Author + information
- aSilicon Valley Cardiology, Palo Alto Medical Foundation, Sutter Health, Palo Alto, California
- bSequoia Hospital, Redwood City, California
- ↵∗Address for correspondence:
Dr. Roger A. Winkle, Silicon Valley Cardiology, 1950 University Avenue, Suite 160, East Palo Alto, California 94303.
- atrial fibrillation
- activated clotting time
- complications of catheter ablation
- direct oral anticoagulant
- pulmonary vein isolation
Periprocedural anticoagulation for ablation of atrial fibrillation (AF) always represents a balance between risk of bleeding, primarily from vascular access and cardiac perforation, which is exacerbated by anticoagulation versus the risk of central nervous system events, which can be partially prevented by anticoagulation. In the early days of AF ablation, the only oral anticoagulants available were vitamin K antagonists, generally warfarin. Warfarin was interrupted and bridged with low-molecular-weight heparin and supplemented by intraprocedural intravenous heparin monitored by the activated clotting time (ACT). Intraprocedural heparin was allowed to wear off, and sheaths were pulled hours later when the ACT had fallen below a certain level. By then the local anesthetic had worn off, and it was difficult to compress the groins without patient discomfort, which often lead to vasovagal episodes and groin hematomas. Without the benefit of randomized trials or meta-analyses, many physicians adopted changes such as reversing intraprocedural heparin with protamine and using uninterrupted warfarin. A subsequent meta-analysis suggested uninterrupted warfarin might nearly eliminate the risk of stroke (1). However, almost all small nonrandomized series using uninterrupted warfarin continued to show a risk of stroke of approximately 1% (2,3). Other studies have suggested that interrupted oral anticoagulation could be just as good or better than uninterrupted anticoagulation (4,5).
There are several things to keep in mind regarding undesired central nervous system events associated with AF ablation. Not all these events are due to thromboembolic clot from the left atrium, which could be expected to be minimized by anticoagulation. Some neurological events are due to unintended air emboli, watershed ischemia due to intracranial vascular disease related to procedural hypotension, or spontaneous intracranial hemorrhage. It is also important to remember that physicians can do things other than give more anticoagulation to prevent strokes associated with AF ablation. These alternatives include preflushing all left atrial sheaths with high-concentration heparin (6), infusing high-concentration heparin through all left atrial sheaths, minimizing the number of left atrial sheaths, reducing left atrial and procedural times, and performing a transesophageal echocardiogram in the electrophysiology laboratory immediately before the ablation procedure. Likewise, groin bleeding can be minimized by using micropuncture or ultrasonography for access and using only a single groin to do the ablation.
In virtually all series, whatever the anticoagulation strategy, the cerebral event rate is fortunately so low that it may never be possible to perform a proper randomized trial to resolve these important issues. Such a study would need thousands or tens of thousands of patients to be powered to show a difference in these infrequent events.
With the introduction of direct oral anticoagulants (DOACs), physicians had a whole new series of decisions to make about periprocedural anticoagulants. Not all published studies have provided clarity for the physician trying to make their own decisions. One early study comparing dabigatran to warfarin showed a striking increase in both bleeding and thromboembolic events with the use of dabigatran compared to warfarin (7). That study could have scared physicians so much that they might have stuck with warfarin. Undaunted, many physicians followed their own experience and judgment and continued to use dabigatran for periprocedural anticoagulation. Most subsequent studies and meta-analyses showed that dabigatran and, subsequently, other DOACs were safe and generally comparable to warfarin (8,9). DOACs have become the predominant oral anticoagulants used in the periablation period. DOACs were initially used in an interrupted manner due to concerns about lack of reversal agents, but recent reports have evaluated uninterrupted use of DOACs. One landmark randomized trial of uninterrupted dabigatran versus uninterrupted warfarin should have resolved most of the clinicians’ questions and made decision making easy (10). That study concluded dabigatran was associated with fewer bleeding complications than warfarin. However, as a practicing physician, when I read that study, I am struck by the very large number of pericardial tamponades seen in the warfarin group (1 in every 53 ablations), which seems difficult to explain and makes me wonder just how applicable that study may be to my center, where tamponade occurs only once every few hundred cases, regardless of the anticoagulant used. Furthermore, although the dabigatran therapy in that study was supposed to be uninterrupted, only 41.3% of patients received a final dose of dabigatran <4 h before the transseptal puncture and 36.6% from 4 to <8 h before, and 19.6% of subjects received their last dabigatran dose more than 8 h before the ablation. The major bleeding rate for those given dabigatran <8 h before the transseptal puncture was 2.0% and was 0% for those given dabigatran more than 8 h before. Thus, even a randomized trial designed well by highly regarded investigators does not always help resolve all of the issues plaguing the practicing physician. Physicians must still exercise their own judgment about the relative risks of warfarin versus those of dabigatran in their laboratories and whether to give the last dose of dabigatran at least 8 to 12 h before the ablation.
Although we are in an era of guideline-driven medicine, which often seems to force all clinicians to choose the same treatments, there may not be a one-size-fits-all approach to periprocedural anticoagulation. One study suggested that low-risk patients may take aspirin or not take any oral anticoagulation prior to their procedure and need be given an oral anticoagulant only post-procedure (4). Another study found that for low-risk patients, aspirin could be used as the only anticoagulant both pre and post ablation without occurrence of thromboembolic events (11). A physician may choose to take an approach in a low-risk patient undergoing a relatively short ablation for paroxysmal AF that is much different from that in a high-risk patient undergoing a more complex and lengthier left atrial ablation.
In this issue of JACC: Clinical Electrophysiology, Gorla et al. (12) provide us with excellent data to help physicians make clinical decisions in the form of a meta-analysis of 8,362 patients undergoing DOAC therapy before and after AF ablation. The study evaluates 3 pre-ablation anticoagulation strategies. One strategy is uninterrupted; the second strategy is mildly interrupted, where the last dose is given <12 h before the procedure; and the third strategy is interrupted, where the drug is held for at ≥12 h before the procedure. This study shows that all 3 anticoagulation strategies are similar in regard to major bleeding and thromboembolic events. As most physicians might predict, overall bleeding was lowest in the group of patients undergoing the interrupted strategy.
Results of this meta-analysis should be welcomed by clinicians performing AF ablations. The study suggests that whatever approach an individual physician or laboratory chooses, they are probably not gaining or losing anything compared to other physicians or laboratories who may use a slightly different technique of interrupting the DOACs. The guideline writers should be careful not be too dogmatic about recommendations when multiple approaches may be equivalent or nearly equivalent.
Each physician or laboratory needs to carefully monitor their own outcomes and complications. For low-volume operators, it may be difficult to make judgments about the best anticoagulation strategy in their cases, and that may be one reason to restrict AF ablation to high-volume centers. High-volume centers tracking their own bleeding and thromboembolic complications should be able to use whichever strategy seems best in their hands and in their patient population. If they are using uninterrupted DOACs and having bleeding issues, perhaps switching to a more interrupted strategy would be appropriate. Physicians may also want to use an interrupted approach in a patient with a CHA2DS2-VASC (Congestive heart failure, Hypertension, Age ≥75 years, Diabetes mellitus, Prior stroke, transient ischemic attack [TIA], or thromboembolism, Vascular disease, Age 65–74 years, Sex category [female]) of 0 or 1 and use an uninterrupted approach for patients with high CHA2DS2-VASC scores. At least now physicians can feel justified in choosing whatever DOAC anticoagulation interruption strategy they deem best for their laboratory and for each individual patient.
↵∗ Editorials published in JACC: Clinical Electrophysiology reflect the views of the authors and do not necessarily represent the views of JACC: Clinical Electrophysiology or the American College of Cardiology.
Dr. Winkle has reported that he has no relationships relevant to the contents of this paper to disclose.
The author attests he is in compliance with human studies committees and animal welfare regulations of the author’s institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Santangeli P.,
- Di Biase L.,
- Horton R.,
- et al.
- Lakkireddy D.,
- Reddy Y.M.,
- Di Biase L.,
- et al.
- Calkins H.,
- Willems St,
- Gerstenfeld E.P.,
- et al.
- Gorla R.,
- Dentale F.,
- Crippa M.,
- et al.