As data continue to accumulate, we are learning that uninterrupted anticoagulation prior to electrophysiologic procedures (such as pacemaker placement or ablation for atrial fibrillation [AF]) appears safer than interruption of anticoagulation, ie, bridging with parenteral agents followed by resumption of oral anticoagulation postoperatively. The peaks and troughs of anticoagulation that are created with an interruption/resumption strategy can result in rebound hypercoagulable effects or periods without protection at all.
In April 2014, the COMPARE1 study, which was published in Circulation, prospectively randomized 1584 patients with AF and a CHADS2 score ≥1 who were going to receive radiofrequency catheter ablation to either uninterrupted warfarin (INR 2-3 for 3 to 4 weeks prior to ablation) or warfarin with periprocedural bridging using subcutaneous enoxaparin followed by intravenous heparin (with the process reversed after the procedure). The rates of stroke/TIA were significantly lower in those in whom warfarin was maintained (0.25%) compared with those in whom the warfarin was withdrawn (4.9%) (P<.001). Similarly, there was less minor bleeding (4.1% vs 22%, P<.001, respectively) and (statistically insignificant) less major bleeding (0.38% vs 0.76%, P=.31, respectively). At that time, the authors cautioned that these results with a vitamin K antagonist could not be extrapolated to the new oral anticoagulants (NOACs [rivaroxaban, dabigatran, apixaban]).
Pilot data presented at the Heart Rhythm Society Scientific Sessions 2015 by the same study group now suggest that results may indeed be similar when NOACs are being used. In a 2015 study known as the VENTURE-AF trial2 (simultaneously published in the European Heart Journal), 248 patients with AF were randomized to uninterrupted rivaroxaban 20 mg or uninterrupted warfarin prior to catheter ablation and for 4 weeks post-procedure. In this small study, there were no thromboembolic events in the rivaroxavan group and 2 events in the warfarin group. Only 1 major bleeding event occurred in the warfarin group. Note that this was a proof-of-principle study and larger studies are needed to confirm these results.
At this time, caution should be used with this NOAC strategy because unlike warfarin, there is no reversal agent available yet for the NOACs in the event of significant periprocedural bleeding. On the other hand, results from the ROCKET-AF trial do suggest that another potential strategy—switching from rivaroxaban to warfarin—can create a period of high-risk for thromboembolism. For patients who are receiving rivaroxaban, then, continuation without interruption may be the best strategy. These data make a compelling case for a larger trial of this strategy, particularly because the NOACs are in ever-wider use and because an antidote to these novel agents (ANNEXA-R) may be in the pipeline.
1. Di Biase L, Burkhardt D, Santangeli P, et al. Periprocedural stroke and bleeding complications in patients undergoing catheter ablation of atrial fibrillation with different anticoagulation management: Results from the "COMPARE" randomized trial. Circulation. 2014;129:2638-2644. doi: 10.1161/CIRCULATIONAHA.113.006426. Epub 2014 Apr 17.
2. Cappato R, Marchlinski FE, Hohnloser SE, et al. Uninterrupted rivaroxaban vs uninterrupted vitamin K antagonists for catheter ablation in non-valvular atrial fibrillation. Eur Heart J 2015; DOI:10.1093/eurheartj/ehv177.