Procedural outcomes and learning curve of cardiac arrhythmias catheter ablation using remote magnetic navigation: Experience from a large‐scale single‐center study

Abstract Background Remote magnetic navigation (RMN)‐guided ablation has become an inspiring method of catheter ablation for tachyarrhythmias. Hypothesis Data from a large‐scale single center may provide further insight into the safety of and the learning curve for RMN‐guided ablation. Methods A total of 1003 catheter ablation procedures using RMN for conditions including supraventricular ventricular tachycardia, atrial tachyarrhythmias, and premature ventricular contraction/ventricular tachycardia (PVC/VT) were retrospectively analyzed from an ablation registry. Procedural outcomes, including procedure time, mapping time, X‐ray time, and RF time, were assessed. The complications were classified into two categories: major and minor. A subanalysis was used to illustrate the learning curve of RMN‐guided ablation by assessing procedure time and total X‐ray time of 502 atrial fibrillation (AF) ablation procedures. Results Among these procedures, 556 (55.4%) were AF and 290 (28.9%) were PVC/VT. Electrical pulmonary vein isolation was achieved in 99.0% of AF procedures, and acute success reached 90.3% in PVC/VT procedures. The overall complication rate was 0.5%. In the subanalysis of AF procedures, the overall procedure time and X‐ray time of procedures were short (125.9 ± 54.6 and 5.3 ± 3.9 minutes, respectively) and proceeded to decrease from the initial 30 procedures to about 300 procedures, where the learning curve reached plateau, demonstrating maximum procedure efficiency. Conclusions RMN‐guided ablation is safe, as verified by very low overall complication rate and reduced X‐ray time. In our study, even the first AF procedures had a relatively low procedure time and total X‐ray time, and procedure efficiency improved during the learning curve.


| INTRODUCTION
Catheter ablation has been well established in the treatment of a variety of tachyarrhythmias. 1 Such an anatomy-based therapy has placed high requirements on the precision of localization, stability of catheter contact, and catheter flexibility. Even experienced cardiac electrophysiologists performing certain ablation procedures may encounter difficulties in reaching complex anatomical regions and maintaining the stability of the catheter, leading to impaired safety and success rate. [2][3][4] The remote magnetic navigation (RMN) system for catheter ablation offers the advantages of precise and flexible catheter navigation, 5,6 reduction in peri-procedure complications, 7 and X-ray exposure. 8 However, previous studies regarding RMN-guided ablation have been limited by small sample size and few arrhythmia types. In addition, no large sample data targeted the analysis of a learning curve for procedure efficiency. Here, we have studied over 1000 RMNguided ablation cases in a single center, with the objectives of evaluating its safety and the procedural outcomes among patients with different types of arrhythmias, and determining the learning curve for RMN-guided ablation.
F I G U R E 1 An example of mapping and ablation for premature ventricular contraction (PVC) using remote magnetic navigation (RMN). A, Surface ECG of PVC and intracardiac recordings of the earliest activation site in coronary sinus (great cardiac vein). B, The combination of activation CARTO map of PVC and X-ray image in "Navigant" system. C, Fluoroscopic view in the posterior-anterior oblique view. The ablation catheter is pointing toward LV summit after reaching LV cavity through a transseptal approach, this forming a "reverse S-curve." D, Fluoroscopic view in the left anterior oblique view. The ablation catheter is pointing toward left coronary cusp through a retrograde transaortic approach. E, The ablation catheter was manipulated with RMN in the great cardiac vein. The catheter tip is equipped with three small magnets. By changing the relative orientation of the external magnets, effective control of the catheter deflection can be achieved 2 | METHODS

| Patient population
Between May 2010 and October 2019, 1003 consecutive cases, ranging from supraventricular tachycardia (SVT) to atrial tachycardia (AT), atrial flutter (AFL), atrial fibrillation (AF), premature ventricular contraction (PVC), and ventricular tachycardia (VT), targeted for ablation were included in a registry at Ruijin Hospital, Shanghai Jiao Tong University School of Medicine. The diagnoses of patients included in the registry were all established according to the current guidelines. [9][10][11] All the patients signed informed consent prior to the RMNguided ablation procedure. Any antiarrhythmic drug was discontinued before the ablation procedure for at least 5 half-lives except the betablockers for the control rate for persistent AF and VT storm.

| Preparation for procedure
For patients with SVT, internal jugular and left femoral veins were selected for puncture. A decapolar catheter, a bipolar catheter, and a quadripolar catheter (St. Jude Medical, Inc., St. Paul, Minnesota) were placed within the coronary sinus, at the right ventricle apex, and at the His bundle, respectively.
For patients with AF, standard anticoagulation therapy was given prior to the procedure. All patients underwent effective anticoagulation therapy for at least 1 month prior to the procedure, either with an uninterrupted vitamin-K antagonist (warfarin) or with non-vitamin-K antagonist oral anticoagulants, dabigatran, or rivaroxaban. INR was not allowed to exceed 3 prior to ablation for the patients administered with warfarin. Preprocedural transesophageal echocardiography was used to exclude atrial thrombosis. A decapolar catheter and a bipolar catheter

| Mapping and ablation with RMN
The RMN Niobe ES (Stereotaxis Inc., St. Louis, Missouri), the CARTO respectively. Doctor's X-ray time was calculated as the time difference between the total X-ray time and the control room's X-ray time. RF applications and RF time reflected the total sum of the number and minutes of ablation burns in the procedure, respectively.
Acute procedural success was the endpoint of the study.

| Learning curve
We divided consecutive AF procedures performed by one operator into three phases for analyzing the learning curve with two parameters of procedural outcomes: procedure time and X-ray time. The

| Complications
Adverse events or peri-procedural complications in this study were divided into two categories: major and minor. Major complications con-

| Complications
A detailed description of the complications is given in

| Procedural outcomes
Procedural outcomes data included 821 complete record sets that were automatically derived from the RMN system (Table 2) from May 2015.
All the SVT and focal AT cases were successfully ablated with RMN. Electrical PVI was achieved in 99.0% of the studied AF procedures. Acute success reached 90.3% in PVC/VT procedures.

| Learning curve
An analysis was performed to determine the learning curve of the procedure and the total X-ray time for consecutive AF ablation procedures by one operator and the results are shown in Figure 4A,C (R 2 = 0.97) and (R 2 = 0.98), respectively. The curves of procedure time and total X-ray time started to decrease from the initial 30 procedures, and reached a plateau after 300 and 350 procedures, respectively, where both regression coefficients became indistinguishable from zero (P = .46and P = .87).
Comparisons of the procedure and total X-ray time among three phases are given in Figure 4B

| Major findings
To our knowledge, this is the first large-scale study reporting the safety, the learning curve, and the procedural outcomes among different types of arrhythmias for RMN-guided ablation in a single center.
The main findings of this study are as follows. RMN-guided ablation is safe, as verified by the low overall complication rate (0.5%) and reduced X-ray time. The learning curve of RMN is relatively short.
Even at the very beginning of the study, low procedure time and total X-ray time could be achieved and procedure efficiency could be improved along the learning curve.

| Learning curve of RMN-guided ablation
In this study, AF cases, with the largest sample size, were used to demonstrate the effect of the learning curve on RMN-guided ablation procedure outcomes. To our knowledge, this might be the first largescale study regarding RMN learning curve. We have established that as the experience grows, both procedure time and total X-ray time decreases significantly. Previously, Pappone et al 12  The complication rate in our series is remarkably low; therefore, in general, RMN can be quickly adopted by electrophysiologists without the cost of compromised safety.

| Safety of RMN-guided ablation
Earlier studies have demonstrated that compared with manual procedures, RMN-guided ablation has a lower complication rate. 13,14 However, previous reports mostly focused on a single type of arrhythmia.
In this retrospective study, 1003 cases were analyzed and arrhythmia types included SVT, AT, AFL, AF, PVC, and VT. The overall complication rate was 0.5% and the major complication rate was even lower (0.1%). Only one case underwent cardiac tamponade, and no procedure-related death occurred. Our data demonstrated the general safety of RMN ablation for all above-mentioned arrhythmias. A recent epidemiological study that calculated the incidence of complications in patients who underwent catheter ablation from 2000 to 2013 in the United States reported that the total incidence of in-hospital periprocedural complications was 5.46%, and VT ablation was associated with the highest rate of complication (9.90%). 15 Of note, the safety endpoint in our study is independent of the type of arrhythmia ablated. The superior safety profile of RMN can be attributed to the following reasons. First, the floppy and steerable ablation catheter can achieve a relatively low tissue contact force, reducing the possibility of cardiac perforation. Second, the RMN system can maintain catheter stability despite changes in cardiac rhythm or cardiorespiratory movement, which remarkably reduces the incidence of pericardial tamponade. [16][17][18] Hematoma or hemorrhage at access site is the most common complication of ablation. 15 Compared with the manual procedure, the sheath of RMN system is fixed to the QuikCAS Cardiodrive system and remains relatively still with the femoral vein.
This allows for fewer sheath maneuvers and can reduce endothelium injury and thrombosis at the tip of the sheath. Finally, the relatively large number of procedures in our center may also contribute to the low complication rate.

| Fluoroscopy and procedure time of RMN-guided ablation
Multiple studies have shown that exposure to radiation is associated with tumorigenesis among interventional cardiologists. 19,20 Meanwhile, studies comparing RMN and manual ablation have consistently reported that application of RMN can reduce radiation exposure. 21,22 A recent meta-analysis also reported a significant reduction in fluoroscopy time. 23 The mean total X-ray time (5.3 ± 3.9 minutes) was remarkably low and X-ray time continued to decrease as the learning curve traversed. Moreover, we noted that differences in X-ray time among ablations for differing types of arrhythmias was more related to doctor's X-ray time than to control room's X-ray time. This finding further supports that when the operator is in the control room, utilizing the RMN system, radiation exposure can be kept low regardless of the type of arrhythmia ablated.
There are conflicting data regarding procedure time of RMNguided ablation. Some 24 have reported prolonged procedure time while others 25 have disagreed, reporting lower and desirable procedure time. Even published meta-analyses have reached differing conclusions, possibly due to dissimilar inclusion criteria. 26,27 The heterogeneity between studies regarding procedure time may be influenced by studies completed by different generations of RMN, during an extended time period in which globally, there was neither understanding nor consensus regarding optimal application of the technology. Additionally, when compared with conventional ablation and other RMN studies, procedure time is relatively low in this study. This may be related to the effect of the learning curve, increased use of RMN best practices, and large sample size.
Although our data are promising, larger randomized controlled trials are needed to demonstrate the superiority of RMN over traditional manual pull-wire catheter navigation, for the ablation treatment of cardiac arrhythmias.

| LIMITATIONS
This is a single-center retrospective observational study. Second, in AF procedures, additional ablations such as linear ablation and CAFÉ ablation, which could potentially influence the procedural outcomes, were not analyzed. Third, this was not a controlled study to compare the procedural outcomes with conventional manual procedures.

| CONCLUSIONS
RMN-guided ablation is safe, which is confirmed by a very low overall complication rate and reduced X-ray time. The procedure and X-ray times decreased along the learning curve of RMN.