Ablation of drivers in patients with persistent atrial fibrillation refractory to pulmonary vein antral isolation: Target characteristics and 3-year outcomes

To the Editor: Pulmonary vein antral isolation (PVAI) is the cornerstone of catheter ablation techniques used in the treatment of patients with atrial fibrillation (AF). However, the outcomes of a PVAI-only strategy for persistent AF have proven suboptimal. The majority of patients with persistent AF may require further atrial ablation beyond PVAI. Ablation of AF drivers, a patient-tailored ablation strategy for persistent AF with a procedural endpoint of AF termination, has garnered increasing attention in recent years.[1] We aimed to determine the electrogram (EGM) characteristics of AF drivers that can predict AF termination. We prospectively enrolled 147 consecutive patients who underwent the index procedure of catheter ablation for persistent AF at Beijing Chaoyang Hospital from September 2017 to January 2021. The study was approved by the Ethics Committee of Beijing Chaoyang Hospital (2015-Ke-80) and written informed consent was obtained from all patients. After PVAI, intravenous low-dose (0.004 mg/kg) ibutilide was administered within 5 min in patients who experienced ongoing AF. If persistent AF was terminated within 30 min, no further atrial ablation was performed. Only patients whose persistent AF was refractory to PVAI combined with low-dose ibutilide administration underwent further atrial high-density mapping and ablation of AF drivers.[2] A three-dimensional electroanatomic mapping and navigation system (CARTO 3, Biosense Webster Inc., Diamond Bar, CA, USA) and a 20-pole, 5-spline mapping catheter (PentaRay, Biosense Webster Inc.) were used for atrial geometry reconstruction and AF driver mapping. After PVAI, a high-density remap of the left atrium was created using a PentaRay catheter. Atrial regions that presented with the following EGM characteristics were indicative of AF drivers: clusters of EGMs that displayed interelectrode time and space dispersion at a minimum of three adjacent bipoles such that activation spread >90% of the AF cycle length (AFCL). At each bipole in a dispersion area, one or more of the following fractionated or non-fractionated EGM morphologies were found: (1) continuous (>50 ms), low-voltage (<0.3 mV) fractionated EGMs; (2) bursts of fractionated EGMs; and (3) fast, non-fractionated EGMs (AFCL <120 ms) [Figure 1].Figure 1: Electrogram characteristics of definite AF drivers. (A) The PentaRay (Penta) (Biosense Webster Inc., Diamond Bar, CA, USA) catheter was on the anterior roof of the left atrium; long-duration potentials were recorded; (B) The PentaRay catheter covered a low-voltage area on the posteroinferior wall of the left atrium. The signal amplitude was less than 0.1 mV in a majority of splines. (C) The AFCL at the base of the right atrial appendage was 102 ms, which was significantly shorter than that in the coronary sinus (165 ms). AF: Atrial fibrillation; AFCL: AF cycle length.All tagged areas were ablated using a 3.5-mm irrigated-tip catheter (TheromoCool SmartTouchTM, Biosense Webster Inc.). The procedural endpoint was AF termination, which was defined as either conversion of AF to sinus rhythm or as organization of AF to atrial tachycardia during ablation. Persistent AF was terminated by direct current cardioversion if the arrhythmia continued after all AF drivers were ablated. We defined three types of drivers to reflect the consequences of driver ablation: definite driver: AF was converted to sinus rhythm or organized to stable atrial tachycardia when ablating the drivers; plausible driver: AF was converted to transient atrial tachycardia or AFCL was prolonged in the coronary sinus during the ablation of the drivers; and pseudo driver: the ablation of the drivers had no effect on the maintenance of AF or the prolongation of the AFCL in the coronary sinus. We compared the following EGM characteristics of the last definite drivers found in patients whose AF was terminated by ablation (the termination group) with those of the pseudo drivers found in patients whose AF was not terminated by ablation (the non-termination group): the presence of long-duration (>50 ms) EGMs, short AFCL (defined as AFCL at the target area that was shorter than the AFCL in the coronary sinus by at least 10 ms), and the presence of low-voltage (<0.3 mV) signals for each AF driver region. Of the 147 consecutive patients with persistent AF who underwent the index ablation, AF was terminated by PVAI alone or by PVAI combined with low-dose ibutilide in 34 patients. Those 34 patients were excluded from the final analysis. The 113 (76.9%) patients who underwent AF driver ablation following PVAI combined with low-dose ibutilide were eventually included in this study. The mean age of the participants was 63.8 ± 11.0 years. The mean duration of continuous AF was 14.7 ± 19.4 months. Intraprocedural acute AF termination was achieved in 83 (73.5%) patients. A total of 238 AF drivers were identified in the 113 patients. In summary, there were 168 drivers (mean: 2.02 per patient) in the termination group (n = 83), which included 83 definite drivers, 50 plausible drivers, and 35 pseudo drivers. The 83 definite drivers include 63 drivers that converted AF to stable sinus rhythm and 20 drivers that converted AF to stable atrial tachycardias. In the non-termination group (n = 30), we found 70 drivers (mean: 2.33 per patient), including 47 plausible drivers and 23 pseudo drivers. We compared the EGM characteristics of the definite drivers (n = 83), plausible drivers (n = 97), and pseudo drivers (n = 58) in all the patients. Long-duration EGMs, short AFCL, and low-voltage EGMs were present in 42.2%, 23.1%, and 34.7% of the 83 definite drivers, respectively; in 31.6%, 32.5%, and 36.0% of the 97 plausible drivers, respectively; these proportions were significantly higher than those associated with the 58 pseudo drivers (52.7%, 13.5%, and 33.8%, respectively; P = 0.018 for all). After a median follow-up of 973 (371–1601) days, the rate of freedom from atrial tachyarrhythmias was 71.1% (59/83) in the termination group, which was significantly higher than that of the cardioversion group (50.0%, 15/30, P <0.05). The main findings of this study demonstrated that: (1) the majority of cases of persistent AF refractory to PVAI combined with low-dose ibutilide can be terminated by further ablation of AF drivers. Importantly, intraprocedural AF termination resulted in better 3-year outcomes; (2) in addition to spatiotemporal dispersion, the EGM characteristics indicative of definite AF drivers include long duration, low amplitude, and shorter AFCL. Several approaches have been proposed as methods of mapping AF drivers.[3] These approaches have been associated with a wide range of findings in terms of AF driver characteristics, percentage of acute AF termination, and long-term outcomes. However, considering the chaotic nature of atrial activity during persistent AF and the limitations of different mapping technologies, some AF drivers identified by these cutting-edge tools may not be definite drivers. This may explain why ablation of some AF drivers can terminate AF or prolong the AFCL, whereas ablation of other AF drivers has almost no impact on AF maintenance. In this study, we performed AF driver mapping using high-density contact mapping combined with the administration of low-dose ibutilide. Theoretically, the strength of this mapping approach stems from the fact that AF drivers may be more easily identified by reducing the degree of atrial disorganization during AF without terminating the key AF drivers. We described this approach as a method of identifying responders to PVAI among patients with persistent AF.[2] We found that it was impossible for such a low dose (0.004 mg/kg) of ibutilide to terminate persistent AF when used as a sole strategy. However, this intervention facilitated AF termination after the completion of PVAI in 30% of patients. Importantly, although a PVAI-only strategy was adopted in those patients whose AF was terminated after the administration of low-dose ibutilide, their long-term outcomes were comparable with those of patients with paroxysmal AF who underwent only PVAI as well. Based on this finding, we hypothesize that low-dose (0.004 mg/kg) ibutilide may not terminate persistent AF maintained by AF drivers outside the pulmonary vein antrum, but could potentially facilitate AF driver mapping. In line with the findings from previous studies about AF driver ablation, we also found that the distribution of AF in the left atrium seems to favor the anterior roof (usually close to the left atrial appendage), the posteroinferior wall, and the anterior wall. This suggests that these regions should be highlighted during the mapping of AF drivers. When it comes to the EGM characteristics of AF drivers, Seitz et al[4] first reported the use of spatiotemporal dispersion to represent the EGM fingerprints of AF drivers. However, we consider that it is necessary to further rank the regions presenting with dispersed EGMs because such EGMs can also be recorded in some areas with complex atrial myocardial architecture. In this study, we analyzed the EGM characteristics of the 83 driver regions in which ablation directly terminated AF, which is the strongest identifier of definite AF drivers. We found that the following three characteristics may be associated with acute AF termination: long duration, low amplitude, and shorter AFCL. The mechanisms underlying these EGM characteristics are still unknown. The sites that presented with low amplitude and long-duration potentials are likely related to the local atrial substrate. Of the 83 definite drivers in the termination group, 47 (56.7%) were located at the border between healthy atrial tissue and low-voltage zones, which were defined as regions with a voltage <0.3 mV. However, we speculate that the sites that presented with shorter AFCLs may be related to an autonomous substrate of the atrium. To date, the majority of studies that explored AF driver ablation had follow-up periods of approximately 1 year.[5] This is the first study to report 3-year outcomes of this ablation strategy used in the management of persistent AF. Of note, the mean duration of continuous AF in the 113 patients with persistent AF was only 14.5 months. Moreover, intraprocedural AF termination was more easily achieved in patients with a short duration of persistent AF as compared with patients in whom sinus rhythm was restored by cardioversion (13.0 ± 18.2 months vs. 19.5 ± 21.8 months, P = 0.014). This indicates that this ablation strategy may be best suited for patients with early persistent AF. Funding This study was supported by grants from the Beijing Natural Science Foundation (No. 7202050) and the 1315 personal training plan (CYMY-2017-03).