Author + information
- Received July 30, 2018
- Revision received January 15, 2019
- Accepted January 17, 2019
- Published online March 18, 2019.
- Hiro Yamasaki, MD∗ (, )
- Kazutaka Aonuma, MD,
- Yasutoshi Shinoda, MD,
- Yuki Komatsu, MD,
- Keita Masuda, MD,
- Naoaki Hashimoto, MD,
- Eikou Sai, MD,
- Fumi Yamagami, MD,
- Yuta Okabe, MD,
- Yasuaki Tsumagari, MD,
- Yuichi Hanaki, MD,
- Hiroaki Watanabe, MD,
- Takeshi Machino, MD,
- Kenji Kuroki, MD,
- Yukio Sekiguchi, MD,
- Akihiko Nogami, MD and
- Masaki Ieda, MD
- ↵∗Address for correspondence:
Dr. Hiro Yamasaki, Department of Cardiology, Faculty of Medicine, University of Tsukuba, 1-1-1 Tennodai, Tsukuba 305-8575, Japan.
Objectives This study sought to determine the feasibility of a novel simplified ablation protocol targeting only the pulmonary vein antrum using the radiofrequency hot-balloon catheter in patients with paroxysmal atrial fibrillation.
Background Radiofrequency hot-balloon (RHB) catheter has been recently introduced into clinical practice for pulmonary vein isolation (PVI). The authors hypothesized that a novel simplified ablation protocol targeting only the PV antrum with energy application for a longer time (single-shot technique) could be an alternative approach to achieve PVI, while avoiding unnecessary energy application at the PV ostium.
Methods A total of 61 consecutive paroxysmal atrial fibrillation patients (age 64.1 ± 10.9 years, 48 male) who underwent antrum RHB-PVI were enrolled. Energy applications were performed following the pre-specified protocol only targeting the PV antrum. If the PVI was not achieved after 2 energy applications using the RHB, a touch-up ablation was performed.
Results Of 241 PV, including 3 left common PV, 194 (80%) were isolated exclusively using the RHB. The target PVI average per group of 15 consecutive procedures improved from 75% (initial 15) to 89% (last 16) of patients. The injected volume was greatest in the right superior PV (13.1 ± 2.0 ml) and the smallest in the left inferior PV (10.8 ± 1.1 ml), and 23 PV (9.5%) required over 15 ml (estimated balloon diameter of 30 mm). Periprocedural complications were noted in 3 patients (4.9%), but phrenic nerve injury was not observed. Sinus rhythm maintenance at 12-month follow-up was achieved in 57 patients (93%).
Conclusions A novel simplified antrum RHB-PVI appears to be a feasible technique for the treatment of paroxysmal atrial fibrillation.
- antrum pulmonary vein isolation
- learning curve
- paroxysmal atrial fibrillation
- phrenic nerve injury
- radiofrequency hot-balloon ablation
Radiofrequency hot-balloon catheter (RHB) (SATAKE Hot-Balloon, Toray Industries, Inc., Tokyo, Japan) consists of a high compliant size-adjustable balloon (diameter 25 to 33 mm) that helps to achieve better contact with irregular-shaped pulmonary vein (PV) ostia and creates contiguous and transmural lesions (1). The feasibility of the pulmonary vein isolation (PVI) using the radiofrequency hot-balloon catheter (RHB-PVI) has been proven in a randomized study (2). During the clinical trial, the central balloon temperature and radiofrequency energy delivery time was determined empirically, and energy applications were routinely performed not only at the antrum, but also at the ostium and carina (3). As a result, acute complete PVI was achieved in 98% of the PV. However, phrenic nerve injury (PNI) and severe PV stenosis (PVS; >70%) were observed in total of 8.9% of the patients with a smaller injection volume (<10 ml). Especially, the incidence of severe PVS of 5.2% was significantly higher compared to that with the other balloon technologies (4,5). In addition, asymptomatic but moderate PVS (50% to 70%) was found in 22.4% of the patients (6). Energy application at the PV ostium increases the risk of collateral damage; thus, establishment of an alternative approach was strongly desired.
Based on the RHB principle, longer energy application creates a deeper lesion (1). We hypothesized that a longer energy application targeting the PV antrum with larger injection volume could be an alternative approach to achieve PVI, while avoiding unnecessary energy application at the PV ostium. Accordingly, we developed a novel simplified RHB-PVI protocol targeting only the PV antrum (single-shot technique). In this study, we aim to elucidate the feasibility of the simplified technique, its efficacy, and the safety profile.
This is a retrospective single-center study to evaluate the feasibility of the simplified ablation protocol targeting only the PV antrum. After the approval of the RHB system in April 2016, 108 patients underwent RHB-PVI at the Tsukuba University Hospital from April 1, 2016, to March 31, 2017. Initially, 10 patients were treated with an ablation protocol used during the clinical trial. Then, we began to adopt the simplified technique targeting only the PV antrum (single-shot technique). In this retrospective analysis, a total of 61 consecutive patients, with a history of paroxysmal atrial fibrillation were included. Patients with a history of persistent atrial fibrillation (n = 28), structural heart disease (n = 3), chronic renal failure on dialysis (n = 2), and follow-up period <6 months (n = 4) were excluded. Pre-procedural imaging was not mandatory and antrum RHB-PVI was performed irrespective of the PV variants. As a result, patients with left long common PV (n = 3) were included. All antiarrhythmic drugs were discontinued at least 5 half-lives before the procedure. At the beginning, only 1 surgeon (H.Y.) performed the procedure. From September 2016, 5 fellows of electrophysiology (Y.S., Y.K., K.M., N.H., E.S.) performed the ablations. In those cases, the most experienced doctor in the team (H.Y.) gave the final decision on the injected volume and the balloon position. All patients provided their written informed consent before the procedure. The study protocol was approved by the ethics committee of the Tsukuba University Hospital.
The procedure was performed under conscious sedation using dexmedetomidine and fentanyl. After groin punctures, 5,000 units of heparin were given. A steerable mapping catheter (Bee AT, Japan Lifeline Co., Ltd., Tokyo, Japan) was positioned within the coronary sinus. A 10-polar mapping catheter (Inquiry Steerable Diagnostic Catheter; Abbott, St. Paul, Minnesota) was positioned at the His-bundle area and a pig-tail catheter was placed at the noncoronary cusp as a landmark for the transseptal puncture (TSP).
Double TSP were performed using a radiofrequency needle (NRG RF Transseptal Needle, Baylis Medical, Montreal, Quebec) and 8-F long sheath (SL0; Abbott) under fluoroscopic guidance. After the initial TSP, a guidewire was advanced into the left superior pulmonary vein (LSPV) and the course of the guidewire was confirmed in the right anterior oblique projection. If the guidewire was advanced into the LSPV in a coaxial position, the second TSP was attempted slightly anterior to the initial TSP site. If the advanced first guidewire failed to achieve coaxial position to the LSPV, the second TSP was attempted slightly posterior to the initial TSP site to achieve better coaxial position (Figure 1). The bolus heparin was given immediately after the initial TSP followed by continuous heparin to achieve the target activated clotting time between 300 and 350 s.
A direct PV angiogram was performed to visualize the left atrium–PV junction location and PV branches. An oral airway was inserted to prevent deep respiration during the sedation, and an esophageal thermometer (Esophaster, Japan Lifeline Co.,Ltd.) was inserted for continuous esophageal temperature monitoring.
The 8-F long sheath crossing the atrial septum posteriorly was exchanged over a J-tip guidewire (Spring Guide Wire, Toray Medical Co., Ltd., Tokyo, Japan) for a 17-F deflectable guiding sheath (Treswaltz, Toray Industries, Inc.). The balloon was gradually inflated with target injection volume of ≥10 ml (estimated diameter of 25 mm) of contrast medium diluted 1:1 with normal saline in the PV. The curve of the deflectable guiding sheath was adjusted and the coaxial position of the RHB to the PV antrum was confirmed by the 2 different fluoroscopy angles (Figure 2). Along with balloon inflation, the balloon was gradually moved toward the antrum. Using the pre-acquired PV angiogram results as a reference, careful attention was paid to maintain a coaxial position with forward pressure in order to avoid “popping out” from the PV antrum. The largest injection volume with complete occlusion was confirmed with a selective PV angiogram and considered as the optimal balloon size. The energy application was delivered as follows: right superior pulmonary vein (RSPV) (70°C, 3.5 min), right inferior pulmonary vein (RIP) (70°C, 3 min), left inferior pulmonary vein (LIPV) (70°C, 2.0 to 2.5 min), and LSPV (70°C, 4 min) (Central Illustration). To prevent PNI, pacing was applied from an electrode in the superior vena cava (SVC) during energy applications in the right PV. After each energy application, a circular mapping catheter was placed in the PV. If the PV potential remained, a second energy application was considered. In such cases, the guidewire was inserted into a different PV branch (Online Figure 1) to enable better contact with the remaining conduction site. A touch-up ablation with an irrigated ablation catheter (FlexAbility; Abbott) was performed if the PV conduction remained after a maximum of 2 energy applications or reconduction was observed spontaneously. A voltage map using 3-dimensional mapping system was obtained to confirm the isolation area. Then, PV reconnection was tested using isoproterenol (1 μg/min) and adenosine triphosphate (20 mg) and touch-up ablation was performed if necessary. If the patient had a history or inducible atrial flutter, cavotricuspid isthmus ablation was performed. Isolation of the SVC was also performed at the physician’s discretion. Ablation time and total energy application only included energy applications using the RHB and the procedure time was calculated from the time of TSP to the time of protamine administration after ablation (2).
Esophageal temperature monitoring and endoscopy evaluation
To avoid esophageal injury, cooling water (0°C) was injected into the esophagus (initial injection with 10 ml; repeated with 5 ml of injection bolus) when the esophageal temperature exceeded 39°C (7). If the rapid temperature fall was not observed immediately after the cooling water injection, the guiding sheath was slightly turned counterclockwise while maintaining the forward pressure. This maneuver helped to provide space for the cooling water to pass through the esophagus behind the balloon. If the temperature increased over 39°C in every 30 s, it was considered to be frequent and the energy application was shortened from 2.5 min to 2 min during the LIPV ablation.
Endoscopic examination was scheduled as a part of the post-procedural assessment to evaluate the incidence of esophagus-related injury regardless of the symptom. Esophagus-related injury was defined as any injury resulting from energy applications, including esophageal erythema, necrotic ulcerations, atrio-esophageal fistulae, acute pyloric spasms, and gastric hypomotility (8). The endoscopy was performed and assessed by independent, experienced gastroenterologists blinded to the patients’ clinical data within 48 h after the index procedure.
No antiarrhythmic drugs were prescribed after the procedure. If the patient had an immediate recurrence within a blanking period, antiarrhythmic drugs were prescribed at the physician’s discretion and stopped at 3 months. Follow-up visits consisted of a clinical interview, electrocardiograms, and 24-h Holter monitoring at 3, 6, and 12 months. Recurrence was defined as any atrial tachyarrhythmia lasting longer than 30 s, and a 3-month blanking period was applied.
Continuous data are expressed as mean ± SD for normally distributed variables or as median (interquartile range [IQR]) for non-normally distributed variables. Acute success was defined by PVI achieved only using the RHB. The learning curve was analyzed in blocks of 15 patients (the last group included 16 patients) and procedure time was compared using the 1-way analysis of variance. A Kaplan-Meier curve was plotted for the time to first atrial arrhythmia recurrence if any. A probability value of p < 0.05 indicated statistical significance.
Patient and procedural characteristics
Patient and procedural characteristics are shown in Table 1. A total of 61 patients (48 male, 64.1 ± 10.9 years), underwent antrum RHB-PVI using the single-shot technique. The mean left atrial dimension and left atrial volume index was 36.9 ± 4.8 mm and 33.7 ± 12.6 ml/m2, respectively. Pre-procedural imaging was acquired using multidetector computed tomography in 38 patients (62%).
As shown in Table 2, energy applications per procedure were 5.2 ± 1.0. The injected volume was greatest in the RSPV (13.1 ± 2.0 ml) and smallest in the LIPV (10.8 ± 1.1 ml), and over 15 ml (estimated balloon diameter of 30 mm) was required during the RSPV ablations in 16 patients (26%), LSPV ablation in 3 (4.9%), and RIPV ablation in 1 (1.6%). In 2 patients with left common PV, successful isolation was achieved with injection volumes of 20 ml and 17 ml, respectively. Acute PVI was achieved using the RHB in 200 of 241 PV (83%) and 31 patients (51%). During the procedure, spontaneous reconductions were observed in 6 PV (LSPV; n = 2, RSPV; n = 1, RIPV; n = 3). As a result, 194 PV (80%) and 29 patients (49%) were isolated exclusively using the RHB. Touch-up ablations were required in 41%, 5%, 16%, and 10% of LSPV, LIPV, RSPV, and RIPV, respectively. Among those 47 PV requiring touch-up ablation, a single gap was observed in 42 (89%) and most of the sites (83%) were found at the anterior bottom part of the PV where stable contact of the balloon against the PV antrum was difficult during exhalation or coaxial position of the balloon to the PV was difficult to achieve. Among 159 PV of 40 patients (including 1 patient with a left common PV), administration of isoproterenol and adenosine provoked PV reconnection in 2 PV (1.3%) where touch-up ablation was required, whereas PV reconnection was not observed after successful antrum RHB-PVI. The average procedure time was 83.1 ± 13.6 min. In addition to the PVI, SVC isolation and right atrial isthmus ablation were performed in 5 patients (8.2%) and 12 patients (20%), respectively.
Analysis of the learning curve
The learning curve of the procedure was evaluated by comparing the procedural results in groups of 15 patients. The acute success rate per PV was significantly improved from 75% in the initial group to 89% in the last group (p < 0.05), and was apparent in LIPV and RSPV (Figure 3, Online Figure 2). Also, the procedure time reduced from 89 ± 16 to 74 ± 14 min (<0.05). The number of procedures performed by each physician were 41 (H.Y.), 10 (Y.S.), 5 (Y.K.), 3 (K.M.), and 1 each (N.H. and E.S.). Accordingly, new surgeons performed 42% of the procedure after 5 months.
In this series, there were 3 complications (1 pseudo-aneurysm, 2 cardiac tamponades) (Table 3). In 1 patient, cardiac tamponade developed shortly after inadvertent body movement and mechanical trauma from a catheter placed at the right ventricular apex was suspected. In another case, perforation of the left atrial appendage due to the guidewire was observed (9). In each patient, pericardiocentesis stabilized the hemodynamic condition. No patient experienced PNI and none required immediate termination of the energy application due to suspected PNI.
During the procedure, cooling saline was injected for a median of 6 times (IQR: 3.5 to 8.5) with median of 35 ml (IQR: 22.5 to 47.5). Among 61 patients, 54 (89%) underwent endoscopic evaluation. Asymptomatic esophageal erythema and gastric hypomotility were found in 3 (5.6%) and 4 patients (6.6%), respectively. There was no esophageal ulceration, atrio-esophageal fistula, or pyloric spasm (Table 3). In addition, there was no clinically relevant aspiration pneumonia related to the cooling saline injection.
Patients were followed-up for a median of 398 (IQR: 369 to 563) days and recurrence of any type of atrial arrhythmia was observed in 9 patients (14.8%). There was no post-procedural death. The Kaplan-Meier plot with event-free survival is presented in Figure 4. At 12 months, the drug-free rate of freedom from any atrial arrhythmia was 93%. Six patients underwent a repeat ablation at median of 464 (IQR: 386 to 517) days after the initial procedure. Among them, PV reconduction was observed in 6 of 24 PV (25%).
This study provided several new findings. First, a novel antrum RHB-PVI, with the single-shot technique, was found to be feasible and could be performed with a short learning curve. Among 241 PV, 194 (80%) were exclusively isolated using the RHB-PVI. During the study period, acute success rate per PV was significantly improved from 75% to 89% when comparing the initial group to the last group. Second, sinus rhythm was maintained without antiarrhythmic drugs in 93% of the patients at 12-month follow-up. Third, antrum PVI using the size-adjustable RHB successfully avoided PNI irrespective of the shape of the right PV ostium, which remains a significant concern in other balloon-based PVI (4,5).
Antrum RHB-PVI as a single-shot device
This is the first study to describe the detailed technique of the simplified antrum RHB-PVI and evaluate its feasibility for treatment of paroxysmal atrial fibrillation. By adjusting the injection volume while maintaining a coaxial position and forward pressure, RHB enabled the achievement of complete PV occlusion at the level of the PV antrum and successfully isolated PV by prolonging energy application. During the RHB randomized trial, energy applications were delivered at the ostium, antrum, and carina. The injection volume (average injection volume at the PV antrum and ostium) was greatest at the LSPV (10.2 ± 2.9 ml) and smallest at the RIPV (7.9 ± 1.8 ml). A smaller injection volume resulted in a high PV isolation rate (98%) (2) but increased the risk of severe PV stenosis and PNI. During this study, 9.5% of the PV including left common PV, required injected volume over 15 ml (estimated diameter of 30 mm). Although further study is required to establish the optimal energy setting, the unique feature of the size-adjustable RHB enables its use as a single-shot device to create a contiguous lesion at the level of PV antrum irrespective of the PV anatomy.
Learning curve of the antrum RHB-PVI
In this study, the result clearly showed a learning curve and acute PVI was achieved in 89%, exclusively using the RHB in the last group. Surprisingly, the success rate, as well as procedure time, was not affected by new surgeons. This finding was contrary to that of the cryoballoon. In the STOP-AF (Sustained Treatment of Paroxysmal Atrial Fibrillation) trial, the success and complication rates were affected by the number of procedures. In the study, doctors who had performed 12 to 23 procedures had a success rate of 90%, and for those who had performed only 1 or 2 cases, the success rate was 56% (10). We speculated that a deformation of the high-compliant balloon provided a visual understanding of the balloon position at the PV antrum. The visual feedback helped to adjust the injection volume and catheter manipulation to achieve better PV occlusion throughout energy applications; thereby new surgeons could achieve similar results with advice from the experienced doctor.
Chronic success rate after the antrum RHB-PVI
At the 12-month follow-up, sinus rhythm was maintained in 93% of the patients. The antrum RHB-PVI, with greater injection volume, suggested that the balloon was attached to a more proximal area of the PV antrum and an isolated wider area than those of the randomized study. Although touch-up ablations were required in 20% of the PV in this study, most of the PV needed a single energy application to complete the PVI. This implied that contiguous lesions were mostly achieved using the RHB at the PV antrum. The number of patients requiring a second ablation was limited, but durability of the PV isolation was 75% in those patients over 1 year after the procedure and was similar to the cryoballoon ablation (11).
Previous observational studies using the second-generation cryoballoon and laser balloon had demonstrated arrhythmia-free survival of 83.6% and 60.2% after 1 year (12,13), respectively. Though there were some differences in the follow-up protocol and additional ablation, it appeared that the result of the current study was at least comparable or might be better. We speculated that antrum RHB-PVI successfully avoiding PNI suggested a wider isolation area compared with that of other balloon technologies and contributed to a better clinical outcome.
Complications related to the antrum RHB-PVI
PNI is a serious complication during an energy application in the right-sided PV. The incidence of PNI after cryoballoon had been reported to be significantly higher than that after radiofrequency catheter ablation (14). In the Fire and Ice trial, PNI was still observed (2.7%) even if patients with PV diameter of >26 mm in the right side were excluded (8). Size-adjustable RHB was expected to be applicable in variable PV anatomies. Therefore, we have not considered any anatomical exclusion criteria. In this study, the injection volume was greatest (13.1 ± 2.0 ml) during the RSPV isolation and 26% of the RSPV required an injection volume of >15 ml, which successfully prevented PNI. This observation strongly suggested that the simplified antrum RHB-PVI protocol created a contiguous lesion in a more proximal area than the cryoballoon.
The esophageal-related injury is also another concern. In this study, asymptomatic esophageal erythema and gastric hypomotility were found in 7 patients (12.2%). A previous study had reported a higher incidence of esophageal-related injury (32.5%) after second-generation cryoballoon ablation (15). Cool saline injection and time adjustment in patients with frequent esophageal temperature rise effectively prevented esophageal-related injuries. In addition, counterclockwise catheter manipulation in case of frequent esophageal temperature rise may have avoided overheating of the posterior wall while maintaining good contact against the thicker anterior ridge.
The present study described a single-center retrospective experience with a small number of patients reporting the initial experience of antrum RHB-PVI. The follow-up duration was limited and protocol was not as strict and neither telemetry nor 7-day Holter electrocardiogram were used. Second, 5 patients received SVC isolation per the physician’s decision. Empirical SVC isolation had been shown to improve the sinus rhythm maintenance; thereby it may have contributed to greater long-term success (16). Third, the semiquantitative analysis of PVS was not performed systematically. Post-procedural multidetector computed tomography was performed in 40 patients, and severe PVS was observed in 2 patients (5%) (Table 3). Of note, there were no cases of moderate PVS. Both cases of severe PVS occurred at the beginning of the procedure, and RHB may have been pushed inside the PV ostium despite the balloon being inflated by at least 15 ml. Our study showed a significant reduction of the moderate PVS as compared to the randomized clinical study (6), but further study is needed to assess the impact of antrum RHB-PVI on the post-procedural severe PVS. Finally, we adopted a fixed ablation protocol. Animal studies had shown that higher surface temperature of the balloon and longer energy application time are associated with deeper lesions (1). Larger injection volume had a negative correlation with the surface balloon temperature; thus, it would decrease if the balloon was inflated with greater injection volume. Therefore, longer energy application may be required to create a transmural lesion if the balloon was inflated with greater injection volume. Further studies are necessary to determine the optimal ablation protocol needed to improve the acute success rate of RHB-PVI.
A novel simplified approach to achieve antrum RHB-PVI appeared to be feasible with a short learning curve. Antrum RHB-PVI with avoidance of PNI appears a feasible therapeutic option for the treatment of paroxysmal atrial fibrillation.
COMPETENCY IN MEDICAL KNOWLEDGE: Antrum PVI using the size-adjustable RHB catheter with avoidance of PNI is a feasible therapeutic option of paroxysmal atrial fibrillation.
TRANSLATIONAL OUTLOOK: Multicenter study with a large number of patients and new operators are needed to assess the feasibility, safety, and long-term success rate of this novel simplified approach.
Dr. Yamasaki has received a consultant fee from Toray Industries, Inc. Dr. Aonuma has received a grant from and belongs to the endowed department of Toray Industries, Inc. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- interquartile range
- left inferior pulmonary vein
- left superior pulmonary vein
- phrenic nerve injury
- pulmonary vein
- pulmonary vein isolation
- pulmonary vein stenosis
- radiofrequency hot-balloon
- right inferior pulmonary vein
- right superior pulmonary vein
- superior vena cava
- transseptal puncture
- Received July 30, 2018.
- Revision received January 15, 2019.
- Accepted January 17, 2019.
- 2019 American College of Cardiology Foundation
- Sohara H.,
- Ohe T.,
- Okumura K.,
- et al.
- Sohara H.
- Dukkipati S.R.,
- Cuoco F.,
- Kutinsky I.,
- et al.
- ↵Applications for Manufacturing/Marketing Approval for Medical Devices (in Japanese). PMDA (Pharmaceuticals and Medical Devices Agency). Available at: http://www.pmda.go.jp/medical_devices/2015/M20151118001/index.html. Accessed February 21 2019.
- Yamasaki H.,
- Yamagami F.,
- Machino T.,
- et al.
- Bunch T.
- Dukkipati S.R.,
- Kuck K.H.,
- Neuzil P.,
- et al.
- Buiatti A.,
- von Olshausen G.,
- Barthel P.,
- et al.
- Miyazaki S.,
- Nakamura H.,
- Taniguchi H.,
- et al.
- Li J.Y.,
- Jiang J.B.,
- Zhong G.Q.,
- Ke H.H.,
- He Y.