Author + information
- Received January 9, 2018
- Revision received July 3, 2018
- Accepted July 6, 2018
- Published online November 19, 2018.
- Yumei Xue, MDa,b,∗,
- Yang Liu, MDa,b,c,∗,
- Hongtao Liao, MDa,b,
- Xianzhang Zhan, MDa,b,
- Xianhong Fang, MDa,b,
- Hai Deng, MDa,b,
- Feng Wang, MDa,b,
- Wenxiang Huang, MDa,b,
- Yuanhong Liang, MDa,b,
- Wei Wei, MDa,b,
- Yingjie Huang, MDa,b,
- Zili Liao, MDa,b,
- Michael Shehata, MDc,
- Xunzhang Wang, MDc and
- Shulin Wu, MDa,b,∗ ()
- aGuangdong Cardiovascular Institute, Guangdong General Hospital, Guangdong Academy of Medical Sciences, Guangzhou, China
- bGuangdong Provincial Key Laboratory of Clinical Pharmacology, Guangdong General Hospital, Guangdong Academy of Medical Sciences, Guangzhou, China
- cHeart Institute, Cedars Sinai Medical Center, Los Angeles, California
- ↵∗Address for correspondence:
Dr. Shulin Wu, Guangdong Cardiovascular Institute, Guangdong General Hospital, Guangdong Academy of Medical Sciences, 106 Zhongshan Er Road, Guangzhou 510080, China.
Objectives This study aimed to evaluate the electrophysiological mechanisms of post-surgical atrial tachycardias (ATs) during mapping with an automated high-resolution mapping system (Rhythmia, Boston Scientific, Marlborough, Massachusetts).
Background Mapping and ablation of post-operative ATs following previous open-heart surgery is often challenging because the potential mechanisms remain incompletely understood.
Methods Fifty-one consecutive patients underwent mapping and ablation of post-surgical ATs.
Results A total of 64 ATs were identified, and the mechanism was macro re-entry in 58 of 63 (92.1%) ATs, focal in 4 ATs, localized micro re-entry in 1 AT, and undetermined in 1 AT. Of 11 patients who underwent surgical repair of congenital heart disease, 6 (54.5%) had peri-tricuspid re-entrant AT, 5 had either right atrial (RA) free-wall incisional ATs or figure-8 re-entrant ATs, with an isthmus between the tricuspid annulus and the RA free-wall incision or between the incisions, and none had left atrial (LA) or focal ATs. In 32 patients with valve replacement and 8 who underwent valvuloplasty, peri-tricuspid ATs were observed in 14 (43.4%) and 6 (75%) patients, RA free wall or septal incisions-related ATs were seen in 7 and 2 patients, and LA macro re-entrant ATs were observed in 12 patients and 1 patient, respectively. A macro pseudo re-entry pattern was identified in 8 of 51 patients (15.7%). All these activations could be easily excluded by manually moving the window of interest, except in 2 cases with a figure-8 re-entrant configuration.
Conclusions RA macro re-entrant ATs predominate, irrespective of the types of initial surgical procedures, but LA ATs occur more frequently in patients with valve replacement. Pseudo re-entry atrial activation is common and easily recognized by adjusting the mapping window.
Atrial tachycardias (ATs) can develop after open-heart surgery regardless of whether patients have undergone concomitant surgical ablation of atrial fibrillation (AF). The potential mechanism of such arrhythmias remains incompletely understood, but is usually considered to be a re-entry that involves pre-existing scar tissue and anatomical obstacles. Mapping and ablation of these arrhythmias is well known to be challenging because of underlying cardiac disease and iatrogenic lesion sets with a resultant complex atrial substrate.
An automated high-resolution mapping system (Rhythmia, Boston Scientific, Marlborough, Massachusetts) has recently become clinically available. The first preclinical animal study, reported by Nakagawa et al. (1), demonstrated that the new mapping system could accurately and rapidly identify surgical incisions or epicardial radiofrequency lesions, as well as the presence or absence of a gap in linear lesions in a canine right atrial (RA) model. Subsequent clinical studies also confirmed its advantages for mapping and ablation of ATs in patients with previous AF ablation, surgical mitral valve repair, or failed AT ablation (2–5), which suggested its potential capability to guide complex AT ablation. The objective of this study was to determine the mechanisms of complex ATs using the Rhythmia mapping system in patients with a history of open-heart surgery.
Between October 2015 and July 2017, we prospectively enrolled 51 consecutive patients who were referred to our hospital for electrophysiological study of ATs that were refractory to antiarrhythmic drug therapy. All patients had a history of open-heart surgery for congenital anomaly repair (n = 11) or valve replacement (25 with mechanical mitral valves, 7 with bioprosthetic mitral valves) and valvuloplasty (n = 8). Among them, 18 patients who underwent valve procedures received concomitant surgical radiofrequency ablation of AF, and 6 patients with mechanical valves underwent CARTO-guided transvenous catheter (Biosense Webster, Irvine, California) ablation for ATs or AF, including typical atrial flutter in 3, left atrial (LA) flutter in 1, and AF in 2. The detailed data of baseline demographics are shown in Table 1. This investigation was approved by the ethics committee of Guangdong General Hospital and performed in accordance with the Declaration of Helsinki.
Previous open-heart surgical procedure
Procedural data regarding the initial cardiac surgery was obtained for all patients. Extracorporeal circulation was established through central cannulation from both caval veins to the aorta. Three different incisions were applied to the atria to approach the intracardiac structures (Online Figure 1). RA free wall atriotomy was only needed in the 11 patients who had surgical repair of congenital heart disease. In some other patients, both RA and transseptal LA incisions were used to perform surgery at the left side of the heart. An alternative access to the LA through direct left atriotomy in the posterior interatrial groove was performed at the discretion of the surgeons. Overall, the 3 types of incisions were used in 49, 30, and 14 patients, respectively.
Concomitant surgical radiofrequency ablation was used if patients had a pre-operative diagnosis of AF. A bi-atrial lesion set or left-sided only ablation was performed at the discretion of the surgeon. In brief, 2 separate encircling lesions were made around the left and right pulmonary veins (PVs) in all ablated patients. Additional lesion sets included a posterior box lesion in the LA, a connecting line from the right inferior PV to the mitral annulus, closure or excision of the LA appendage, dissection of the ligament of Marshall, a cavo-tricuspid isthmus line, and extension of the right-sided atrial incision to the tricuspid annulus and to the superior and inferior vena cavae.
During the procedure, patients were studied under conscious sedation with continuous intravenous fentanyl. All procedures were performed using the Rhythmia electroanatomic mapping system in conjunction with an Orion mini-basket catheter (IntellaMap Orion, Boston Scientific, Marlborough, Massachusetts) (Online Figure 2). The system and catheter were described previously in detail (6). The atria were mapped using the Orion catheter, which was inserted from the right femoral vein and advanced into the RA via an 8.5-F long sheath (SL1, St. Jude Medical, Minneapolis, Minnesota) and into the LA via a transseptal access.
Cardiac beats were accepted in the maps based on 5 criteria, which consisted of cycle length stability, timing stability between 2 reference electrodes (coronary sinus electrograms), respiration gating, catheter motion, and tracking quality. All selected criteria had to be met for inclusion. Electrographic timing was based on the timing difference between the maximum negative dV/dt of the unipolar electrogram or the maximum amplitude on the bipolar electrogram and the first reference electrode. The voltage map was based on the difference between the maximum and minimum peaks of the signal. The regions with a voltage <0.03 mV were defined as scar, whereas those with a voltage between 0.03 and 0.5 mV were considered low voltage areas. Bipolar electrograms were filtered at 30 and 300 Hz. Unipolar electrograms were filtered at 1 and 300 Hz. The V (for all maps) and P (for paced maps only) blanking windows were applied to the reference trigger channels to prevent inadvertent triggering of the V and the pacing artifact. The V blanking window represented the time period during ventricular depolarization, which was typically characterized by the surface QRS morphology. The P blanking window represented the time period during the pacing artifact of a paced rhythm.
The outermost electrodes of the mini-basket catheter were used for creating the surface geometry of the cardiac chambers based on the electrode locations of accepted beats. The activation maps were generated by including all electrograms recorded within 2 to 5 mm from the surface geometry. The projection distance was set by the operators. Generally, the system was programmed to select electrograms located within 2 mm of the surface for newly created maps and for 5 mm for remapping, or by creating a validation map (vMap), which was designed to rapidly confirm linear block and check for gaps of ablation lines. The mapping window (usual 100% tachycardia cycle length) could be manually shifted to focus on the area of interest at any time during or after the completion of the map.
In most cases, activation mapping was performed initially, and then multisite entrainment maneuvers were used for confirming re-entrant circuits.
ATs were deemed macro re-entry if an activation map exhibited a re-entry around a large central obstacle, that is, either an anatomical structure or scar tissue created surgically or by an underlying substrate. Focal ATs were defined by a centrifugal activation pattern from a discrete source. Micro re-entrant ATs were diagnosed if long-duration fractionated electrograms were recorded at a focal site of successful ablation. Pseudo re-entry was arbitrarily defined as an activation map with an “early-meets-late” pattern but which did not exhibit a true macro re-entrant mechanism. Pseudo re-entry was identified by manually moving the window of interest rightward or leftward without a change of the window duration, which resulted in the disappearance of the early-meets-late activation pattern. The window of interest could be moved by manually dragging the mapping window on the screen and using the mouse to exclude re-entry. Atrial entrainment could be performed further in such a scenario, if needed.
After completing ablation of targeted sites and tachycardia termination to sinus rhythm, programmed atrial stimulation with burst pacing in the presence or absence of isoproterenol was performed to induce other ATs. The endpoint of the procedure was noninducibility of any sustained ATs. Intravenous administration of ibutilide or atrial overdrive pacing was used if the tachycardia did not revert to sinus rhythm during the procedure. Atrial arrhythmia inducibility was not attempted in this scenario. Peri-procedural anticoagulation was administered following our standard protocol (7).
Patients were monitored by 24-h Holter recording at 4-week, 3-month, and 6-month follow-up periods. A 12-lead electrocardiogram was obtained at every post-procedural visit. Additional electrocardiograms were obtained if arrhythmia-related symptoms occurred. Antiarrhythmic drugs were not routinely administered after the ablation procedure, except in cases of AF. Success of maintaining sinus rhythm was defined as freedom from documented ATs lasting ≥30 s without a blanking period.
A total of 112 maps in 51 patients were acquired with the mini-basket mapping catheter, including 80 RA maps and 32 LA maps. Of the 112 maps, 89 were obtained during ATs, 21 during atrial pacing to check the completeness of linear block, and 2 maps were obtained during sinus rhythm. The number of maps per patient averaged 2.2 ± 1.2 (range 1 to 5). These maps took 2.3 to 34.5 min (median: 11.5 min) to complete and contained 123 to 4,708 (median: 921) accepted beats, with 1,804 to 41,827 (median: 9,059) accepted electrograms. The activation maps demonstrated sharply demarcated lines of conduction block, which suggested the location of surgical incisions, linear lesions, and natural anatomic obstacles, manifested by an abrupt change in activation time and color (Figure 1). Electrograms along the lines of conduction blocks demonstrated double atrial potentials with sharp near-field and small far-field potentials on either side of the line.
The conduction gaps were also easily identified on the activation and voltage maps. An example illustrated in Figure 2A clearly demonstrated a conduction gap after initial cavo-tricuspid isthmus ablation. Repeat ablation at this site achieved a complete conduction block across the gap. In Figure 2B, the gap between the RA septal incision and the cavo-tricuspid isthmus scar was the only conduction channel responsible for RA flutter. A single radiofrequency application at this site resulted in termination of the tachycardia. In Figure 2C, the voltage threshold range for scar was reduced to 0.2 to 0.3 mV to facilitate the identification of the surgical incision and conduction channel.
Mechanisms of ATs
During the electrophysiological study, a total of 64 ATs were identified in 51 patients. The number of ATs per patient averaged 1.3 ± 0.6 (range 1 to 3). The mechanism could not be determined in 1 AT because of significant cycle length, and activation sequence instability impeded complete mapping. The mechanisms of ATs were macro re-entry in 58 of 63 (92.1%) ATs, focal source in 4 (6.3%) ATs, and localized micro re-entry in 1 (1.6%) AT. All focal ATs were noted to originate from the LA: the anterior septum for 2 ATs, the right inferior PV in 1, and the posterior mitral annulus in 1 AT. The micro re-entrant AT was localized in the LA anteroseptal area.
Of the 58 macro re-entrant ATs, a single-loop re-entrant circuit was identified in 49 (84.5%) ATs. Rotation around the tricuspid annulus was observed in 26 of 49 (53.1%) ATs, which involved the RA incision either on the free wall (n = 5) or on the septum (n = 6) in 11 (22.4%) ATs, and the mitral annulus in 9 (18.4%) ATs. The re-entrant circuit involved the LA roof in 2 (4.1%) ATs and the LA posterior silent area in 1 (2%) AT.
Nine of 58 (15.5%) macro re-entrant ATs had a figure-8 circuit configuration. The activation proceeded through an isthmus between the tricuspid annulus and the RA free-wall incision for 2 ATs, between 2 RA free-wall incisions in 2 ATs (Figure 3, Online Video 1), between the RA free-wall incision and the septal incision in 1 AT, and between the mitral annulus and the septal incision or the left PVs in 2 ATs and 1 AT, respectively. A gap after surgical ablation resulted in macro-re-entry around the right PVs in 1 patient.
Comparison between patients with different surgical procedures
Of 11 patients with surgical repair of congenital heart disease, 6 (54.5%) had peri-tricuspid re-entrant ATs, 1 (9.1%) patient had RA free-wall incisional ATs, 4 (36.4%) patients had figure-8 re-entrant ATs, with an isthmus of conduction either between the tricuspid annulus and the RA free-wall incision or between the incisions, and no patients had LA or focal ATs.
Of 32 patients with valve replacement, peri-tricuspid ATs were observed in 14 (43.4%) patients, RA free wall or septal incisions-related ATs were seen in 7 (21.8%) patients, LA macrore-entrant ATs were observed in 12 (37.5%) patients, and LA focal and micro re-entrant ATs were seen in 5 (15.6%) patients. Of 12 patients with LA macro re-entrant ATs, 6 had had a previous surgical AF ablation.
Of 8 patients with valvuloplasty, peri-tricuspid ATs occurred in 6 (75%) patients, RA free-wall or septal incision-related ATs were seen in 2 (25%) patients, and figure-8 re-entrant ATs involved the mitral annulus and left PVs in 1 (12.5%) patient.
Ablation of ATs
Of 63 completely mapped ATs, radiofrequency delivery resulted in tachycardia termination and change to sinus rhythm in 50 (79.4%) ATs and transformation to another AT in 8 (12.7%) ATs. Noninducibility of any sustained ATs was achieved in 45 of 51 (88.2%) patients. Programmed stimulation to induce ATs was not performed in the remaining 6 patients, 1 of whom developed AF and who required intravenous administration of ibutilide to restore sinus rhythm after extensive LA ablation. Four of these 6 patients had refractory peri-mitral flutter terminated by either electrical cardioversion or burst pacing after ablation of the anterior line only or both anterior and lateral mitral isthmus lines with (n = 3) or without (n = 1) epicardial attempts through the coronary sinus. One patient with an unstable AT required electrical cardioversion to achieve sinus rhythm. All 4 patients with refractory peri-mitral flutter underwent either mitral valve replacement or valvuloplasty, and 3 of them had rheumatic heart disease.
Identification of macro pseudo re-entry
Macro pseudo re-entry was identified in 8 of 51 (15.7%) patients. The pseudo re-entry circuit was located at the RA in 4 patients and the LA in 4 patients. During RA mapping, a macro re-entrant AT that arose from the LA with complete cavo-tricuspid isthmus conduction block exhibited macro pseudo re-entry around the tricuspid annulus (Figure 4, Online Video 2); the 3 others with peri-tricuspid ATs demonstrated macropseudo re-entry around the RA free-wall incision. During LA mapping, a focal AT that arose from the posterior mitral annulus with previous surgical ablation between the right inferior PV and the mitral annulus showed macro pseudo re-entry around the mitral annulus (Figure 5); the 3 other ATs, either from the RA septal macro re-entrant AT or peri-mitral ATs, exhibited the early-meets-late circular activation in the LA posterior wall due to intrusion of extremely late activation into the early phase of the next cycle (Figure 6).
All macro pseudo re-entry activations could be identified by manually moving the window of interest, except for 2 cases in which the pseudo re-entry circuit had a figure-8 configuration. In 1 such case, 1 loop rotated in a clockwise direction around the tricuspid annulus, and the other loop moved counterclockwise around a RA free-wall incision (Figure 7, Online Video 3). The common channel seemed to be in the RA free wall between the tricuspid annulus and the incision, with the activation proceeding upward, dividing itself into 2 waves in the RA appendage base, with 1 wave going downward in the RA free wall along the incision and the other proceeding around the tricuspid annulus. Entrainment maneuvers confirmed the presence of a peri-tricuspid circuit, but ruled out peri-incisional re-entry. The re-entrant circuit was also ultimately validated by the result of cavo-tricuspid isthmus ablation that achieved sinus rhythm.
There were no embolic complications, including stroke or systemic embolism. Vascular complications such as groin hematomas and atrioesophageal fistulas were not observed. All patients survived to discharge. During a mean follow-up of 10 months (range 1 to 22 months), 42 of 51 (82.3%) patients were free of AT recurrence. Of the 9 patients with recurrence, 5 had previous unsuccessful AT ablation. A repeat ablation procedure due to AT recurrence was performed in 5 of 9 patients. Two patients had recurrence of ATs around the tricuspid annulus and the mitral annulus, respectively, due to recovery of isthmus conduction. Three patients developed new ATs: peri-mitral flutter occurred in 2 patients, 1 of whom had previous extensive LA ablation for AF; and a macro–re-entrant AT around an area of dense LA posterior scar occurred in another patient.
In this present study, we evaluated the mechanisms of ATs using the Rhythmia mapping system in patients with a history of open-heart surgery. There were several important findings relevant to clinical practice. First, this study demonstrated the feasibility to map complex post-surgical ATs using the Rhythmia mapping system. It allowed detailed activation mapping of the re-entrant circuit and clearly determined the underlying mechanisms. Second, the most common type of post-surgical AT was RA macro re-entrant ATs that involved the tricuspid annulus, the RA surgical incisions, or both, irrespective of the types of initial surgical procedures. LA macro re-entrant ATs occur less frequently in patients who only undergo surgical RA intervention, but these ATs are more common in those with valve replacement. Third, macro-pseudo–re-entry activity was not uncommon in the post-surgical setting but could be quickly identified by manual moving the window of interest for activation maps.
Mechanisms of post-surgical ATs
The most common mechanism of post-surgical ATs was macro re-entry in our study, including peri-tricuspid flutter, incisional ATs, and LA flutters. Focal ATs occurred less frequently. The specific type of ATs was dependent in part on the initial surgical intervention. RA macro re-entrant ATs were the most common tachyarrhythmias, irrespective of the types of initial surgical procedures. In patients with surgical repair of congenital cardiac anomalies, the macro re-entrant circuit involved either the tricuspid annulus and/or the RA free-wall incision, which might be attributable to little or no requirement for LA surgical manipulation. In contrast, LA re-entrant ATs were more common in patients who underwent valve replacement and were observed exclusively after operations that involved the LA. Furthermore, one-half of these patients had previous surgical AF ablation. Therefore, the most likely cause was iatrogenic LA lesions and underlying cardiac disease progression that resulted in substrate abnormalities. These results were comparable with the findings of a large study by Pap et al. (8), who evaluated the AT mechanisms in 100 post-surgical patients using a conventional electroanatomical mapping system. In their study, 58% of the ATs were cavotricuspid-isthmus dependent, 22.5% were RA incisional, and 8% were peri-mitral. However, ATs with a multiple-loop circuit were not deduced from their study. Therefore, the figure-8 re-entrant ATs frequently observed in our study might be considered to be 2 independent tachycardias in their study.
Ablation of ATs that involved the mitral annulus was more challenging in our study; the possible reasons for this included a complex LA substrate due to rheumatic heart disease, an iatrogenic pouch, and concern for prosthetic mitral valve damage. Conventional anterior and/or lateral mitral isthmus line(s) ablation was frequently ineffective; therefore, a tailored location of a linear lesion to target the mitral isthmus, based on the LA substrate, should be carefully considered and might be helpful to achieve ablation success in such a scenario.
Identification of macro pseudo re-entry
The macro pseudo re-entry pattern in post-surgical ATs was only reported previously in 2 cases with surgical repair of congenital anomalies (9,10). In these 2 patients, pseudo-typical atrial flutter occurred due to a focal source or an adjacent macro-re-entrant AT with slow areas of conduction. Our study showed that the incidence of macro pseudo re-entry during post-surgical AT mapping was 15.7% among all the study patients. However, the true incidence might be underestimated and highly dependent on study subjects, as well as mapping resolution and how this entity is defined. A recent study by Luther et al. (3) on 15 patients with previous AF ablation demonstrated that the activation pattern of localized re-entry was often pseudo re-entrant in 86% of small rotatory activations within the full spectrum of the color-coded map. Similarly, Ju et al. (11) reported that a macro pseudo re-entry activation pattern was identified in 4 of 13 cases with focal ATs and previous extensive ablation.
In our study, macro-pseudo–re-entry could be arbitrarily divided into 2 categories. Type 1 could be promptly ruled out by moving the window of interest, which suggests a potential relevance to window settings, due to either inadequate window settings (Figures 4 and 5) or extremely slow conduction exceeding the window of interest (Figure 6). Type 2, with a figure-8 activation pattern, could not be excluded by shifting the window of interest (Figure 7). Entrainment maneuvers in this scenario are necessary to determine the actual circuit and to avoid unnecessary ablation.
The understanding of the mechanism of an early-meets-late activation pattern is essential for successful curative therapy and an incorrect interpretation can lead to a more difficult and unnecessary strategy of linear ablation. Such an occurrence of a macro pseudo re-entry activation pattern could potentially increase the complexity of post-surgical ATs. Without careful recognition of this phenomenon, electrophysiologists might be misled to deliver ablation at these pseudo re-entrant sites, which are irrelevant to the clinical ATs and then would result in unsuccessful procedures.
Our study was a single-center experience with a relatively limited number of patients. However, this was a consecutive series that reflected real-world experience. A heterogeneous group of cases were included despite the commonality of all patients having a history of open-heart surgery. Other available mapping systems, including Biosense Webster’s CARTO system, which uses the PentaRay mapping catheter, and the EnSite Precision system (Abbott, St. Paul, Minnesota), which uses the circular mapping catheter or Advisor HD Grid mapping catheter (Abbott), also offer similar algorithms and allow high-density mapping. However, a comparison of mapping time, accuracy of maps, and identification of pseudo re-entry among different systems was not performed in the present study. For certain cardiac anatomic locations, it is possible that other catheters might be easier to manipulate because of their shapes and could potentially acquire more complete data, although the Orion mini-basket catheter might help generate higher resolution arrhythmia mapping. The manual annotation might be needed in the minority of the accepted electrograms because of artifacts, double potentials, and far-field ventricular electrograms around the valve areas, with resultant incorrect automated annotations. However, the small area with incorrect annotations was recognized easily on the high-density map as areas of inconsistent color coding to the large surrounding areas with correct annotations. We adjusted bipolar voltage cutoffs for visualizing scar tissue and conduction gaps (Figure 2C). Further study in association with imaging techniques is necessary to define the optimal voltage threshold and to facilitate substrate mapping. Some re-entrant circuits might be undefined because of complex activation patterns in such patients with previous surgical intervention.
The specific types of ATs were dependent in part on the types of surgical intervention. RA macro re-entrant ATs that involved the tricuspid annulus and/or RA incisions were the most common tachyarrhythmias, irrespective of the types of initial surgical procedures. It might be worthwhile to create prophylactic lesions of the cavo-tricuspid isthmus and other potential circuits related to atrial incisions during open-heart surgery, especially in patients with RA surgical manipulation only. LA macro re-entrant ATs occur more frequently after operations involving the LA. Pseudo re-entry activation is not uncommon in post-surgical patients and can be easily recognized by adjusting the mapping window to focus on the area of interest.
COMPETENCY IN MEDICAL KNOWLEDGE: RA macro re-entrant ATs predominate in patients with prior open-heart surgery irrespective of the types of initial surgical procedures, but LA macro re-entrant ATs occur more frequently in patients with valve replacement. Pseudo re-entry atrial activation is common and easily recognized by adjusting the mapping window.
TRANSLATIONAL OUTLOOK: Prophylactic lesions of cavo-tricuspid isthmus and other potential circuits related to atrial incisions during open-heart surgery might be useful to decrease the occurrence of post-surgical ATs especially in patients with RA surgical manipulation only.
↵∗ Drs. Xue and Liu contributed equally to this article, and are joint first authors.
Dr. Liu is supported by research grants from the National Natural Science Foundation of China (NSFC-81400259) and Guangdong Province (2014A030310470). Dr. Wu is supported by research grants from the Science and Technology Programs of Guangdong Province (No. 2014B070705005) and Guangzhou City (No. 201508020261). Dr. Yang Liu has received an educational fee from Boston Scientific. 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
- atrial fibrillation
- atrial tachycardia
- left atrial
- pulmonary veins
- right atrial
- Received January 9, 2018.
- Revision received July 3, 2018.
- Accepted July 6, 2018.
- 2018 American College of Cardiology Foundation
- Nakagawa H.,
- Ikeda A.,
- Sharma T.,
- Lazzara R.,
- Jackman W.M.
- Latcu D.G.,
- Bun S.S.,
- Viera F.,
- et al.
- Luther V.,
- Sikkel M.,
- Bennett N.,
- et al.
- Anter E.,
- McElderry T.H.,
- Contreras-Valdes F.M.,
- et al.
- Schaeffer B.,
- Hoffmann B.A.,
- Meyer C.,
- et al.
- Mantziari L.,
- Butcher C.,
- Kontogeorgis A.,
- et al.
- Yokota J.,
- Fujiu K.,
- Kojima T.,
- Komuro I.
- Yoshida N.,
- Yamada T.
- Ju W.,
- Yang B.,
- Chen H.,
- et al.