Saving the Brain From Catheter Ablation of Atrial Fibrillation

The Role of Pre- and Peri-procedural Anticoagulation

Jared T. Bunch; Scott C. Woller


Eur Heart J. 2019;40(19):1538-1540. 

Atrial fibrillation (AF) ablation is an established therapy for symptomatic AF. As a non-pharmacological therapy for AF that in ~60–70% of patients provides a durable reduction in AF recurrence risk, there exists hope that AF-related co-morbidities may be reduced. Large observational trials have shown that AF ablation is associated with lower rates of stroke, dementia, heart failure, and mortality.[1,2] However in a meta-analysis of rigorous randomized control trials that compel high-quality anticoagulation, close patient follow-up, and rigorous assessment of arrhythmia recurrence analogous co-morbidity reductions have not been seen.[3] The dichotomy in outcomes between large long-term observational trials and randomized trials goes beyond selection bias to the systematic approach to a patient with AF and its most significant co-morbidity, stroke.

While stroke risk reduction efforts among AF patients undergoing ablation has historically emphasized the post-procedure period, including pre- and peri-procedural risk mitigation is central in any comprehensive approach to lower procedure-associated stroke risk and decreases the likelihood of cognitive decline and dementia years later. Although stroke risk during an AF ablation is relatively uncommon, event rates can be up to 2% and are augmented by baseline risk factors for stroke.[4,5]

Comprehensive peri-procedural stroke risk mitigation requires pre- and peri-procedural anticoagulation.[4] It is generally well accepted that peri-procedural anticoagulation with heparin and a goal activated clotting time (ACT) >300 s lowers the risk of thrombus formation, with more intense heparin therapy required in those patients with baseline spontaneous contrast in the left atrium.[4,6] The pragmatic application of anticoagulation to optimize net clinical benefit (lowest bleeding and thrombosis risk) is less certain.

In a collective analysis that highlighted the potential benefit of pre-procedural anticoagulation, a meta-analysis of nine studies showed that an AF ablation performed with a therapeutic international normalized ratio (INR) was associated with a significant decrease in risk of stroke [odds ratio (OR) 0.10; 95% confidence interval (CI) 0.05–0.23; P < 0.001] compared with discontinuing warfarin.[7] Even when discontinuing warfarin was brief for 2–3 days in a prospective randomized trial, warfarin discontinuation was a strong predictor of peri-procedural thrombo-embolism (OR 13; 95% CI 3.1–55.6; P < 0.001) and did not improve risk of bleeding.[8] In this analysis, discontinuation of anticoagulation was the strongest predictor of peri-procedural stroke. Although warfarin is not immediately reversible, administration of blood products can improve coagulation and potentially minimize risk in the setting of a significant bleed.

Direct oral anticoagulants (DOACs) have improved long-term outcomes compared with warfarin in patients with AF by lowering the risk of intracranial haemorrhage and maintaining or reducing the risk of embolic stroke. Six randomized controlled trials have been carried out comparing strategies of uninterrupted DOACs vs. warfarin anticoagulation for AF ablation.[9] Meta-analysing these six studies (that compared warfarin with either rivaroxaban, apixaban, or dabigatran) among 1903 patients revealed no difference between strategies with regard to risk of stroke or transient ischaemic attack (OR 1.0; 95% CI 0.23–4.40; P = 1.0), silent cerebral events (OR 1.09; 95% CI 0.67–1.75; P = 0.74), or bleeding (OR 1.01; 95% CI 0.78–1.31; P = 0.9).[9]

Broad adoption of uninterrupted DOAC therapies during AF ablation has been slow for a number of reasons. These randomized trials of DOAC therapy were performed in high-volume centres and the overall adverse event rates were very low. Procedural risks of both stroke and bleeding increase in lower volume centres and with less procedural experience.[10] Although antidotes for dabigatran (idarucizumab) and apixaban/rivaroxaban [coagulation factor Xa (recombinant), inactivated-zhzo—formerly known as andexanet alpha] are available, the preparation of coagulation factor Xa (recombinant), inactivated-zhzo causes delays in administration, and both agents are costly and are not broadly available worldwide. Finally, there remains uncertainty amongst cardiologists and physicians regarding the safety of urgent surgery or percutaneous intervention to treat uncontrolled peri-procedural bleeding with or without an antidote or administration of activated prothrombin complex concentrates. Within these reasons exists the need to consider additional study into short-term discontinuation of a DOAC therapies with relatively predictable metabolism kinetics to minimize risk of severe bleeding across the broad global landscape of AF ablation procedures. Furthermore, there are no randomized controlled trials that examine discontinuation of DOAC therapy vs. interrupted DOAC therapy to determine if the same elevated risk of peri-procedural stroke is observed with these anticoagulants as was seen with warfarin discontinuation.[8]

In this issue of the European Heart Journal, Yu and colleagues[11] present an exploratory prospective randomized trial of DOAC anticoagulation strategies for AF ablation that included 24 h skipped, same day skipped, and uninterrupted anticoagulation. The DOAC therapies were not consistent in the 326 study participants, but evenly distributed [dabigatran (n = 108), rivaroxaban (n = 104), and apixaban (n = 114)] and used for at least 3 weeks before the ablation. Dosing of these medications was adherent with the package insert recommendations in ~80% of patients. All ablation was performed with an open-irrigated tip catheter with the power settings 30–35 W. The ablation approach was selected by the individual operators. Heparin titration targeted an ACT of between 350 and 400 s. Patients with interrupted DOAC therapy, whether for 24 h or single dose skipped, required more heparin units to obtain the goal ACT. Importantly, one of five patients randomized to a 24-h DOAC hold received enoxaparin 1 mg/kg (if persistent AF was present).

Peri-procedural complications were rare in all groups. There were no strokes in any group. Of the three patients that experienced tamponade, two patients were randomized to the 24-h hold arm and one patient was randomized to uninterrupted anticoagulation. These patients were treated with percutaneous drainage and did not require surgery. In another two patients, an effusion was detected that did not require treatment, one in the uninterrupted group and another in the same day held DOAC group. Minor bleeding events including haematoma formation were not reported. With these very low event rates, there were no discernible differences in risks between DOAC therapies.

Any clinical conclusions that may be drawn from this exploratory study are highly limited by the small sample size and the low bleeding and stroke rates. Indeed, in experienced centres, the complications of peri-procedural major bleeding and cerebral thrombo-embolism are estimated to be 0.2–5% and 0.2%, respectively.[4] For perspective, the five-fold larger COMPARE trial randomized 1584 patients to either uninterrupted warfarin or continuous warfarin and reported a 0.6% and 1.9% rate of tamponade and 1.9% of stroke, respectively.[8] This study by Yu and colleagues is hypothesis generating. Before these observations may inform clinical practice, validating prospective evidence among AF patients undergoing ablation that receive a DOAC for stroke risk reduction adequately powered for outcomes of bleeding and stroke is required.

Yet even this small study is meaningfully additive to the extant literature. First, the event rates for both stroke and significant bleed were extremely low and suggest the safety and utility of DOAC therapies before AF ablation. Next, there was not a small signal or trend of adverse outcomes with temporarily holding DOAC therapies up to 24 h. In a registry analysis of 290 patients that compared warfarin and dabigatran peri-procedural anticoagulation, significant signals of both stroke and bleeding risks were observed in the dabigatran group which was attributed to no dabigatran administration on the day of the procedure.[12] Finally, these low events data were obtained in an Asian population in which expected stroke and bleeding rates are perhaps higher compared with other races among which the potential benefit of DOAC use may be greater.[13] Take home figure highlights the key insights of this study and strategic approaches to minimize risk of stroke and brain injury in patients pre-, peri-, and post-AF ablation.

Take Home Figure.

Stroke and brain injury reduction strategies by atrial fibrillation ablation.

The authors are to be congratulated on performing a prospective randomized trial in an area of uncertainty with significant pragmatic implications. Although safety and efficacy cannot be concluded in this exploratory analysis, the findings suggest that interrupted DOAC therapy may not impact harm and, in low-volume centres, and those without immediate access to DOAC antidotes or surgical therapies, this message is critical.