, peripheral embolism, myocardial infarction, and all-cause mortality, and individual components of the primary efficacy end-point.

The primary safety outcome was major bleeding. Major bleeding events were defined according to the International Society of Thrombosis and Haemostasis criteria.16 The secondary safety end-point was all bleeding events.

Sample size and statistical analysis

Assuming the risk for thromboembolic events within 30 days after cardioversion in patients assigned to a VKA is 1%, we estimated that between 25 000 and 30 000 patients would be required to establish that rivaroxaban is non-inferior to VKA at a non-inferiority margin of 1.5 with 90% power and a 2:1 randomization in favour of rivaroxaban. We concluded that a trial of this size was not feasible. Using the post hoc analysis of cardioversions in the RE-LY trial with dabigatran in 1270 patients as a guide,11 we decided that a descriptive comparison involving 1500 participants would give clinically meaningful information. The statistical analyses were descriptive. We estimated the risk and risk ratios for outcome events including 95% confidence intervals (CIs). Efficacy analyses were performed using the modified intention-to-treat (mITT) population that excluded patients with a LA thrombus and the intention-to-treat (ITT) population, including all randomized patients. The mITT population was used for the primary efficacy analysis. Safety analyses were performed in the safety analysis population, which included patients who received at least one dose of study medication. The study period for efficacy analyses was defined as the time from randomization until either the date of last dose of study medication plus 2 days (for patients who completed the planned study medication period) or the earlier date of the last planned dose of study medication (e.g. 42 days after cardioversion) and the end of the 30-day follow-up (for patients who prematurely discontinued study medication).

Results

Patient population, treatment assignment, and cardioversion

A total of 1584 patients were screened between 3 October 2012 and 25 September 2013, and 1504 patients were randomized (Figure 1) at 141 centres from 16 countries. Overall, 1002 patients were assigned to rivaroxaban and 502 to VKA. Thirty-five patients withdrew consent during the treatment phase, of whom 17 were confirmed alive at the end of follow-up. Of the remaining 18 for whom no further information was available, five were never treated with the study drug.

Figure 1

Study patient flow for scheduled cardioversion during the target time period (ITT population). ITT, intention-to-treat; LA, left atrial; LAA, left atrial appendage. aAs scheduled; for early cardioversion: 1–5 days after randomization; for delayed cardioversion: 21–25 days after randomization. bNot further specified.

Thirty-four patients (rivaroxaban: 24; VKA: 10) were not included in the mITT population of 1470 patients (rivaroxaban: 978; VKA: 492; Supplementary material online, Figure S1). The characteristics of randomized patients were well balanced between the two treatment groups (Table 1). Forty-three percent of patients were experienced to oral anticoagulants, defined as ≥6 weeks of oral anticoagulation. Before randomization, 53 (3.5%) patients had received dabigatran, 88 (5.9%) rivaroxaban, 2 (0.1%) apixaban, and 51.3% (772/1504) VKAs. Fifty percent (504/1002) of patients were to be transitioned from pre-study treatment with VKAs to rivaroxaban, and 5.2% (26/502) of patients were to be transitioned from rivaroxaban to VKA.

View this table:
Table 1

Demographics (ITT population)

Total by treatmentEarlyDelayed
Rivaroxaban (n = 1002)VKA (n = 502)Rivaroxaban (n = 585)VKA (n = 287)Rivaroxaban (n = 417)VKA (n = 215)
Region, n (%)
 Europe728 (72.7)364 (72.5)414 (70.8)205 (71.4)314 (75.3)159 (74.0)
 North America221 (22.1)111 (22.1)121 (20.7)59 (20.6)100 (24.0)52 (24.2)
 Asia-Pacific53 (5.3)27 (5.4)50 (8.5)23 (8.0)3 (0.7)4 (1.9)
Gender: male, n (%)727 (72.6)367 (73.1)426 (72.8)203 (70.7)301 (72.2)164 (76.3)
Age (years): mean ± SD64.9 ± 10.664.7 ± 10.565.3 ± 10.465.3 ± 10.664.4 ± 10.864.0 ± 10.3
BMI (kg/m2): mean ± SD30.09 ± 5.8330.19 ± 6.0729.75 ± 5.8229.82 ± 6.2430.55 ± 5.8130.69 ± 5.82
Creatinine clearance,a n (%)
 <30 mL/min01 (0.2)0001 (0.5)
 30–≤50 mL/min68 (6.8)30 (6.0)50 (8.5)17 (5.9)18 (4.3)13 (6.0)
 50–<80 mL/min310 (30.9)176 (35.1)174 (29.7)114 (39.7)136 (32.6)62 (28.8)
 ≥80 mL/min616 (61.5)289 (57.6)355 (60.7)152 (53.0)261 (62.6)137 (63.7)
Medical history, n (%)
 Prior stroke34 (3.4)21 (4.2)18 (3.1)14 (4.9)16 (3.8)7 (3.3)
 Prior non-CNS SE10 (1.0)11 (2.2)4 (0.7)6 (2.1)6 (1.4)5 (2.3)
 Prior TIA23 (2.3)17 (3.4)15 (2.6)7 (2.4)8 (1.9)10 (4.7)
 Congestive HF197 (19.7)75 (14.9)124 (21.2)52 (18.1)73 (17.5)23 (10.7)
  NYHA class III/IV48 (4.8)9 (1.8)33 (5.6)8 (2.8)15 (3.6)1 (0.5)
 Arterial hypertension651 (65.0)345 (68.7)398 (68.0)196 (68.3)253 (60.7)148 (68.8)
 Diabetes mellitus203 (20.3)103 (20.5)125 (21.4)68 (23.6)78 (18.7)35 (16.3)
 Vascular disease134 (13.4)56 (11.2)98 (16.8)32 (11.1)36 (8.6)24 (11.2)
 MI90 (9.0)33 (6.6)62 (10.6)17 (5.9)28 (6.7)16 (7.4)
Atrial fibrillation, n (%)
 First diagnosed238 (23.8)106 (21.1)104 (17.8)52 (18.1)134 (32.1)54 (25.1)
 Paroxysmal172 (17.2)114 (22.7)124 (21.2)85 (29.6)48 (11.5)29 (13.5)
 Persistent560 (55.9)251 (50.0)339 (57.9)135 (47.0)221 (53.0)116 (54.0)
 Long-standing persistent30 (3.0)26 (5.2)17 (2.9)12 (4.2)13 (3.1)14 (6.5)
CHADS2 score, n (%)
 0239 (23.9)105 (20.9)119 (20.3)52 (18.1)120 (28.8)53 (24.7)
 1381 (38.0)203 (40.4)235 (40.2)118 (41.1)146 (35.0)85 (39.5)
 ≥2382 (38.1)194 (38.6)231 (39.5)117 (40.8)151 (36.2)77 (35.8)
CHA2DS2-VASc score, n (%)
 0 (or 1, if female only)147 (14.7)65 (12.9)67 (11.5)31 (10.8) 80 (19.2)34 (15.8)
 1 (except for female alone)215 (21.5)118 (23.5)128 (21.9)66 (23.0) 87 (20.9)52 (24.2)
 ≥2640 (63.9)319 (63.5)390 (66.7)190 (66.2)250 (60.0)129 (60.0)
OAC experienced,b n (%)424 (42.3)220 (43.8)280 (47.9)130 (45.3)144 (34.5)90 (41.9)
Antiplatelet agents289 (28.8)153 (30.5)157 (26.8)90 (31.4)132 (31.7)63 (29.3)
 ASA266 (26.5)142 (28.3)145 (24.8)87 (30.3)121 (29.0)55 (25.6)
 Clopidogrel21 (2.1)9 (1.8)13 (2.2)5 (1.7)8 (1.9)4 (1.9)
Antiarrhythmic drugs, n (%)813 (81.1)407 (81.1)471 (80.5)237 (82.6)342 (82.0)170 (79.1)
 Amiodarone181 (18.1)100 (19.9)129 (22.1)62 (21.6)52 (12.5)38 (17.7)
 Dronedarone13 (1.3)8 (1.6)7 (1.2)7 (2.4)6 (1.4)1 (0.5)
 Flecainide50 (5.0)22 (4.4)32 (5.5)14 (4.9)18 (4.3)8 (3.8)
 Propafenone15 (1.5)5 (1.0)8 (1.4)3 (1.0)7 (1.7)2 (1.0)
 Others, classes I and IIIc2 (0.2)1 (0.2)2 (0.3)1 (0.3)00
  • ASA, acetylsalicylic acid; BMI, body mass index; CNS, central nervous system; HF, heart failure; ITT, intention-to-treat; MI, myocardial infarction; NYHA, New York Heart Association; OAC, oral anticoagulant; SD, standard deviation; SE, systemic embolism; TIA, transient ischaemic attack; VKA, vitamin K antagonist.

  • aCreatinine clearance calculated by Cockcroft–Gault formula.

  • bOAC experienced: oral anticoagulant use for 6 weeks or longer prior to first study medication intake.

  • cDisopyramide, dofetilide, or quinidine.

Overall, 872 (58%) patients were scheduled to undergo early cardioversion with transesophageal echocardiography performed in 564/872 (64.7%; rivaroxaban: 377; VKA: 187) patients. A total of 632 (42%) patients were scheduled to undergo delayed cardioversion, with transesophageal echocardiography performed in 64/632 (10.1%; rivaroxaban: 33; VKA: 31) patients.

Overall, 1167 patients (77.6%) underwent electrical (97.6%) or pharmacological (2.4%) cardioversion within the target time range of 1–5 days (early) or 21–25 days (delayed cardioversion) after randomization. In the delayed group, 321/417 (77.0%) patients in the rivaroxaban arm compared with 78/215 (36.3%) patients in the VKA arm were cardioverted within the target time range (P < 0.001), primarily due to failure to achieve adequate anticoagulation (rivaroxaban: 1 patient, VKA: 95 patients). The acute cardioversion success rate was 86.8% (1013/1167) and was similar by treatment arm [rivaroxaban: 735/841 (87.4%); VKA: 278/326 (85.3%)] as well as by cardioversion strategy, 86.5 and 87.5% in the early and delayed cardioversion groups, respectively.

Patients who could not be cardioverted within the target time ranges continued on study treatment and 115 patients (rivaroxaban: 21; VKA: 94) had cardioversion performed at a later visit during the treatment phase. In 116 (7.7%) patients (rivaroxaban: 76; VKA: 40), spontaneous cardioversion was observed before an interventional cardioversion was performed.

Overall, the time between randomization and cardioversion was similar or shorter in patients assigned to rivaroxaban [early: median = 1 (interquartile range: 1–2) vs. 1 (1–3) days, P = 0.628; delayed: 22 (21–26) vs. 30 (23–42) days, P < 0.001].

Efficacy outcomes

Primary outcome events were experienced in 10/1470 (0.68%; 95% CI 0.36–1.21%) patients in the mITT population. The cumulative risk for this composite outcome was 5/978 (0.51%; 95% CI 0.20–1.17%) for patients assigned to receive rivaroxaban and 5/492 (1.02%; 95% CI 0.40–2.34%) for patients assigned to receive VKA, with a risk ratio for rivaroxaban to VKA of 0.50 (95% CI 0.15–1.73) (Table 2). Table 3 reports individual outcome events. There were two patients with strokes in each treatment group (rivaroxaban: 0.20%; 95% CI 0.04–0.71%; VKA: 0.41%; 95% CI 0.07–1.41%), one patient with systemic embolism in the VKA group, one patient with myocardial infarction in each treatment arm, and six patients with cardiovascular deaths (four in the rivaroxaban group and two in the VKA group). One additional non-cardiovascular (cancer-related) death was reported in each treatment arm. In the early cardioversion strategy, primary efficacy outcome events occurred in 4/567 (0.71%; 95% CI 0.24–1.76%) rivaroxaban-treated patients and 3/277 (1.08%; 95% CI 0.30–3.06%) VKA-treated patients, whereas in the delayed cardioversion strategy group they occurred in 1/411 (0.24%; 95% CI 0.01–1.29%) patients and 2/215 (0.93%; 95% CI 0.17–3.26%) patients in the rivaroxaban and VKA groups, respectively. Out of the 10 patients who experienced primary efficacy outcome events, nine events occurred within the first 21 days after cardioversion; one patient died before cardioversion. When the outcome events occurred, 9 of the 10 patients were on study treatment; one cardioverted patient discontinued study treatment 10 days after randomization and died 10 days later. The cumulative incidence risk for the composite outcome of stroke, non-central nervous system embolism, transient ischaemic attack, myocardial infarction, and all-cause mortality was 6/978 (0.61%; 95% CI 0.27–1.29%) in patients receiving rivaroxaban and 6/492 (1.22%; 95% CI 0.53–2.51%) in patients receiving VKAs (risk ratio 0.50; 95% CI 0.16–1.55).

View this table:
Table 2

Number of patients with outcome events

Total by treatmentEarlyDelayed
RivaroxabanVKARR (95% CI)RivaroxabanVKARivaroxabanVKA
Efficacy, n (%)an = 978n = 492n = 567n = 277n = 411n = 215
 Primary end-point5 (0.51)5 (1.02)0.50 (0.15–1.73)4 (0.71)3 (1.08)1 (0.24)2 (0.93)
  Stroke2 (0.20)2 (0.41)2 (0.35)1 (0.36)01 (0.47)
  Haemorrhagic stroke2 (0.20)02 (0.35)000
  Ischaemic stroke02 (0.41)01 (0.36)01 (0.47)
  TIA000000
  SE01 (0.20)01 (0.36)00
  MI1 (0.10)1 (0.20)1 (0.18)001 (0.47)
  Cardiovascular death4 (0.41)2 (0.41)3 (0.53)2 (0.72)1 (0.24)0
 All-cause death5 (0.51)3 (0.61)3 (0.53)3 (1.08)2 (0.49)0
Safety, n (%)bn = 988n = 499n = 575n = 284n = 413n = 215
 Major bleeding6 (0.61)4 (0.80)0.76 (0.21–2.67)3 (0.52)3 (1.06)3 (0.73)1 (0.47)
  Fatal1 (0.10)2 (0.40)1 (0.17)2 (0.70)00
  Critical site2 (0.20)3 (0.60)2 (0.35)2 (0.70)01 (0.47)
   ICH2 (0.20)1 (0.20)2 (0.35)001 (0.47)
  Hb decrease ≥2 g/dL4 (0.40)1 (0.20)1 (0.17)1 (0.35)3 (0.73)0
  Transfusion ≥2 units RBCs or whole blood3 (0.30)1 (0.20)1 (0.17)1 (0.35)2 (0.48)0
  • Cumulative incidence risk for adjudicated outcomes from randomization up to the date of last study medication plus 2 days. Per protocol, study treatment was to be continued up to 42 days after cardioversion.

  • CI, confidence interval; CV, cardiovascular; Hb, haemoglobin; ICH, intracranial haemorrhage; MI, myocardial infarction; N/A, not applicable; RBC, red blood cells; RR, risk ratio; SE, systemic embolism; TIA, transient ischaemic attack; VKA, vitamin K antagonist.

  • amITT population.

  • bSafety population.

View this table:
Table 3

Adjudicated individual efficacy outcome events

Adjudicated event(s)Age/genderNYHA class III/IVRelevant medical historyCHADS2 scoreCHA2DS2-VASc scoreCardioversion strategyCardioversion performed/number of attempts/successCardioversion typebDays after randomizationDays before(−)/after(+) cardioversionTEE performed
Rivaroxaban groupa
 Sudden unexpected death69/MNoCHF/HTN/liver disease/bladder cancer23EarlyYes/1/yesE7+6Yes (no thrombus)
 MI/CV death76/MNoHTN/stroke/syncope/renal disease/pacemaker45EarlyYes/1/yesE6 (MI)

7(death)
+1 (MI)

+2 (death)
Yes (no thrombus)
 Stroke/intracerebral bleed, non-fatalc79/MNoDM/gastric ulcer/prostate cancer/angina pectoris/sarcoidosis23EarlyYes/2/yesE18+15Yes (no thrombus)
 Stroke/intracerebral bleed, fatalc69/FNoHTN/ischaemic stroke/CHD/cardiomyopathy/breast cancer35EarlyYes/–/yesP21 (stroke)

23 (death)
+16 (stroke)

+18 (death)
Yes (no thrombus)
 CV death/CHF89/MNoPacemaker/hypothyroid/HF12Delayed–/–/–None33
 All-cause death/lung cancer70/MNoLower urinary tract symptoms01DelayedYes/1/yesE79+58
Adjudicated event(s)Age/genderNYHA class III/IVRelevant medical historyCHADS2 scoreCHA2DS2-VASc scoreCardioversion strategyCardioversion performed/number of attempts/successCardioversion typebDays after randomizationDays before (−)/after (+) cardioversionTEE performedINR at time of event
VKA group
 All-cause death/lung cancerc79/MNoHTN/prostate cancer/pulmonary squamous cell carcinoma23EarlyYes/2/yesE26+26Yes (no thrombus)1.2
 Ischaemic stroke, non-fatal76/MNoHTN/RBBB23EarlyYes/2/yesE4+2Yes (no thrombus)
 Non-CNS SE/CV death/all-cause death (autopsy documented, fatalc)72/MNoHTN/complex aortic plaque/PAD/dyslipidaemia/left leg above knee amputation; aortobifemoral bypass24EarlyYes/1/yesE9+9Yes (no thrombus)7.4
 CV death63/MYesCHF/MI/cardiomyopathy/pacemaker12EarlyYes/–/noP21+19Yes (no thrombus)
 Ischaemic stroke, non-fatal55/MNoHTN11DelayedYes/1/yesE32+4
 MI, non-fatal58/MNoChronic obstructive pulmonary disease /Crohn's/renal disease/GERD/fatty liver00DelayedYes/2/noE48+131.7
  • AF, atrial fibrillation; CHD, coronary heart disease; CHF, congestive heart failure; CV, cardiovascular; HD, heart disease; HTN, hypertension; INR, international normalized ratio; MI, myocardial infarction; NSTEMI, non-ST-elevation myocardial infarction; NYHA, New York Heart Association; RBBB, right bundle branch block; SE, systemic embolism; TEE, transesophageal echocardiogram.

  • a20 mg once daily, down-titrated to 15 mg once daily if creatinine clearance 30–49 mL/min.

  • bCardioversion type: E, electrical; P, pharmacological.

  • cEvent also qualified as major bleeding (refer to Table 4).

In OAC-naïve or untreated patients, the primary efficacy outcome events occurred in 4/565 (0.71%) patients in the rivaroxaban group and 3/273 (1.10%) patients in the VKA group. In patients with prior OAC use, the respective incidences were 1/413 (0.24%) in the rivaroxaban group and 2/219 (0.91%) in the VKA group. Patients undergoing transesophageal echocardiogram (TEE) experienced seven primary efficacy events (four in the rivaroxaban arm, and three in the VKA arm).

In the ITT population, the same 10 patients as in the mITT population experienced adjudicated primary efficacy outcomes because no outcome events were observed in patients excluded due to an LA thrombus (Supplementary material online, Table S1). This resulted in a marginally lower estimated risk of the primary efficacy outcome in the larger ITT population (0.50; 95% CI 0.15–1.72). The results were consistent across a large number of pre-specified subgroups. Of 1415 patients (0.28%), 4 experienced a primary outcome event during post-treatment follow-up.

Safety outcomes

Major bleeding occurred in 6/988 (0.61%; 95% CI 0.26%–1.27%) patients in the rivaroxaban group and 4/499 (0.80%; 95% CI 0.27–2.00%) patients in the VKA group (risk ratio 0.76; 95% CI 0.21–2.67) (Table 2). Intracerebral bleeding occurred in two (0.2%) patients in the rivaroxaban group and one (0.2%) patient in the VKA group. Fatal bleeding was reported in one (0.1%) patient in the rivaroxaban group and two (0.4%) patients in the VKA group. In patients scheduled for early cardioversion, the incidence of major bleeding was 3/575 (0.5%) patients in the rivaroxaban group and 3/284 (1.1%) patients in the VKA group (Table 4). The risk of the secondary safety outcome (any confirmed bleeding events) was similar between the two treatment arms (8.9 and 7.2% for the rivaroxaban and VKA groups, respectively). Patients undergoing TEE experienced four primary safety events (two in the rivaroxaban arm, and two in the VKA arm).

View this table:
Table 4

Adjudicated individual major bleeding events on treatment

Adjudicated event(s)Age/genderNYHA class III/IVRelevant medical historyCHADS2 scoreCHA2DS2-VASc scoreCardioversion strategyCardioversion performed/number of attempts/successCardioversion typeaDays after randomizationDays before (−)/after(+) cardioversionTEE performed
Rivaroxaban group
 Major intracerebral bleed, non-fatalb79/MNoDM/gastric ulcer/prostate cancer/angina pectoris/sarcoidosis23EarlyYes/2/yesE18+15Yes (no thrombus)
 Major GI bleed, non-fatal54/MNoHTN/DM/anaemia22EarlyYes/3/yesE3+3
 Major intracerebral bleed, fatalb69/FNoHTN/ischaemic stroke/CHD/cardiomyopathy/breast cancer35EarlyYes/–/yesP21+16Yes (no thrombus)
 Major GI bleed, non-fatal72/FNoCHF/CAD36DelayedYes/1/yesE13−8
 Major GI bleed, non-fatal79/FNoHTN/TIA/gastric ulcer/GI bleed46DelayedYes/1/yesE64+12
 Major vaginal bleed, non-fatal51/FNoCholecyst-ectomy/atrial septal defect/right eye amaurosis01Delayed–/–/–None4
Adjudicated event(s)Age/genderNYHA class III/IVRelevant medical historyCHADS2 scoreCHA2DS2-VASc scoreCardioversion strategyCardioversion performed/number of attempts/successCardioversion typeaDays after randomizationDays before (−) /after (+) cardioversionTEE performedINR at time of event
VKA group
 Major pulmonary bleed/lung cancer, fatalb79/MNoHTN/prostate cancer/pulmonary squamous cell carcinoma23EarlyYes/2/yesE24+24Yes (no thrombus)5.3
 Major vitreous haemorrhage, non-fatal54/MNoHTN11EarlyYes/1/yesE31+30
 Major GI bleed, fatalb72/MNoHTN/complexaortic plaque/PAD/left leg above-knee amputation/dyslipidaemia/aortobifemoral bypass24EarlyYes/1/yesE9+9Yes (no thrombus)7.4
 Major subdural haematoma, non-fatal62/MNoHTN11DelayedYes/1/yesE58+361.8
  • AF, atrial fibrillation; DM, diabetes mellitus; HTN, hypertension; INR, international normalized ratio; NYHA, New York Heart Association; PAD, peripheral artery disease; TIA transient ischaemic attack; TEE, transesophageal echocardiography; VKA, vitamin K antagonist.

  • aCardioversion type: E, electrical; P, pharmacological.

  • bEvent also qualified as efficacy outcome events (refer to Table 3).

Treatment-emergent serious adverse events were reported in 8.8% of patients in total, of which 1.1% were assessed to be drug related. No clinically important differences in the overall cumulative incidence of adverse events and serious adverse events by treatment assignment or by cardioversion strategy were observed.

Discussion

X-VeRT is the first completed prospective trial of a novel oral anticoagulant in patients with AF undergoing elective cardioversion. Rivaroxaban administered de novo, or as ongoing therapy, or as a replacement for VKAs or another anticoagulant agent was associated with thromboembolic and bleeding risks that were low and similar to those observed with VKA treatment. This observation applied to both the early and delayed cardioversion strategies. A net clinical benefit outcome (the composite of stroke, non-central nervous system systemic embolism, transient ischaemic attack, myocardial infarction, cardiovascular death, and major bleeding) occurred in 6/978 (1.06%) patients receiving rivaroxaban and 5/492 (1.81%) patients receiving VKA (risk ratio 0.49; 95% CI 0.14–1.69). In the delayed cardioversion group, rivaroxaban allowed cardioversion after a shorter treatment period (mean 25 days) compared with VKAs (mean 34 days) because of the inability to achieve adequate anticoagulation prior to cardioversion in the VKA group at 3 weeks (95 patients compared with 1 patient in the rivaroxaban group). In the early cardioversion group, rivaroxaban administered at least 4 h before cardioversion provided effective and safe anticoagulation. Results were consistent across all analysis sets (mITT, ITT, and safety) and in pre-specified subgroups.

The use of VKAs before and after cardioversion is the current standard practice endorsed by guideline recommendations.1,6 A major obstacle to using a VKA is the observation that >3 weeks are required to achieve stable therapeutic INR values.17 This finding was confirmed in this study. The pharmacological characteristics of the novel oral anticoagulants are particularly useful in the setting of elective cardioversion. Their rapid onset of action (2–4 h), short half-life, and predictable pharmacokinetics and pharmacodynamics allow a more rapid cardioversion strategy. This study adds to the data from phase III clinical trials from which post hoc analyses using the direct thrombin inhibitor dabigatran,11 or direct factor Xa inhibitors rivaroxaban and apixaban,12,13 were conducted. In the RE-LY post hoc analyses, the largest to date, the frequencies of stroke and major bleeding within 30 days after cardioversion were found to be low on chronic treatment with dabigatran similar to those observed with warfarin.11 Similar findings were found in the other post hoc analyses involving apixaban and rivaroxaban.12,13

X-VeRT was underpowered to provide statistically rigorous results and was thus exploratory in nature. Several findings, however, substantiate the strength of the results of X-VeRT. The estimated risk ratios consistently indicated a trend towards lower incidences of thromboembolic events and major bleeding events in favour of rivaroxaban in the total population as well as in the early and delayed cardioversion subgroups (Table 2). The 95% upper confidence limits of incidences for thromboembolic events (1.17%) and major bleeding events (1.27%) in the rivaroxaban arm correspond to efficacy and safety incidences, which are well within the range of those reported in previous series of VKA-treated patients.4,5 The risks for efficacy events in the rivaroxaban arm were much lower than those reported in the absence of anticoagulant therapy (5–7%).18 The practical advantage of using rivaroxaban was demonstrated by the short time to cardioversion and the low number of patients failing to achieve adequate anticoagulation pre-cardioversion at 3 weeks in the delayed cardioversion group. When considering the results, it should be noted that the study was conducted in a heterogeneous real-world population. The open-label randomization used here could have introduced a bias in the reporting and/or adjudication of outcome events. In order to reduce this bias, several validated procedures, including blinded evaluation of outcome events, were employed, as previously reported in RE-LY.7 A similar proportion of patients (15.6% in the rivaroxaban arm and 20.3% in the VKA arm) discontinued drug treatment owing to suboptimal compliance.

In summary, oral rivaroxaban appears to be an effective and safe alternative to VKA and may allow prompter cardioversion.

Funding

This work was supported by Bayer HealthCare Pharmaceuticals and Janssen Scientific Affairs LLC. The authors would also like to thank Judy Spahr and Paulette Trent who provided editorial support, also with funding from Bayer HealthCare Pharmaceuticals and Janssen Scientific Affairs LLC.

Conflict of interest: M.D.E. is a consultant and speaker for Boehringer Ingelheim and a consultant for Pfizer, Sanofi-Aventis, Bristol-Myers Squibb, Portola, Bayer, Daichii Sankyo, Medtronic, Aegerion, Merck, Johnson & Johnson, Gilead, Janssen Scientific Affairs, Pozen Inc., and Coherex. A.L.K. has received honoraria from book publishers Elsevier and LWW. C.S.M. reports no relevant disclosures. R.C. has received consultancy fees from Boston Scientific, Medtronic, St. Jude, Biosense Webster, ELA Sorin, Boehringer Ingelheim, Bayer HealthCare, and Abbott; Pfizer; speaker's bureau fees from Boston Scientific, Medtronic, St. Jude, Biosense Webster, BARD, sanofi-aventis, Boehringer Ingelheim, Bayer HealthCare, and Abbott; investigator fees from Medtronic, Biosense Webster sanofi-aventis, Cameron Health, BARD, Bayer HealthCare, Abbott, and Pfizer; grants from Boston Scientific, Medtronic, St. Jude, Biosense Webster, BARD, and ELA Sorin; and holds equity and intellectual property rights with Cameron Health. A.J.C. has received personal fees from Sanofi-Aventis, Boehringer Ingelheim, Bayer, Bristol-Myers Squibb/Pfizer, Daiichi Sankyo, GlaxoSmithKline, Meda, and Servier. J.Y.L.H. has received consultancy fees from Sanofi-Aventis, Boehringer Ingelheim, Bayer, Bristol-Myers Squibb/Pfizer, Daiichi Sankyo, GlaxoSmithKline, Meda, and Servier. M.T. has received consultancy fees and research funding from Bayer, Boehringer Ingelheim, and Bristol-Myers Squibb. M.S. has received consultancy and lecture fees from Bayer, Bristol-Myers Squibb/Pfizer, Johnson & Johnson, St Jude Medical, and Boston Scientific. P.E.V. has received educational grants and consultancy fees from Servier, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Medtronic, Menarini, Respicardia, and Pfizer. P.K. has received consultancy fees and honoraria from 3M Medica, AstraZeneca, Bayer, Boehringer Ingelheim, Boston Scientific, Bristol-Myers Squibb, Cardiome, Daiichi Sankyo, MEDA Pharma, Medtronic, Merck, Otsuka, Pfizer, Sanofi-Aventis, Servier, Siemens, and Takeda; and research grants from 3M Medica, Cardiovascular Therapeutics, Daiichi Sankyo, MEDA Pharma, Medtronic, OMRON, and St Jude Medical. S.H.H. has received consultancy and lecture fees from Sanofi-Aventis, Boehringer Ingelheim, Bayer, Bristol-Myers Squibb/Pfizer, Johnson & Johnson, St Jude Medical, Cardiome, Gilead, Servier, Boston Scientific, and Medtronic. V.L., M.H., I.L.M., and M.vE. are employees of Bayer HealthCare; P.W. is an employee of Janssen Scientific Affairs, LLC and a shareholder in Johnson & Johnson.

Acknowledgements

The authors would like to thank Susanne Hess of Bayer HealthCare for her assistance in the project, Dr Kathrin Schmidt and Stephan Cichos of Clinical Data Management, Bayer HealthCare, and Albert A. Volkl of Janssen for his contribution to the concept and design of the study. The authors would also like to thank the three members of the Data Monitoring Committee (DMC)—Dr Alain Leizorovicz (DMC Chairman), Prof. Dr Günter Breithardt, and Prof. Dr Hans-Christoph Diener—and the Clinical Events Committee, chaired by Prof. Dr Martin Prins.

References

View Abstract