Periprocedural Management of New Oral Anticoagulants in Atrial Fibrillation Ablation

Abstract

Background:

Patients who undergo catheter ablation for atrial fibrillation (AF) are at increased risk of developing thromboembolic and bleeding complications periprocedurally. Many patients are now on newer oral anticoagulants (NOACs), but data regarding their safety and efficacy during AF ablation are limited.

Methods and Results:

This article reviews the literature in PubMed from 1998 to 2014 and includes clinical trials and meta-analysis that analyzed the safety and efficacy of NOACs during AF catheter ablation. Dabigatran seems to be as effective and safe as warfarin, although most data are from single-center studies, with small samples and very low overall bleeding and thromboembolic complications. Periprocedural anticoagulation protocols also vary greatly between studies. Some recent meta-analysis has shown that warfarin could still be a safer and more effective alternative. There are fewer studies with rivaroxaban in AF ablation, and there have been no meta-analysis yet comparing rivaroxaban to warfarin or dabigatran. There seems to be no significant differences in safety or efficacy of rivaroxaban compared to warfarin. Interestingly, there are no available data for apixaban in AF ablation yet.

Discussion:

There are no consensus guidelines regarding the use of NOACs during AF ablation. Dabigatran and rivaroxaban seem as safe and effective as warfarin, although larger studies with standardized protocols are needed, as available studies may be underpowered to detect small differences in bleeding and thromboembolic rates.

Introduction

Patients with atrial fibrillation (AF) undergoing catheter ablation are at increased risk of thromboembolic events during, immediately postprocedure, and weeks to months after the procedure. ^1,2 Despite transesophageal echocardiography (TEE) and periprocedural anticoagulation, thromboembolic complications range between 0.5% and 2.8% and depends on the type of AF, duration of AF, left atrial (LA) size, and patient’s stroke risk profile based on CHADS₂ or CHA₂DS₂VASc score. ^{3 –6} Historically, patients undergoing catheter ablation for AF had warfarin discontinued 3 to 5 days prior to the procedure and were bridged with heparin or a low-molecular-weight heparin after the procedure until the patients were therapeutically anticoagulated with warfarin. ^5,7 However, studies have also demonstrated that use of uninterrupted anticoagulation with warfarin during AF catheter ablation was associated with similar bleeding complications and less periprocedural thromboembolic events. ^{8 –12} With introduction and increasing use of the newer oral anticoagulants (NOACs) that offer a better and more predictable pharmacokinetic profile, studies have demonstrated similar efficacy as warfarin in thromboprophylaxis in patients with nonvalvular AF having lower bleeding rates (especially hemorrhagic stroke). ^{13 –15} However, there are a limited data regarding the safety and efficacy of periprocedural use of NOACs during catheter ablation of AF. It is also not clear whether these agents still hold the same safety advantages compared to warfarin in this context. The aim of this article is to review the existing data regarding the safety and efficacy of use of NOACs during catheter ablation of AF, including a review of the current guidelines about their use.

Anticoagulation Strategy During and Following AF Ablation

Catheter ablation is a thrombogenic procedure and increases the risk of thromboembolic events (defined as transient ischemic attack, stroke, or any systemic arterial thromboembolism) even in patients who are initially deemed to be at a low thromboembolic risk. Several mechanisms have been proposed—(1) thrombus formation on the catheter or within the sheath following transeptal puncture to access the LA. ^{3,16 –18} (2) Endothelium damage secondary to catheter ablation, which may act as a nidus for thrombus formation and (3) “stunning” of the atrial tissue. ¹⁹ Anticoagulation is thus crucial in order to minimize the thromboembolic risks inherent to the procedure. The 2012 Heart Rhythm Society/European Heart Rhythm Association/European Cardiac Arrhythmia Society (HRS/EHRA/ECAS) expert task force recommends 3 weeks of systemic anticoagulation at a therapeutic level prior to catheter ablation in patients with AF for more than 48 hours, or for unknown duration of AF; and, if not the case, a TEE is advised to screen for thrombus. ²⁰

Optimal safe level of anticoagulation is essential throughout the procedure, as thrombi can form almost immediately after crossing the septum on the transseptal sheath and the electrode catheter and heparinization significantly decreases this risk. ^{3,7,17,21 –23} A heparin loading dose should be administered immediately after the vascular access, prior to transseptal puncture, followed by a standard heparin infusion adjusted to maintain an activated clotting time (ACT) of at least 300 to 400 seconds. ²⁰ Although no data exist to guide the frequency with which ACT levels should be monitored, writing group recommends that ACT levels should be monitored every 10- to 15-minute intervals until therapeutic anticoagulation is achieved and subsequently 15- to 30-minute intervals for the duration of procedure. It is also recommended to infuse heparinized saline continuously through the transseptal sheaths to further reduce the risk of thrombi formation. ¹⁷ The risk of embolization of the thrombus formed on a sheath may be reduced by withdrawal of transseptal sheath into the right atrium once a catheter is positioned in the LA. Once all catheters are removed from the LA, the heparin infusion is discontinued. The sheaths are removed from the groin once the ACT is less than 200 to 250 seconds. ²⁴

Although anticoagulation during AF catheter ablation minimizes the periprocedural thromboembolic risk, anticoagulation is also held accountable for most of the bleeding complications associated with the procedure, which include hemopericardium, cardiac tamponade, and bleeding complications at the vascular access site. ^4,12,25 It is therefore imperative to achieve an optimal level of anticoagulation that minimizes the bleeding risk while also protecting against thromboembolic complications. Classically, anticoagulation with warfarin would be discontinued 3 to 5 days prior to the ablation, and the patients would be bridged with heparin or enoxaparin prior to and following the ablation procedure. Although this strategy was widely accepted, it was observed that this approach was associated with a higher bleeding risk (especially at the vascular access site) and later resulted in trend of performing AF catheter ablation on uninterrupted warfarin with less thromboembolic complications and similar or less bleeding complications. ^{9,26 –28} Kok et al in 2002 estimated the incidence of thromboembolic complications to be up to 5%, which may be secondary to interruption of oral anticoagulation prior to the procedure, insufficient intraprocedural anticoagulation, or due to the use of nonirrigated catheters. ²⁹ Over time and with better anticoagulation strategies during AF ablation, the risk of thromboembolic events has declined. In 2006, Oral et al estimated the incidence of thromboembolic events to be 1.1% with a strategy of interrupted warfarin with bridging therapy ⁵ when compared to Hussein et al who demonstrated 0.09% thromboembolic events with a strategy of uninterrupted anticoagulation with warfarin in patients undergoing ablation for AF. ¹⁰ The approach to anticoagulate with uninterrupted warfarin is gaining wider acceptance and is now endorsed in the 2012 HRS/EHRA/ECAS expert consensus recommendations. ²⁰ According to 2012 HRS/EHRA/ECAS expert consensus recommendations, if warfarin was interrupted before ablation, it can be reinitiated within 4 to 6 hours after all sheaths (both venous and arterial) are removed, and heparin or enoxaparin should be used as a bridging therapy to achieve a target INR of 2 to 3. Alternatively, if warfarin was uninterrupted, use of heparin or enoxaparin can be avoided and warfarin is continued to maintain INR of 2 to 3.

The 2012 HRS/EHRA/ECAS and the 2012 ESC guidelines also recommend to continue with systemic anticoagulation for all patients for at least 2 months following AF catheter ablation as early recurrences of AF are not uncommon and can be asymptomatic. ^20,30 Although studies have shown that low-risk patients (CHADS₂ score 0-1) can be discharged on aspirin alone, ³¹ it is not currently recommended. Likewise, it is recommended that the decision to anticoagulate beyond first 2 months postablation should be made based on the patient’s individual risk factors for stroke and not on the presence or type of AF. Long-term anticoagulation is recommended indefinitely in patients with high risk of stroke as estimated by CHADS₂ or CHA₂DS₂VASc score and especially in patients who are ≥75 years of age or have a prior history of transient ischemic attack or stroke. ³⁰ This recommendation is based on the observation that recurrences of AF are common both early and late following AF ablation, often may be asymptomatic AF postablation, and lack of any randomized controlled trial assessing the safety of stopping anticoagulation in this patient population. Also, the patients who are at increased risk of stroke and in whom the decision to discontinue anticoagulation is being considered, such patients should undergo additional continuous electrocardiogram monitoring for asymptomatic AF/atrial flutter ³⁰ (Table 1).

Table 1.

Anticoagulation Strategies Pre-, During, and Postatrial Fibrillation Ablation.^a

Preablation Prior to undergoing an atrial fibrillation ablation, transesophageal echocardiogram should be performed to screen for thrombus in all patients with atrial fibrillation more than 48 hours in duration or for unknown duration if adequate systemic anticoagulation has not been maintained for at least 3 weeks prior to the ablation procedure. The presence of left atrial thrombus is a contraindication of atrial fibrillation ablation.

During ablation A heparin-loading dose should be administered immediately after the vascular access, prior to transseptal puncture, followed by a standard heparin infusion adjusted to maintain an activated clotting time (ACT) of at least 300-400 seconds. Activated clotting time should be monitored every 10-15 minute interval until therapeutic anticoagulation is achieved and subsequently 15-30 minutes intervals for the duration of procedure. Heparinized saline should be infused continuously through the transseptal sheaths to reduce the risk of thrombi formation. Systemic anticoagulation with warfarin does not alter the need for intravenous heparin to maintain a therapeutic ACT during atrial fibrillation ablation.

Postablation In patients whom warfarin was interrupted before ablation, it can be reinitiated within 4 to 6 after all sheaths are removed and heparin or enoxaparin should be used as a bridging therapy to achieve a target INR 2-3. Alternatively, if warfarin was uninterrupted, use of heparin or enoxaparin can be avoided and warfarin is continued to maintain INR of 2-3. Systemic anticoagulation with warfarin or a direct thrombin inhibitors or factor Xa inhibitor is recommended for at least 2 months following an AF ablation procedure. Decision to anticoagulate beyond first 2 months postablation should be made based on the patient’s individual risk factors for stroke and not on the presence or type of AF. Long-term anticoagulation is recommended indefinitely in patients with high risk of stroke as estimated by CHADS2 or CHA2DS2VASc score and especially in patients who are 75 years of age or older or have a prior history of transient ischemic attack or stroke.

Abbreviation: AF, atrial fibrillation.

^aAdapted from 2012 HRS/EHR/ECAS Expert Consensus Statement on Catheter and Surgical Ablation of Atrial Fibrillation.²⁰

Dabigatran

Dabigatran etexilate (Pradaxa) is one of the new orally active direct thrombin inhibitors approved by the Food and Drug Administration (FDA) in October 2010 for prevention of stroke in patients with nonvalvular AF, and in April 2014, it was also approved for the treatment and reduction in the risk of recurrence of deep venous thromboembolism (DVT) and pulmonary embolism (PE). Dabigatran has a bioavailability of 6%, time to peak levels of 3 hours, and a half-life of 12 to 17 hours (in healthy volunteers with normal renal function), with 80% renal excretion. ¹³ Its half-life allows for twice daily dosing, and since the maximum anticoagulation effects are achieved within 2 to 3 hours of administration, bridging therapy is not necessary. There is no specific way to reverse its anticoagulant effects. Due to its predictable pharmacokinetics and fewer drug interactions, dabigatran has become an attractive alternative to warfarin for prevention of thromboembolic events in patients with nonvalvular AF. In the RE-LY trial, ¹³ a noninferiority, open-label, prospective randomized trial comparing 2 different doses of dabigatran (110 mg b.i.d or 150 mg b.i.d) to warfarin, 150 mg b.i.d was associated with lower rates of stroke and systemic embolism (1.69% a year with warfarin, compared to 1.11% with dabigatran 150 mg twice a day, relative risk [RR] 0.66, 95% confidence interval [CI] 0.53 to 0.82; P < .001 for superiority) with no significant differences in the rates of major bleeding (3.36% a year with warfarin, compared to 3.11% a year with dabigatran 150 mg twice a day P = .31). Dabigatran of 110 mg twice a day was associated with similar rates of stroke and systemic embolism than warfarin (1.53% a year with dabigatran 110 mg twice a day, RR 0.91, 95% CI 0.74-1.11; P < .001 for noninferiority, P = .34 for superiority) but with lower rates of major bleeding—2.71% a year with dabigatran 110 mg twice a day, P = .003. The rate of hemorrhagic stroke was lower in the dabigatran groups compared to warfarin—0.38% per year in the warfarin group, as compared to 0.12% with 110 mg twice a day dabigatran (P < .001 for superiority) and 0.10% with 150 mg twice a day dabigatran (P < .001 for superiority). The FDA approved the 150 and 75 mg twice a day dose (for severe renal impairment), while the EMA has approved both the 110 and 150 mg twice a day doses.

As for the use of dabigatran during AF catheter ablation, it is unclear whether this drug still holds its advantages compared to warfarin as shown in the RE-LY trial. The existing data regarding its use in AF ablation are limited and most of it comes from observational and/or single-center studies. ^{32 –41} In 2012, Kaseno et al published a study where they evaluated 211 patients undergoing AF ablation, with 110 patients on dabigatran (110 mg twice a day) and 101 patients on warfarin; warfarin was uninterrupted during the ablation, while dabigatran was discontinued on the morning of the procedure and resumed the next morning. ³² No symptomatic thromboembolic complications were found in either group, and postprocedural brain magnetic resonance imaging in 60 patients (dabigatran group, n = 31; warfarin group, n = 29) demonstrated 1 silent cerebral infarction detected in each group; the total bleeding complications seemed to occur less often with dabigatran (4.5%) than with warfarin (12.9%; P < .05), suggesting that dabigatran could be a safer alternative to warfarin. Another study published by Bassiouny et al in 2013 ³⁶ prospectively reviewed data of 999 patients undergoing AF ablation at their center between December 2010 and July 2012, with 376 patients on dabigatran 150 mg twice a day and 623 patients on warfarin with therapeutic INR prior to the procedure; 1 or 2 doses of dabigatran were held prior to the procedure and restarted as soon as the patient was transferred to the floor; a cohort of 344 patients was generated in each group using propensity score matching. Of note, the mean ACT was significantly lower in the dabigatran group compared to those on warfarin, but no significant differences in thromboembolic or hemorrhagic complications were noted between the 2 groups. These results are in contrast to the findings reported by Lakkireddy et al a multicenter, prospective, observational study from 8 high-volume electrophysiology laboratories between January 2010 and July 2011. ³³ They included 145 patients who underwent catheter ablation of AF receiving dabigatran and an equal number of patients on warfarin matched by age, sex, and type of AF. Warfarin was uninterrupted during the ablation, while dabigatran was held on the morning of the procedure and was restarted within 3 hours after achieving hemostasis. No significant differences were found in the thromboembolic complications—3 (2.1%) in the dabigatran group compared to none in the warfarin group (P = .25)—but major bleeding complications were higher in the dabigatran group (6% vs 1%; P = .019) as well as the total bleeding rate (14% vs 6%; P = .031) and the composite of bleeding and thromboembolic complications (16% vs 6%; P = .009) compared to the warfarin group.

In recent years, meta-analyses addressing the question of the safety and efficacy of dabigatran compared to warfarin have been published. ^{42 –46} Bin Abdulhak et al demonstrated no significant differences in the incidence of thromboembolic (dabigatran 5 [0.4%], warfarin 2 [0.1%]; odds ratio [OR] 2.15, 95% CI 0.58-7.98; P = .54) or bleeding events (dabigatran 58 [5.4%], warfarin 103 [5.2%]; OR 0.92, 95% CI 0.55-1.45; P = .11), concluding that dabigatran seemed as safe and effective during AF catheter ablation as warfarin. Another 2 meta-analyses arrived to similar conclusions. ^43,45 In a meta-analysis of 3648 patients (2241 on warfarin and 1407 on dabigatran) Honhloser et al ⁴³ demonstrated 3 thromboembolic events with warfarin and 9 with dabigatran (OR 2.38, 95% CI 0.82-6.85; P = .11) and no significant difference in terms of major (OR 1.06, 95% CI 0.62-1.80; P = .85) or minor bleeding events. Similarly, Providencia et al included 4782 patients (1823 on dabigatran and 2959 on warfarin); the rates of thromboembolic events in dabigatran arm were 0.55% versus 1.35% with warfarin (RR 1.78, 95% CI 0.66-4.80; P = .26) and major bleeding complications were 1.48% versus 1.35% (RR 1.07, 95% CI 0.51-2.26; P = .86). ⁴⁵ But, contrary to the previous results, Steinberg et al ⁴⁴ and Sardar et al ⁴⁶ concluded that dabigatran might be associated with a higher frequency of adverse neurological events and higher risk of thromboembolic complications including stroke and transient ischemic attack. Steinberg et al included 10 cohort studies with 1501 patients on dabigatran and 2356 patients on warfarin and demonstrated higher neurological events in the dabigatran group when compared to warfarin group (0.7%, n = 10 vs 0.2%, n = 4) during AF catheter ablation (absolute risk difference 0.0047, 95% CI 0.0007-0.0099), whereas major bleeding outcomes were not significantly different (34 [1.6%] in the dabigatran group vs 40 [1.7%] in the warfarin group, absolute risk difference 0.0010, 95% CI −0.0090-0.0076). In a meta-analysis of 5513 patients by Sardar et al, 14 strokes or transient ischemic attacks were reported in the dabigatran group (0.65%) compared to 4 (0.12%) in the warfarin group (Peto OR [POR] 3.94, 95% CI 1.54-10.08; P = .004, number needed to harm = 284 patients), concluding that dabigatran had significantly more ischemic complications. ⁴⁶ The risk of all thromboembolic complications was also higher in the dabigatran group (POR 2.81, 95% CI 1.23-6.45; P = .01), although no significant differences were observed for the risk of major bleeding, pericardial tamponade, or groin hematoma (33 events in the dabigatran group vs 53 events in the warfarin group, OR 0.99, 95% CI 0.55-1.78; P = .98). It must be noted, though, that the incidence of adverse events such as strokes was very low (18 strokes in a total of 5513 patients in Sardar et al) and with a wide CI (1.54-10.08). On that subject, Sardar et al commented that they could not exclude the possibility that the higher incidence of strokes in the dabigatran group could be explained by chance also indicated for study sequential analysis for a 150% increase in the POR. ⁴⁶ Other probable explanation could be due to patient noncompliance. It poses a problem for AF ablation procedure where inadequate anticoagulation from poor compliance could lead to a higher embolic complication during periprocedural period.

Despite multiple randomized controlled trials, extensive meta-analyses with mixed results, there is no consensus guidelines regarding the recommended strategy for management of NOACs during AF catheter ablation. However, the task force recommends warfarin or NOACs for all patients for at least 2 months postablation. ³⁰

Rivaroxaban

Rivaroxaban (Xarelto) is another of the new oral anticoagulants, but, unlike dabigatran, which targets thrombin; rivaroxaban pertains to the group of the direct factor Xa inhibitors. The FDA approved it on July 2011 for pharmacological prophylaxis of DVT in patients undergoing knee or hip replacement, on November 2011, it was also approved for stroke prophylaxis in patients with nonvalvular AF, and, finally, in 2012, its approval was extended for management of DVT and PE. It has 60% to 80% oral bioavailability, a maximum effect achieved 3 hours after ingestion and a half-life of 5 to 13 hours in healthy volunteers. Two-thirds of the drug undergoes hepatic metabolism and one third is eliminated unchanged in urine. After its onset of action, factor Xa activity does not return to normal for up to 24 hours, therefore administration of rivaroxaban can be done once daily. Similar to dabigatran, there is no specific antidote for reversal of its anticoagulant effects. Its predictable pharmacokinetics obviates the need for monitoring and its rapid onset of action avoids the need for bridging therapy. The clinical trial that leads to approval of rivaroxaban in patients with AF was the ROCKET-AF trial, ¹⁴ a multicenter, randomized, double-blind, double-dummy, event-driven trial conducted at 1178 participating centers in 45 different countries and including 14 264 patients with nonvalvular AF who were at moderate to high risk of stroke. Rivaroxaban showed to be noninferior to warfarin in the prevention of ischemic stroke and systemic embolism (incidence of stroke of 2.1% per year with rivaroxaban, compared to 2.4% per year with warfarin; hazard ratio 0.88; 95% CI 0.74-1.03; P < .001 for noninferiority). Of note, there was a significant decrease in hemorrhagic strokes and intracranial hemorrhage in the rivaroxaban group (0.5% vs 0.7%, P = .02).

Unfortunately, the available data regarding the use of rivaroxaban in AF catheter ablation are lesser than that of dabigatran. There are only a few studies on the subject samples are usually small, studies are mostly single center, and there have been no meta-analysis done to this date. ^{47 –51} Eitel et al ⁴⁷ conducted a prospective observational study with 259 patients (from July 2010 to June 2012) after AF catheter ablation who were anticoagulated with either dabigatran 110 mg twice a day (38% of patients), dabigatran 150 mg twice a day (56% of patients), or rivaroxaban 20 mg daily (6% of patients); dabigatran was held the morning of the day before the procedure, rivaroxaban was held the evening before the ablation, and both drugs were resumed after hemostasis was achieved postprocedure; 54 (21%) patients were already on new oral anticoagulants prior to the procedure. During a mean follow-up time of 311 days, no stroke, systemic embolism, or major bleeding was noted in either group; periprocedurally, there were 4 events of major bleeding and thromboembolism, but none occurred in patients previously taking dabigatran or rivaroxaban. Another study looked at elevations in the d-dimer as a sign of hypercoagulability after AF catheter ablation in patients randomized to receive either dabigatran 110 mg twice a day (n = 30) or rivaroxaban 15 mg once daily periprocedurally (n = 30). ⁴⁹ With no initial differences in the baseline d-dimer level, the rivaroxaban group had significantly increased d-dimer levels following the ablation (mean ± standard deviation 0.62 ± 0.16-1.09 ± 0.38 µg/mL vs from 0.59 ± 0.08-0.75 ± 0.17 mg/mL in the dabigatran group; P < .0001). Providencia et al in a single-center study with 556 patients undergoing AF ablation from October 2012 to September 2013 (192 patients on warfarin, 188 on rivaroxaban, and 176 on dabigatran) found no differences between these NOACs in safety of efficacy when compared to warfarin in a 30-day follow-up period: Overall the rate of events was low, with 1.3% thromboembolic events (warfarin group 2.1%, rivaroxaban group 1.1%, and dabigatran group 0.6%; P = .410), 2.3% major bleeding events (warfarin 4.2%, rivaroxaban 1.6%, and dabigatran 1.1; P = .112), and minor bleeding in 1.4% (warfarin 2.1%, rivaroxaban 1.6%, and dabigatran 0.6%; P = .464). ⁵¹ In a multicenter, observational, prospective study of 642 patients by Lakkireddy et al comparing the use of uninterrupted rivaroxaban versus warfarin anticoagulation in patients undergoing AF ablation, no significant differences in the number of thromboembolic events (one in each group, 0.3%; P = 1.0), major bleeding (5 [1.6%] vs 7 [1.9%] P = .772) or minor bleeding events (16 [5.0%] vs 19 [5.9%]; P = .602) were observed. Of note, patients in the warfarin group had a slightly higher HAS-BLED score than the rivaroxaban group (1.70 ± 1.0 vs 1.47 ± 0.9; P = .032).

Pitfalls in Periprocedural Anticoagulation During AF Ablation

Nowadays, the new oral anticoagulants—thrombin inhibitors and direct factor Xa inhibitors—are an attractive alternative for prevention of stroke and systemic embolism in patients with nonvalvular AF. Their predictable pharmacokinetics and better interaction profile compared to warfarin make them a preferred option. In the context of AF catheter ablation, some of their advantages include a rapid onset of action, which obviates the need for bridging therapy when administered after hemostasis is achieved. These drugs also eliminate the concern with warfarin about sub- or supratherapeutic INR prior to the procedure, which often leads to procedure cancellations and delays.

There are, however, limited data on the safety and efficacy of these drugs during AF catheter ablation. There is a concern for increased risk of periprocedural thromboembolic events due to difficulties encountered in achieving target ACT levels when NOACs are used. ³⁶ However, this is based on a surrogate end point, and no large clinical trials have been done to address this concern. Absence of a reversal agent for NOACs should be kept in mind when a bleeding complication is anticipated or encountered during the procedure. Most of the available studies on NOACs in AF ablation are mostly on dabigatran and are often observational, single-center, and nonrandomized studies. ^{32,34 –36,39} There are also a few meta-analyses published in the literature, ^{42 –46} but they have yielded conflicting results. The largest and the most recent meta-analysis so far published in 2014 by Sardar et al concluded that periprocedural use of dabigatran for AF ablation was related to a higher risk of thromboembolic complications including stroke and transient ischemic attack. ⁴⁶ The above-mentioned findings by Sardar et al are not in agreement with the previous and somehow smaller meta-analyses, ^42,43,45 which didn’t find any differences in the safety or efficacy during AF ablation—except for the meta-analysis by Steinberg et al, ⁴⁴ which associated dabigatran with a higher frequency of periprocedural neurological events. Furthermore, there are very few studies of rivaroxaban in this context, with only 1 study comparing uninterrupted rivaroxaban with uninterrupted heparin during ablation. ⁵⁰ There are no data to this date on the use apixaban during AF ablation, and we may have to wait a few more years before we see studies with this medication.

Also, important differences have been noted in the study protocols of the different trials as well as in the characteristics of the patients, which makes the comparison and analysis of the obtained results difficult. Some studies compare uninterrupted dabigatran versus uninterrupted warfarin, ³² while others use interrupted warfarin with bridging therapy instead. ^52,53 Different studies had different protocols for holding dabigatran and rivaroxaban prior to the ablation as well as for restarting them. Some studies would hold dabigatran for 3 doses prior to the procedure and restart it the morning after the ablation, ³² while others would only hold the morning dose on the day of the ablation and then restart the medication 3 hours after hemostasis was achieved. ³³ Interestingly, the last approach was adopted by Lakkireddy et al where there was increased bleeding and thromboembolic risk with dabigatran. ³³ Other studies with a more interrupted approach did not find differences in the rate of bleeding or embolic complications. ^35,40,52,54 Nin et al used a lower INR target for the warfarin group than that accepted in western countries (the Japanese guidelines favor an INR between 1.6 and 2.5 for patients >70 years). ⁵⁴ Also, there have been differences noted in the required intraprocedural anticoagulation that could affect the rates of bleeding complications. Bassiouny et al noticed higher requirements of intraprocedural heparin to achieve a target ACT of 350 to 450 in the dabigatran group; the cumulative effect of dabigatran with higher intra-procedure heparin doses could account for the excessive bleeding risk that was observed in some studies. ³⁶ Equally important, the reported incidence of some of the adverse outcomes are very low, accounting for wide CIs; the studies may have been underpowered to detect differences in rates of rare adverse events (such as thromboembolic events) between groups.

Finally, it is worth mentioning that, although the new oral anticoagulants are gaining popularity in different fields of cardiology—including during AF catheter ablation, these new medications still do not have a role in patients with particular medical conditions such as severe renal or hepatic dysfunction or in patients with prosthetic heart valves. Additionally, there are still obstacles to overcome like the absence of an available antidote to reverse the anticoagulation effects in case of a life-threatening bleed, especially in the periprocedural setting. Prothrombin complex concentrate has been studied in healthy volunteers to reverse the anticoagulant action of rivaroxaban, ⁵⁵ but its role in the clinical practice has not yet been evaluated. Also, the need for dose adjustment in patients with renal dysfunction ⁵⁶ and the need to discontinue dabigatran earlier than usual prior to a procedure in such patients still need to be better characterized. ⁵⁷ Routine anticoagulation tests are altered in patients on direct thrombin inhibitors or factor Xa inhibitors, but their alterations have a poor correlation with the circulating concentrations of the medications, making the tests unreliable for monitoring. ^{58 –60} This may be challenging in patients going for AF catheter ablation who are not compliant with their medication and will be at increased risk of thromboembolic events with no objective way to assess their level of systemic anticoagulation prior to or after the procedure.

It is to be expected that, with availability of NOACs, more and more patients in the next coming years will be on one of these new oral anticoagulants prior and following AF ablation. Therefore, implementing guidelines and recommendations regarding the management of periablation anticoagulation with these agents is essential.

In conclusion, the role of the new oral anticoagulants during AF catheter ablation is still to be determined. There are no consensus guidelines regarding the use of NOACs during AF ablation. Dabigatran and rivaroxaban seem as safe and effective as warfarin although some studies suggest an increased risk of bleeding and embolic events.