Roof‐dependent atrial flutter with epicardial conduction pathway masked by left atrium posterior wall debulking ablation

Abstract Roof‐dependent atrial flutter (AFL) is a major tachyarrhythmia rotating in the left atrium (LA). Here, we describe a case of roof‐dependent AFL during atrial fibrillation ablation. LA posterior wall (LAPW) debulking ablation was performed before the induction. Atrial tachycardia (AT) was induced by burst pacing, and the 3D mappings showed a focal pattern from the LA inferior area. The post‐pacing interval from the roof and bottom line corresponded to the AT cycle length. The LAPW debulking ablation masked roof‐dependent AFL due to the lack of endocardium potentials in the LAPW. We report that roof‐dependent AFL connected by epicardium fibers.

Previous studies have reported several reentry patterns, including pulmonary vein (PV) gap reentry, scar-related reentry, Marshall bundle reentry, and bi-atrial tachycardia, associated with the existing substrate or previous ablation procedure (Hayashi et al., 2016;Kitamura et al., 2020;Ogawa et al., 2018;Satomi et al., 2008;Vlachos et al., 2019). Roof-dependent AFL is a major reentrant tachycardia rotating in the left atrium (LA), and detailed observation using 3D mappings and electrophysiological pacing maneuvers are crucial to identify the reentry circuit of AFL (Casado et al., 2018).
Left atrium posterior wall (LAPW) isolation during AF ablation is a common procedure. The feasibility, safety, and efficacy were demonstrated in previous studies (Thiyagarajah et al., 2019).
Here, we report a case where roof-dependent AFL was masked by LAPW debulking ablation due to the lack of endocardium potentials in the LAPW. We also provide evidence that ethanol infusion into the vein of Marshall (VOM) effectively ablated epicardium fibers.

| C A S E REP ORT
A 68-year-old woman with frequent episodes of palpitation was diagnosed with persistent AF. Initial catheter ablation was performed as AF was unresponsive to antiarrhythmic drug. At the time of the outpatient consultation, 12-lead electrocardiogram revealed rapid AF with a heart rate of 160 bpm. A transthoracic echocardiogram revealed a low normal left ventricular ejection fraction of 54% and a left atrial diameter of 45 mm. The patient was administered edoxaban of 60 mg as oral anticoagulant drug and bisoprolol fumarate of 5 mg as antiarrhythmic drug. Before the ablation procedure, a single dose of edoxaban was skipped and bisoprolol fumarate was discontinued for at least five half-lives. External cardioversion was performed by delivering 150 J in advance of catheter ablation to detect bradycardia.
The ablation procedure was performed under sinus rhythm.
The strategy was comprised of PV followed by linear ablations of the roof, bottom, mitral isthmus, and cava-tricuspid isthmus line using an irrigated tip ablation catheter (TactiCath™ contact force ablation catheter, Sensor Enabled™; Abbott) under general anes- (Agilis™ NxT steerable introducer; Abbott) to gain adequate contact force and stability. Chemical ablation using anhydrous ethanol (5 ml) infused into the VOM with posterior branch was performed before the PV isolation. Debulking ablation was used to complete the LAPW isolation because we had difficulty in LAPW isolation with roof and bottom-line ablation. LAPW isolation was confirmed through capture loss by high-output pacing (10 V and 1.0 ms).
At the end of the procedure, atrial tachycardia (AT) was induced by atrial burst pacing (pacing cycle length of 200 ms; Figure 1a,b).
The two mapping catheters placed in the coronary sinus and around the tricuspid annulus indicated the proximal-to-distal sequence and the tachycardia cycle length of 262 ms (Figure 1b Garcia et al. (2019) reported roof-dependent AFL cases that they failed to terminate despite LAPW isolation. An epicardial component of the septo-pulmonary bundle was also considered as an AFL conduction pathway. Miyazawa et al. (2019) also demonstrated that AFL detoured the epicardium at the anterior wall of the LA. These findings suggested that epicardial fibers could be a common conduction pathway of AFL rotating the LA. In this case, debulking ablation eliminated endocardial potentials of LAPW, and high-output pacing (10 V and 1.0 ms) could not capture the epicardial myocardium. Therefore, the scared area masked AFL propagation in the activation map. PPIs measured at the roof and bottom line indicated the possibility that tachycardia was roofdependent AFL, which was confirmed by termination following ablation.

| DISCUSS ION
We found that ethanol infusion into the VOM effectively ablated epicardium fibers. Jiang et al. (2019) demonstrated that epicardial potentials remained in after LAPW isolation in 33% (3/9 patients) despite endocardial capture loss. They suggested that endocardial ablation can be insufficient to make complete transmural lesion, but extra ablations along the original lines using a higher power for a longer duration could terminate AFLs. In this case, we were able to terminate AFL with two additional radiofrequency applications for the right-side bottom line. We also found that ethanol infusion into the VOM can effectively ablate the epicardium, especially in the leftside LAPW (Figure 3a,b). Thus, ethanol infusion might help terminate roof-dependent AFL detouring epicardial fiber.

| CON CLUS IONS
Left atrium posterior wall debulking ablation masked roofdependent AFL because of the lack of endocardium potentials.
Ethanol infusion into the VOM is possible to effectively ablate epicardium fibers.

AUTH O R CO NTR I B UTI O N
Masayuki ishimura and Kayo Yamamoto conceived the idea of the case report. Masayuki Ishimura also drafted the original manuscript.
Masashi Yamamoto, Toshiharu Himi, and Yoshio Kobayashi supervised the process of writing this case report. All authors reviewed the manuscript draft and revised it critically on intellectual content. All authors approved the final version of the manuscript to be published.

Yoshio Kobayashi has received remuneration for lectures from
Amgen Astellas BioPharma Co., Ltd., Bristol-Myers Squibb Co., and Boehringer Ingelheim and scholarships from Medtronic Japan Co.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.