Effects and Risks of Heparin Bridging and Different Modes of Interruption/Resumption of Rivaroxaban on Perioperative Complications of Catheter Ablation for Atrial Fibrillation　― Another Analysis of the JACRE-R Registry ―

Abstract

Background: The JACRE-R Registry, in which 42 Japanese institutions participated, monitored the efficacy and safety of rivaroxaban in catheter ablation (CA) of atrial fibrillation (AF). In the present analysis, we sought to elucidate the effects and risks of heparin bridging and different patterns of interruption/resumption of rivaroxaban on complications of CA.

Methods and Results: We administered rivaroxaban during the perioperative period and recorded the incidence of complications up to 30 days after CA. A total of 1,118 patients were registered; 546 received heparin bridging and 572 did not. The bridging group showed a significantly higher incidence of non-major bleeding than the no-bridging group (4.03% vs. 0.87%; P=0.001). In the group receiving their last dose of rivaroxaban at 8–28 h before CA, neither thromboembolism nor major bleeding was observed during or after CA and the incidence of non-major bleeding was low (4/435, 0.92%). The incidence of non-major bleeding was significantly higher in the group resuming rivaroxaban ≥12 h after CA than in the group resuming <12 h (1.79% vs. 0.27%, P=0.045).

Conclusions: Heparin bridging increased the risk of non-major bleeding perioperatively. It was safe to stop rivaroxaban 8–28 h before the CA procedure, whereas resumption of the drug within 12 h of CA was associated with a lower incidence of non-major bleeding.

Catheter ablation (CA) is an established therapy for drug-refractory atrial fibrillation (AF). Direct oral anticoagulants (DOACs) have been approved since 2011 for nonvalvular AF (NVAF) and widely used as an alternative to warfarin for the prevention of thromboembolism in patients with NVAF who are expected to undergo CA. Recent studies have demonstrated that in terms of efficacy and safety, DOACs are comparable to warfarin for thromboembolism prophylaxis.^1–6 Several reports state that uninterrupted perioperative treatment with warfarin is both effective and safe;⁷ however, heparin bridging for warfarin is associated with an increased risk of bleeding.^8,9 Meanwhile, the efficacy and safety of heparin bridging for DOACs are not well known.¹⁰ In addition, a comparative study to determine when to interrupt and resume DOACs is needed. Rivaroxaban given once daily is widely used today and these clinical questions about DOACs need to be addressed immediately. We prospectively enrolled and followed up patients receiving rivaroxaban during the CA perioperative period and investigated the clinical details of perioperative heparin bridging and the patterns of interruption/resumption of rivaroxaban, evaluating its efficacy and safety.

Methods

A prospective registry, Japanese Anti-Coagulation Regimen Exploration in AF Catheter Ablation Registry-Rivaroxaban cohort (JACRE-R), consisted of 42 Japanese high-volume institutions performing CA from July 2014 to September 2015. The study outline was deposited in the University Hospital Medical Information Network Clinical Trials Registry (UMIN000014817). NVAF patients receiving rivaroxaban for ≥3 weeks before a scheduled CA were consecutively enrolled and followed up for 30 days after CA. The eligibility criteria were previously described.¹¹ The study protocol was approved by the ethical committees of the participating institutions and all patients provided written informed consent.

Perioperative Anticoagulation Management and Ablation Procedures

The patients received rivaroxaban once daily according to the approved dosage and administration in Japan. Perioperative interruption of rivaroxaban and use of heparin bridging were left to the investigators’ discretion. In addition, no restriction was imposed on ablation procedures. Activated clotting time (ACT) was maintained >300 s during the ablation procedure by administering heparin at appropriate doses.¹¹

Efficacy and Safety Endpoints

Efficacy was evaluated by the incidence of thromboembolic complications, including transient ischemic attacks (TIAs), cerebral infarction, and other systemic embolisms. Safety was gauged by bleeding complications associated with CA procedures and that occurred within 30 days of completion. Major bleeding events were defined as follows: pericardial effusion requiring cardiac drainage; hematoma at the puncture site requiring surgical treatment; and other hemorrhagic complications requiring transfusion. Non-major bleeding events included complications not requiring invasive treatment. The primary endpoint of this study was the composite of thromboembolism and major bleeding events during follow-up; the secondary endpoints included thromboembolism, major bleeding, and non-major bleeding.

On-site sample source data verification was performed for 5% of the JACRE-R patients. An adjudication committee, consisting of 2 neurologists and 2 cardiologists independent of the steering committee, the protocol committee and Bayer Yakuhin, Ltd., judged the blinded data and finally determined the primary and secondary endpoint events.

ACT at sheath removal was also measured and analyzed in association with stratification/outcome variables.

Statistical Analysis

In the statistical analysis, patients were stratified according to heparin bridging, ACT at sheath removal, and mode of interruption/resumption of rivaroxaban. The patients’ characteristics were compared among the groups using Pearson’s chi-square test or Fisher’s exact test for nominal variables and the Mann-Whitney U-test for ordered and continuous variables. The incidence of events was compared using the log-rank test. To identify factors associated with the occurrence of endpoint events, we conducted univariate Cox regression analysis and subsequent multivariate analysis for the parameters considered significant. In the Cox models, baseline imbalance was adjusted with the propensity score method and adjusted hazard ratios (HR) were calculated. To address the sparseness of events, Firth’s penalized likelihood method was applied.¹² In the statistical analysis, we excluded the patients with reasonable confounding factors that might have affected the postoperative treatments or outcomes. To examine the association between the drug withdrawal period (last dose of rivaroxaban before the procedure) and complication occurrence, we conducted a receiver-operating characteristic (ROC) analysis. The threshold of withdrawal period, sensitivity, specificity, and area under the curve (AUC) with 95% confidence interval were calculated. Multiplicity was corrected using the Bonferroni method.

Two-sided P-values <0.05 were considered significant. Statistical analyses were carried out using IBM SPSS Statistics 22.0 (IBM Corp., Armonk, NY, USA) and R version 3.0.2 (R Foundation for Statistical Computing, Vienna, Austria).

Results

A total of 1,118 patients were enrolled in the rivaroxaban cohort; 1,113 (99.6%) completed follow-up. Rivaroxaban dosing was not resumed postoperatively in 5 patients because of laboratory abnormalities necessitating a change to another drug (2 patients), patient request (1), discontinuation because of major bleeding (1), and death (1). Complications in the intraoperative period (between sheath insertion and sheath removal) occurred in 7 patients. Postoperative complications included thromboembolism in 2 patients (0.2%), major bleeding in 5 (0.4%), and non-major bleeding in 27 patients (2.4%).

Effect of Heparin Bridging

Heparin bridging was done in 546 patients (48.8%): both before CA (before insertion of the sheath) and after CA (after removal of the sheath) in 209 patients (18.7%); only before the operation in 122 (10.9%); and only after the operation in 215 patients (19.2%). The other 572 patients (51.2%) did not receive heparin bridging therapy. No significant difference in thromboembolism or major bleeding was observed between patients with and without heparin bridging (Table 1). In the stratification analysis between the bridging and no-bridging groups, the bridging group manifested a significantly higher incidence of non-major bleeding than the no-bridging group (4.03% vs. 0.87%; P=0.001; Table 2). No significant difference in the incidence of non-major bleeding was observed in terms of preoperative heparin bridging. The postoperative heparin bridging group had a significantly higher incidence of non-major bleeding than the group without postoperative heparin bridging (3.80% vs. 1.01%; P=0.002; Table 2). Clinical data of patients with and without postoperative heparin bridging are summarized in Table S1.

Table 1. Associations of Presence/Absence and Timing of Heparin Bridging With Complications After CA for NVAF

Heparin bridging		n	Thrombo embolism and major bleeding	Thromboembolism	Major bleeding	Non-major bleeding
Preoperative	Postoperative
+	+	209	1 (0.48%)	1 (0.48%)	0 (0.00%)	7 (3.35%)
+	−	122	2 (1.64%)	0 (0.00%)	2 (1.64%)	3 (2.46%)
−	+	215	0 (0.00%)	0 (0.00%)	0 (0.00%)	12 (5.58%)
−	−	572	4 (0.70%)	1 (0.17%)	3 (0.52%)	5 (0.87%)
Total		1,118	7 (0.63%)	2 (0.18%)	5 (0.45%)	27 (2.42%)
For 1,111 patients, excluding 7 patients with intraoperative occurrence of an event
+	+	208	1 (0.48%)	1 (0.48%)	0 (0.00%)	6 (2.88%)
+	−	122	2 (1.64%)	0 (0.00%)	2 (1.64%)	3 (2.46%)
−	+	213	0 (0.00%)	0 (0.00%)	0 (0.00%)	10 (4.69%)
−	−	568	1 (0.18%)	1 (0.18%)	0 (0.00%)	4 (0.70%)
Total		1,111	4 (0.36%)	2 (0.18%)	2 (0.18%)	23 (2.07%)

+, Heparin bridging; −, no heparin bridging. CA, catheter ablation; NVAF, nonvalvular atrial fibrillation.

Table 2. Associations of Presence/Absence and Timing of Heparin Bridging With Non-Major Bleeding After CA for NVAF

	Heparin bridging		n	Non-major bleeding	%	Log-rank test P value
	Preoperative	Postoperative
All patients
Comparison between patients with and without heparin bridging	+	+	546	22	4.03	0.001
	+	−
	−	+
	−	−	572	5	0.87
Comparison between patients with and without preoperative heparin bridging	+	+	331	10	3.02	0.397
	+	−
	−	+	787	17	2.16
	−	−
Excluding 7 patients with intraoperative occurrence of an event
Comparison between patients with and without postoperative	+	+	421	16	3.80	0.002
	−	+
	+	−	690	7	1.01
	−	−

+, Heparin bridging; −, no heparin bridging. Abbreviations as in Table 1.

Influence of ACT at Sheath Removal

The mean ACT at sheath removal was 314.7±45.0 s (mean±standard deviation) (n=1,100), and the majority of patients (n=716, 65.5%) showed ACT ≥300 s at sheath removal. Of this patient group with ACT ≥300 s at sheath removal (n=716), 218 (30.4%) received postoperative heparin bridging, whereas of the group with ACT <250 s (n=59), 41 (69.5%) had heparin bridging (P<0.001, Pearson’s Chi-squared test with Yates’ continuity correction) (Table S2).

Because postoperative heparin bridging was likely to affect the occurrence of bleeding complications, we excluded these patients and examined the influence of ACT at sheath removal on subsequent events. The patient group with ACT ≥350 s at sheath removal showed a significantly higher incidence of major bleeding events (1.3% vs. 0%, P=0.011) (Table 3), which included pericardial effusion and hematoma at the puncture site (1 patient each).

Table 3. ACT at Sheath Removal and Complications of CA for NVAF

	ACT <350 s (n=521)	ACT ≥350 s (n=160)	Log-rank test P value
	n (%)	n (%)
Composite event of thromboembolism and major bleeding	1 (0.2)	2 (1.3)	0.077
Thromboembolism	1 (0.2)	0 (0.0)	0.582
Major bleeding	0 (0.0)	2 (1.3)	0.011
Non-major bleeding	5 (1.0)	2 (1.3)	0.750

Analysis set: 681 patients, excluding those with postoperative heparin bridging, intraoperative occurrence of an event, and unavailable ACT. ACT, activated clotting time. Other abbreviations as in Table 1.

Multivariate Cox Regression Analysis of the Events

In this study, occurrences of outcome events were quite sparse, adversely affecting the robustness of the multivariate Cox models despite the use of Firth’s correction method. We therefore focused on non-major bleeding, which showed a relatively high incidence (Table 4). Postoperative heparin bridging and oral dosing of antiplatelet drugs as compared with those without these factors was associated with a significantly higher event risk (HR, 2.644, P=0.037 and HR, 4.192, P=0.012, respectively). We found a significantly higher event risk with longer total duration of the procedure (HR, 1.007 (per hour), P=0.025).

Table 4. Multivariate Cox Regression Analysis for Non-Major Bleeding After CA for NVAF

	HR	95% CI	P value
AF classification [persistent]	0.813	0.312–1.936	0.650
AF classification [long-standing persistent]	0.140	0.001–1.036	0.056
AST	1.017	0.992–1.037	0.166
Oral dosing of antiplatelet drug	4.192	1.428–10.489	0.012
Post-treatment heparin bridging	2.644	1.061–7.209	0.037
Total treatment duration	1.007	1.001–1.013	0.025
Institutional volume ≥40*	0.528	0.174–1.465	0.224

Analysis set: 1,111 patients, excluding 7 patients with intraoperative occurrence of an event. *Institutional volume is number of patients registered in an institution during the study period. AF, atrial fibrillation; AST, aspartate aminotransferase; CI, confidence interval; HR, hazard ratio. Other abbreviations as in Table 1.

Pre-CA Withdrawal Period

The most common pre-CA withdrawal period was 24–28 h, which was seen in approximately half the patients (Table 5). No complications occurred in the 109 patients with a withdrawal period of 8–24 h.

Table 5. Event Incidence at 4-h Intervals From Final Oral Dosing of DOAC to CA Initiation

Drug withdrawal period (h)	<4	≥4	≥8	≥12	≥16	≥20	≥24	≥28	≥32	≥36	Total
		<8	<12	<16	<20	<24	<28	<32	<36
No. patients	22	20	1	37	35	36	534	334	47	52	1,118
Thromboembolism and all types of bleeding	1	1	0	0	0	0	13	13	3	3	34
%	4.5	5.0	0.0	0.0	0.0	0.0	2.4	3.9	6.4	5.8	3.0

CA, catheter ablation; DOAC, Direct oral anticoagulant.

The optimal thresholds of withdrawal period for complications yielded by ROC analysis are presented in Table S3. Although some of the AUCs were not at a high enough level to predict the outcome events, the upper threshold of the withdrawal period was considered to be 26–30 h in this cohort. We assumed that the upper and lower thresholds were 28 and 8 h, respectively. The lower threshold represents the boundary between occurrence and non-occurrence of complications (Table 5). We excluded patients with postoperative heparin bridging and oral dosing of antiplatelet drugs in this analysis, because these were considered to have an effect on the occurrence of complications. The group of patients with a pre-CA withdrawal period <8 h showed a significantly higher incidence of major bleeding (18.18%) than the group of patients with a period of 8–28 h (Table 6). Major bleeding in the group of patients with a withdrawal period <8 h was associated with pericardial effusion and hematoma at the puncture site (1 patient each).

Table 6. Comparison of Event Incidence Between NVAF Patients With Pre-CA Withdrawal Period <8 h, ≥8 h and <28 h and ≥28 h

	1	2	3	Log-rank test P value
	<8 h (n=11)	≥8 h and <28 h (n=435)	≥28 h (n=212)
	n (%)	n (%)	n (%)	1 vs. 2	2 vs. 3
Composite event of thromboembolism and major bleeding	2 (18.18)	0 (0.00)	4 (1.89)	<0.001†	0.008†
Thromboembolism	0 (0.00)	0 (0.00)	1 (0.47)	−	0.148
Major bleeding	2 (18.18)	0 (0.00)	3 (1.42)	<0.001	0.026†
Non-major bleeding	0 (0.00)	4 (0.92)	3 (1.42)	1.000†	1.000†

Analysis set: 658 patients, excluding those with postoperative heparin bridging and oral dosing of antiplatelet drugs. ^†Adjusted P value (Bonferroni correction). Abbreviations as in Table 1.

The group with a pre-CA withdrawal period ≥28 h showed a significantly higher incidence of major bleeding (1.42%) than the group with a period ≥8 and <28 h (Table 6). In the group with a withdrawal period ≥28 h, the thromboembolism in 1 patient was a TIA and the major bleeding in 3 patients was pericardial effusion. In the group with a withdrawal period of 8–28 h, neither thromboembolism nor major bleeding occurred and the incidence of non-major bleeding was low (0.92%).

Time to Post-CA Resumption of Rivaroxaban

The time from the end of CA (sheath removal) to rivaroxaban resumption was 12–24 h in the majority of patients (57.2% (640/1,118)) (Table 7). To examine the association between the time to resumption and complication occurrence, we used 12 h as the threshold. In this analysis as well, we excluded patients with postoperative heparin bridging or oral dosing of antiplatelet drugs; we also excluded those who did not resume rivaroxaban. The incidence of non-major bleeding was significantly higher for a time to postoperative resumption ≥12 h (Table 8).

Table 7. Event Incidence at 4-h Intervals From the End of CA to Rivaroxaban Resumption in NVAF Patients

Time to resumption (h)	<4	≥4	≥8	≥12	≥16	≥20	≥24	≥28	≥32	≥36	Not applicable	Subtotal
		<8	<12	<16	<20	<24	<28	<32	<36
No. patients	197	199	45	193	262	185	11	2	3	16	5	1,118
Thromboembolism and all types of bleeding	1	2	0	8	12	4	0	0	1	4	2	34
%	0.5	1.0	0.0	4.1	4.6	2.2	0.0	0.0	33.3	25.0	40.0	3.0

Abbreviations as in Table 1.

Table 8. Comparison of Event Incidence Based on the Time From End of CA to Rivaroxaban Resumption (<12 h vs. ≥12 h) in Patients With NVAF

	<12 h (n=372)	≥12 h (n=279)	Log-rank test P value
	n (%)	n (%)
Composite events of thromboembolism and major bleeding	1 (0.27)	1 (0.36)	0.835
Thromboembolism	1 (0.27)	0 (0.00)	0.388
Major bleeding	0 (0.00)	1 (0.36)	0.247
Non-major bleeding	1 (0.27)	5 (1.79)	0.045

Analysis set: 651 patients, excluding those with postoperative heparin bridging, use of antiplatelet drugs, intraoperative occurrence of events, and non-resumption of rivaroxaban. Abbreviations as in Table 1.

Discussion

This study demonstrated a significantly higher incidence of non-major bleeding in the group with postoperative heparin bridging than in those without it, but showed no difference in the incidence of thromboembolism and major bleeding for the groups with and without heparin bridging. Also, both a pre-CA withdrawal period of 8–28 h and resumption of the drug within 12 h of CA were associated with a lower incidence of bleeding complications.

Heparin bridging has been reported to be associated with risks of bleeding and adverse events in anticoagulation management using warfarin.¹³ Although heparin bridging is considered necessary when anticoagulant drugs are interrupted for a prolonged period in patients at high risk of thrombosis,¹⁴ the European Heart Rhythm Association (EHRA) Practical Guide states that bridging is not necessary in DOAC-treated patients despite the limited clinical evidence evaluating the benefit and risk of the heparin bridge strategy, because the predictable waning of the anticoagulation effect theoretically allows properly timed short-term cessation and re-initiation of DOAC therapy before and after surgery.¹⁵

Among recent studies of anticoagulation management using rivaroxaban, some used uninterrupted treatment with a withdrawal period of 24 h^5,6,16,17 and others used interrupted treatment with a withdrawal period of >24 h.^18–20 The incidence of complications is reportedly 4.9–17.1% for uninterrupted treatment and 2.7–6.1% for interrupted treatment. Heparin bridging was not conducted in those studies. The Dresden NOAC registry, which enrolled the patients undergoing invasive procedures with perioperative anticoagulation management with DOACs, demonstrated heparin bridging as an independent predictor of major bleeding (odds ratio 5.0, 95% CI 1.2–20.4, P=0.023).¹⁰ However, the Dresden NOAC registry did not mention the risk of complication in AF ablation. Because CA for AF is thought to be a high-risk procedure for both bleeding and thromboembolism,²¹ specific consideration is required for this procedure. In the present study, based on a prospective multicenter registry for AF ablation, approximately half the patients underwent heparin bridging, and thus an evaluation of the risk of the heparin bridge strategy was possible. In this study, the complications of thromboembolism were quite rare, even in those without the heparin bridge strategy. On the other hand, heparin bridging increased the risk of non-major bleeding (4.03% vs. 0.87%; P=0.001). Interestingly, preoperative heparin bridging was not associated with an increase in complications (Table 2), but post-treatment heparin bridging was an independent predictor of non-major bleeding in the multivariate Cox regression model (Table 4). Many of the hemorrhagic complications recorded in this study were non-major bleeding events related to the puncture site, and heparin bridging seems to have increased these complications. Therefore, a post-treatment heparin bridge strategy should be avoided in AF ablation patients taking rivaroxaban in the periprocedural period.

The optimal interruption period of DOACs in the periprocedural period is another point of interest. With regard to the influence of both the withdrawal period and the time to resumption on the incidence of complications in this cohort, the group of patients with a preoperative withdrawal period of 8–28 h showed a lower incidence of major bleeding (Tables 5,6). In the group of patients with a pre-CA withdrawal <8 h and ≥28 h, major bleeding, including pericardial effusion in 4 patients and hematoma at the puncture site in 1, occurred, which possibly represented an increase in the number of procedure-associated complications. On the other hand, no difference specific to the duration of the pre-CA withdrawal period was observed for thromboembolism incidence. Therefore, according to these results and considering bleeding risk, rivaroxaban, a DOAC taken once daily, should be taken the day preceding the procedure and should not be taken before treatment on the day of the procedure. The EHRA Position Paper²² suggests that the final dose of rivaroxaban should generally be administered 24–36 h before CA but that the interruption period for DOACs can be safely shortened. Time to post-CA resumption <12 h was associated with a lower incidence of non-major bleeding in patients, excluding those with postoperative heparin bridging, use of antiplatelet drugs, intraoperative events, and non-resumption of rivaroxaban (Tables 7,8). Although we could not clarify the mechanism of the decrease in complications in this study, postoperative resumption of this drug after hemostasis on the day of procedure seems to be a safe option. The patients who were taking rivaroxaban in the morning until the day before treatment and receiving it after treatment on the day of ablation comprised approximately 40% of study population. The results of this study supported that such a perioperative regimen of DOAC prescription was safe and valid.

Study Limitations

First, because there were few episodes of thromboembolism and major bleeding, the statistical results should be examined carefully. Our study population consisted of AF patients with relatively low CHADS₂ score (mean±standard deviation 0.9±1.0) and heparin bridging might be effective in patients with a high risk of thrombosis. Second, this study was conducted in a clinical setting and the protocol did not define the dosage regimen. As a result, the patients who received the drug orally in the morning accounted for 85.4%. In addition, the approved dose of rivaroxaban in Japan is 15 mg. The clinical setting was specific to Japan, so our results should be understood in that context. Finally, ACT ≥350 s at sheath removal was associated with a significantly higher incidence of major bleeding. This may be primarily caused by the procedure. It remains to be elucidated whether neutralization of heparin with protamine is effective in reducing the major bleeding events in this specific group of patients. Further investigation is required.

Conclusions

In the perioperative CA anticoagulation management using rivaroxaban in patients with NVAF, postoperative heparin bridging possibly increases the risk of non-major bleeding. A pre-CA withdrawal period of 8–28 h and a time to post-CA resumption <12 h was associated with a lower incidence of bleeding complications.

Acknowledgment

This study was financially supported by Bayer Yakuhin, Ltd., Japan.

Disclosures

K.H. has received speaker honoraria from Bayer, Daiichi-Sankyo, Biotronik Japan, WIN International; K.K. from Boehringer-Ingelheim, Bayer, Daiichi-Sankyo, Bristol-Myers, and Pfizer; K.I. from Boehringer-Ingelheim, Bayer, Daiichi-Sankyo, and Bristol-Myers; M.K. from Johnson & Johnson; and K.O. from Boehringer-Ingelheim, Bayer, Daiichi-Sankyo, Bristol-Myers, Pfizer, and Johnson & Johnson. K.A., Y.M., and E.T. have nothing to disclose.

Supplementary Files

Supplementary File 1

Table S1. Clinical characteristics of NVAF patients treated with and without postoperative heparin bridging (after sheath removal)

Table S2. ACT values and heparin bridging in patients with NVAF

Table S3. Thresholds of drug withdrawal periods related to complications in ROC analysis

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-17-0509

References

• 1.   Hohnloser SH, Camm AJ. Safety and efficacy of dabigatran etexilate during catheter ablation of atrial fibrillation: A meta-analysis of the literature. Europace 2013; 15: 1407–1411.
• 2.   Phan K, Wang N, Pison L, Kumar N, Hitos K, Thomas SP. Rivaroxaban versus warfarin or dabigatran in patients undergoing catheter ablation for atrial fibrillation: A meta-analysis. Int J Cardiol 2015; 185: 209–213.
• 3.   Santarpia G, De Rosa S, Polimeni A, Giampa S, Micieli M, Curcio A, et al. Efficacy and safety of non-vitamin K antagonist oral anticoagulants versus vitamin K antagonist oral anticoagulants in patients undergoing radiofrequency catheter ablation of atrial fibrillation: A meta-analysis. PLoS One 2015; 10: e0126512.
• 4.   Piccini JP, Stevens SR, Lokhnygina Y, Patel MR, Halperin JL, Singer DE, et al. Outcomes after cardioversion and atrial fibrillation ablation in patients treated with rivaroxaban and warfarin in the ROCKET AF trial. J Am Coll Cardiol 2013; 61: 1998–2006.
• 5.   Lakkireddy D, Reddy YM, Di Biase L, Vallakati A, Mansour MC, Santangeli P, et al. Feasibility and safety of uninterrupted rivaroxaban for periprocedural anticoagulation in patients undergoing radiofrequency ablation for atrial fibrillation: Results from a multicenter prospective registry. J Am Coll Cardiol 2014; 63: 982–988.
• 6.   Cappato R, Marchlinski FE, Hohnloser SH, Naccarelli GV, Xiang J, Wilber DJ, et al. Uninterrupted rivaroxaban vs. uninterrupted vitamin K antagonists for catheter ablation in non-valvular atrial fibrillation. Eur Heart J 2015; 36: 1805–1811.
• 7.   Wazni OM, Beheiry S, Fahmy T, Barrett C, Hao S, Patel D, et al. Atrial fibrillation ablation in patients with therapeutic international normalized ratio: Comparison of strategies of anticoagulation management in the periprocedural period. Circulation 2007; 116: 2531–2534.
• 8.   Siegal D, Yudin J, Kaatz S, Douketis JD, Lim W, Spyropoulos AC. Periprocedural heparin bridging in patients receiving vitamin K antagonists: Systematic review and meta-analysis of bleeding and thromboembolic rates. Circulation 2012; 126: 1630–1639.
• 9.   Kim TH, Kim JY, Mun HS, Lee HY, Roh YH, Uhm JS, et al. Heparin bridging in warfarin anticoagulation therapy initiation could increase bleeding in non-valvular atrial fibrillation patients: A multicenter propensity-matched analysis. J Thromb Haemost 2015; 13: 182–190.
• 10.   Beyer-Westendorf J, Gelbricht V, Forster K, Ebertz F, Kohler C, Werth S, et al. Peri-interventional management of novel oral anticoagulants in daily care: Results from the prospective Dresden NOAC registry. Eur Heart J 2014; 35: 1888–1896.
• 11.   Okumura K, Aonuma K, Kumagai K, Hirao K, Inoue K, Kimura M, et al. Efficacy and safety of rivaroxaban and warfarin in the perioperative period of catheter ablation for atrial fibrillation: Outcome analysis from a prospective multicenter registry study in Japan. Circ J 2016; 80: 2295–2301.
• 12.   Firth D. Bias reduction of maximum likelihood estimates. Biometrika 1993; 80: 27–38.
• 13.   Steinberg BA, Peterson ED, Kim S, Thomas L, Gersh BJ, Fonarow GC, et al. Use and outcomes associated with bridging during anticoagulation interruptions in patients with atrial fibrillation: Findings from the Outcomes Registry for Better Informed Treatment of Atrial Fibrillation (ORBIT-AF). Circulation 2015; 131: 488–494.
• 14.   Ferrandis R, Castillo J, de Andres J, Gomar C, Gomez-Luque A, Hidalgo F, et al. The perioperative management of new direct oral anticoagulants: A question without answers. Thromb Haemost 2013; 110: 515–522.
• 15.   Heidbuchel H, Verhamme P, Alings M, Antz M, Hacke W, Oldgren J, et al. European Heart Rhythm Association Practical Guide on the use of new oral anticoagulants in patients with non-valvular atrial fibrillation. Europace 2013; 15: 625–651.
• 16.   Dillier R, Ammar S, Hessling G, Kaess B, Pavaci H, Buiatti A, et al. Safety of continuous periprocedural rivaroxaban for patients undergoing left atrial catheter ablation procedures. Circ Arrhythm Electrophysiol 2014; 7: 576–582.
• 17.   Nagao T, Inden Y, Yanagisawa S, Kato H, Ishikawa S, Okumura S, et al. Differences in activated clotting time among uninterrupted anticoagulants during the periprocedural period of atrial fibrillation ablation. Heart Rhythm 2015; 12: 1972–1978.
• 18.   Providencia R, Marijon E, Albenque JP, Combes S, Combes N, Jourda F, et al. Rivaroxaban and dabigatran in patients undergoing catheter ablation of atrial fibrillation. Europace 2014; 16: 1137–1144.
• 19.   Winkle RA, Mead RH, Engel G, Kong MH, Patrawala RA. Peri-procedural interrupted oral anticoagulation for atrial fibrillation ablation: Comparison of aspirin, warfarin, dabigatran, and rivaroxaban. Europace 2014; 16: 1443–1449.
• 20.   Stepanyan G, Badhwar N, Lee RJ, Marcus GM, Lee BK, Tseng ZH, et al. Safety of new oral anticoagulants for patients undergoing atrial fibrillation ablation. J Interv Card Electrophysiol 2014; 40: 33–38.
• 21.   Heidbuchel H, Verhamme P, Alings M, Antz M, Diener HC, Hacke W, et al. Updated European Heart Rhythm Association Practical Guide on the use of non-vitamin K antagonist anticoagulants in patients with non-valvular atrial fibrillation. Europace 2015; 17: 1467–1507.
• 22.   Sticherling C, Marin F, Birnie D, Boriani G, Calkins H, Dan GA, et al. Antithrombotic management in patients undergoing electrophysiological procedures: A European Heart Rhythm Association (EHRA) position document endorsed by the ESC Working Group Thrombosis, Heart Rhythm Society (HRS), and Asia Pacific Heart Rhythm Society (APHRS). Europace 2015; 17: 1197–1214.