Radiofrequency ablation of premature ventricular contractions guided by robotic magnetic navigation combined with pattern matching filter

Abstract Background This study's intent is to evaluate the usefulness of pattern matching filter (PMF) function combined with robotic magnetic navigation (RMN) in guiding the ablation of premature ventricular contractions (PVCs). Hypothesis Assume that PMF can improve the outcomes of PVCs ablation using RMN. Methods A retrospective analysis was completed consisting of 118 consecutive patients with PVCs who underwent radiofrequency ablation guided by RMN. According to the application of PMF, patients were divided into two groups: 20 patients underwent ablation without PMF (group A), and another 98 patients received ablation incorporating PMF (group B). Results Compared with group A, the procedure time (135.0 ± 28.3 min vs. 106.3 ± 37.9 min, p = 0.02) in group B was significantly decreased, while the X‐ray exposure time (6.0 ± 2.6 min vs. 6.5 ± 3.6 min, p = 0.705) and dose (3.2 ± 2.4 gycm2 vs. 3.9 ± 2.7 gycm2，p = 0.208) had no significant difference. Group B had a more than twofold number of points acquired (66.9 ± 23.0 vs. 143.9 ± 68.3, p < 0.001) and required a shorter radiofrequency ablation time (13.2 ± 3.5 min vs. 8.1 ± 2.9 min, p < 0.001). There were no serious complications in either group. The acute success rate was similar [90.0% (18/20) vs. 87.8% (86/98), p = 1.000] in two groups, and the success rate was also similar in the long‐term follow‐up [83.3% (15/18) vs. 87.2% (75/86), p = 0.776]. Conclusions The ablation of PVCs guided by RMN is safe and effective. Combined with the functional capability of PMF, both procedure time and radiofrequency ablation time were significantly decreased.


| INTRODUCTION
Radiofrequency ablation therapy guided by robotic magnetic navigation (RMN) is a well-studied and acknowledged treatment for premature ventricular contractions (PVCs) and other complex arrhythmias with high efficacy, fewer complications and less X-ray exposure versus traditional manual catheter ablation, and a short technology learning curve. [1][2][3][4][5][6][7] Since 2013, our center has been employing RMN combined with three-dimensional electroanatomic mapping system (CARTO, Biosense Webster Inc) in ablation therapy across a wide range of complex arrhythmias including PVCs and ventricular tachycardia. During our center's early experience, the mapping aspect of PVCs procedures presented challenges. With the innovation of CARTO technology, new functional software called pattern matching filter (PMF, Biosense Webster Inc) has been offered which enables a correlation to be made between the real time PVCs and the predefined template morphology. Limited studies have reported that PMF can improve the outcomes in manually controlled catheter ablation. 8 The object of this study was to evaluate the usefulness of PMF in combination with RMN for PVCs ablation.

| Study population
Between May 2017 and May 2022, in Wuxi People's Hospital affiliated to Nanjing Medical University, a total of 118 patients with symptomatic and medically refractory PVCs underwent first time ablation therapy using RMN and were included in this retrospective study. All procedures were performed using RMN in our center due to the system's superior safety and efficacy (Before May 2017, our center had completed ablation of complex arrhythmia guiding by RMN for more than 150 cases.). 9

| Mapping and ablation
After five half-lives withdrawal of antiarrhythmic drugs, procedures were performed with patients in a fasting and conscious state. In the PMF mapping process, the first step was to collect 12.5 s 12lead ECG information in the pattern bank of the Carto system, including at least one clinical PVCs ( Figure 1A). Second, the target PVCs was selected in the pattern bank window and included only one QRS complex of the target PVCs in the range of the PMF interest window ( Figure 1A). Third, the PMF function was set and enabled in the preferences user interface, and the correlation percentage of 97% selected ( Figure 1B). When the morphological similarity between the real time PVCs and the template PVCs is at or above 95%, the system automatically takes a point for activation mapping ( Figure 1C,D). After completing the above steps, the RMN catheter was navigated around the ventricular 3D map and the PMF function automatically obtained points during activation mapping. Subsequently, the earliest activation point gradually appeared in the 3D space. Finally, when required pace mapping was used to confirm the earliest activation point of the PVCs.
In the non-PMF group A, instead of leveraging software the mapping process required the traditional manual method of taking points when the target PVCs presented.
Ablation points were delivered according to the following criteria and all criteria were met at the same time: local activation at least 20 ms before QRS, greater than 95% of QRS matches during pacing mapping, and unipolar ECG assumed QS morphology. The radiofrequency (RF) power setting ranged from 20 to 40 W, depending on the region being targeted. The temperature control setting was set with an upper limit of 43°C, and the flow setting of normal saline during ablation was 17-20 mL/min. If PVCs reduction or elimination was observed within 10 s after ablation energy delivery, we continued to ablate for 90-120 s, otherwise terminated ablation, and mapped to find the new ablation site. 12 After the first ablation, isoproterenol was injected intravenously, and the patient observed for 30 min. If there was no PVCs recurrence, the procedure was defined as an acute success. In the event of PVCs recurrence, further diagnosis and ablation continued. If all identified ablation points were ablated and this did not terminate the PVCs, the procedure was defined as unsuccessful.
Procedure-related parameters were recorded, including X-ray exposure time, X-ray dose, procedure time, ablation time, and complications. Procedure time was defined as the time from the first femoral vein puncture to the sheath removal.

| Complications
Complications were classified into two categories: major and minor.

| Characteristics of mapping and ablation
Characteristics of mapping and ablation are shown in Table 2 and Supporting Information: Figure S1. Compared with Group A, the procedure time (135.0 ± 28.3 min vs. 106.3 ± 37.9 min, p = 0.02) in Group B was significantly decreased. However, the X-ray exposure time (6.0 ± 2.6 min vs. 6.5 ± 3.6 min, p = 0.705) and dose (3.2 ± 2.4 gycm 2 vs. 3.9 ± 2.7 gycm 2 ，p = 0.208) had no significant difference. Group B had a more than twofold increase in the number of points acquired (66.9 ± 23.0 vs. 143.9 ± 68.3, p < 0.001) and required a shorter radiofrequency ablation time (13.2 ± 3.5 min vs. 8.1 ± 2.9 min, p < 0.001). There were no serious complications in either group.

| Distribution of PVCs
The distribution of PVCs origin points is listed in Supporting Information: Table S1. A considerable percentage of PVCs originate from the outflow tract (OT) of either the RV or the LV, and this study yielded similar results. 13 The proportion of PVCs originating from the RVOT alone accounted for more than half at 57.6% (68/118).
Therefore, in our analysis PVCs were generally divided into RVOT origin and non-RVOT origin for interpretation of results.

| Procedural outcomes
Procedural outcome parameters of acute success and recurrence during follow-up are listed in Table 3 and Figure 2.  variation in manual dexterity between operators and the associated impact on safety and efficacy. 18,19 Several studies have shown similar success rates for PVCs ablation under the guidance of RMN when compared to manual ablation. 16,20 The key advantage of RMN ablation is consistent reduction in X-ray exposure for physicians and patients, and no serious complications were reported. 16 Figure S1). Then, the accuracy of radiofrequency ablation energy delivery was also increased. Therefore, the duration of ablation energy delivery in PMF using group reduced.
In terms of outcomes, no statistically significant results were found.
Compared with the success rate of manual ablation reported in previous studies (ranging from 50% to >90%), there is no significant difference in the success rate of PVCs ablation using RMN in our center, which is consistent with our expectations. 15,16 However, our study shows that the application of PMF function failed to improve the acute success rate and failed to reduce the recurrence, which is inconsistent with our expectations. After analyzing the data, we found that we conducted detailed mapping for all patients as much as possible without PMF. indicates that the automatic mapping function is significantly faster than manual point-by-point 3D mapping and results in shorter mapping time and higher point density. 23 Although the automatic mapping function cannot increase the success rate of surgery, the reduction of operation time can also bring benefits to patients, such as reducing the pain and psychological burden, and potentially reducing the possibility of complications. For the operator, reducing the operation time means that more patients can be treated in unit time.
In terms of X-ray application, the results of our center showed that the X-ray exposure of RMN ablation was limited. There was no significant difference in X-ray exposure between the two groups because the application of X-ray was not involved in the mapping stage.

| LIMITATIONS
Limitations for this study are acknowledged as follows. First, RMN ablation of PVCs combined with the PMF should be further evaluated in randomized controlled trials and compared to manual ablation. Second, the number of patients in this study was relatively small, and therefore the comparison is inherently underpowered. Third, this study is limited by the inherent nature of a retrospective study and the results need to be confirmed by prospective studies. Fourth, it is unknow about the PVC burden during procedure in both groups. Therefore, the comparison of the number of acquired points in unit time may exist statistical bias.

| CONCLUSION
In conclusion, PVCs ablation using RMN combined with the function of PMF is safe and effective. Utilizing the PMF function, both total procedure time and ablation time are significantly reduced. However, 572 | LIU ET AL.
RMN itself is still the technology driving safety and efficacy of the ablation procedure.

ACKNOWLEDGMENTS
This study was supported in part by grants from the National Natural Science Foundation of China (No. 82000317 to Xiao-yu Liu and