Evaluation of the input site and characteristics of the antegrade fast pathway based on three-dimensional bi-atrial stimulus-ventricle mapping

Purpose Previous studies examined the right atrial (RA) input site of the antegrade fast pathway (AFp) (AFpI). However, the left atrial (LA) input to the atrioventricular (AV) node has not been extensively evaluated. In this study, we created three-dimensional (3-D) bi-atrial stimulus-ventricle (St-V) maps and analyzed the input site and characteristics of the AFp in both the RA and LA. Methods Forty-four patients diagnosed with atrial fibrillation or WPW syndrome were included in this study. Three-dimensional bi-atrial St-V mapping was performed using an electroanatomical mapping system. Sites exhibiting the minimal St-V interval (MinSt-V) were defined as AFpIs and were classified into seven segments, four in the RA (F, S, M, and I) and three in the LA (M1, M2, and M3). By combining the MinSt-V in the RA and LA, the AFpIs were classified into three types: RA, LA, and bi-atrial (BA) types. The clinical and electrophysiological characteristics were compared. Results AFpIs were most frequently observed at site S in the RA (34%) and M2 in the LA (50%), and the BA type was the most common (57%). AFpIs in the LA were recognized in 75% of the patients. There were no clinical or electrophysiological indicators for predicting AFpI sites. Conclusions Three-dimensional bi-atrial St-V maps could classify AFpIs in both the RA and LA. AFpIs in the LA were frequently recognized. There were no significant clinical or electrophysiological indicators for predicting AFpI sites, and 3-D bi-atrial St-V mapping was the only method to reveal the precise AFp input site.


Introduction
The exit site of the retrograde fast pathway (RFp) from the atrioventricular node (AVN) can be determined by mapping the earliest activation site in the atrium during constant pacing from the ventricle. On the other hand, the input site of the antegrade fast pathway (AFp) to the AVN has been very difficult to determine and only stimulus-His (St-H) mapping can identify the input site of the AFp to the AVN (AFpI) [1][2][3][4][5]. The use of a three-dimensional (3-D) mapping system makes identification of the exact location of the AFpI easier to determine [4,5]. Previous studies, however, have examined the AFpI only in the right atrium (RA) (AFpI [RA]). The left atrial (LA) input to the AVN has not been extensively evaluated and only Gonzalez et al. have suggested that an AFpI exists in the LA (AFpI[LA]) with high probability in 95% of patients [6]. In this study, we created 3-D bi-atrial stimulus-ventricle (St-V) maps, which were equivalent to St-H maps, and analyzed the input site and characteristics of the AFp in both the RA and LA.

Study population
From June 1, 2019, to August 31, 2020, forty-four consecutive patients (31 men and 13 women, 64 ± 14 years, range 25 to 80 years) with atrial fibrillation (AF) (18 paroxysmal and 16 persistent) or Wolff-Parkinson-White (WPW) syndrome with a left free wall accessory pathway who underwent catheter ablation were included in this study. Patients with first degree atrioventricular block and left bundle branch block were excluded because an AFp might not exist in the former and because transient AV block during mapping might occur in the latter. The institutional review board of Tokushima University approved the study protocol. Written informed consent was obtained from all patients.

Electrophysiological study
Patients with atrial fibrillation were studied under deep sedation with a continuous infusion of propofol. They were on a ventilator and underwent continuous monitoring of the blood pressure, oxygen saturation, and bispectral index. Patients with WPW syndrome were studied under local anesthesia using xylocaine. Venous access was established percutaneously from the right jugular vein and right femoral vein to introduce electrode catheters into the RA, right ventricle (RV), coronary sinus (CS), and LA. A His bundle electrogram recording catheter (His) was used in the case of WPW syndrome, but not in the case of AF. All patients required a transseptal left heart catheterization for a pulmonary vein isolation or accessory pathway ablation. To record the electrocardiogram from the RA and CS, a 6-Fr catheter with 20 electrodes (BeeAT; Japan Lifeline, Tokyo) was inserted via the right jugular vein into the CS. A 5-Fr catheter with 5 electrodes (Arma, Century Medical Inc., Tokyo, Japan) was positioned in the RV apex (RVA). The CARTO®3 System (Biosense Webster, Irvine, California) was used for 3-D mapping in all patients. The bipolar electrocardiograms were filtered at 30-400 Hz for the electrophysiological analysis (CardioLab, GE Healthcare Japan, Tokyo).

Identifying the input site of the antegrade fast pathway using 3-D bi-atrial St-V maps
During the waiting period after the ablation, 3-D bi-atrial St-V maps for identifying the AFpI were created in all patients during constant atrial pacing at 100 ppm (600 ms) from a 7-Fr irrigation catheter (Thermocool SmartTouch SF, Biosense Webster). The pacing output was set at 3-4 V/1 ms in order to avoid capturing the AVN, His bundle, and ventricles. As shown in ), respectively. The sites where the His bundle electrocardiogram was recorded were also stored on the 3-D map during sinus rhythm to identify the apex of the triangle of Koch (TOK) without the electrocardiographic information, and the A-H interval during sinus rhythm was measured at those sites. The anatomical information related to the roof and floor of the CS ostium, tricuspid annulus, and mitral annulus were also stored on the 3-D map.

Classification of the input site of the antegrade fast pathway
As shown in Fig. 2, for the classification of the AFpI, the atrial septum was subdivided into seven segments, four on the RA aspect and three on the LA side. In the RA, the TOK was classified into three equidistant parts: superior, middle, and inferior thirds defined as sites S, M, and I, respectively. Also, the part posterior to the TOK was defined as site F. If the AFpI was located at site M or I, the AFpI was defined as an inferiorly dislocated AFp (IDF). In the LA, the AFpI was classified into three segments along the mitral annulus (MA): the 6-7 o'clock, 7-8 o'clock, and 8-9 o'clock directions along the mitral annulus were labeled as sites M1, M2, and M3, respectively.

Classification of the type of input site of the antegrade fast pathway by combining the right and left atria
As shown in

Statistical analysis
Data are expressed as the mean ± SD. Differences between the means were compared using a one-way analysis of variance or the Kruskal-Wallis test. A Levene's test was used to check for the equality of variance. The differences in the proportions were compared using the chi-square test. A P < 0.05 was considered statistically significant. All statistical analyses were performed with the Statistical Package for the Social Sciences for Windows software (SPSS, version 27, Chicago, IL, USA).

Clinical and electrophysiological characteristics of the study population
As shown in Table 1, 66% (29/44) of the patients in this study population continued antiarrhythmic drugs during the procedure. In patients with WPW syndrome, the antiarrhythmic drugs were discontinued preoperatively. Twenty-nine of 34 patients with AF continued on antiarrhythmic drugs. Class I drugs were used in 13 patients (cibenzoline 7, pilsicainide 5, and propafenone 1), class II (bisoprolol) in 7 patients, and class IV in 9 patients (verapamil 4 and bepridil 5). No class III drugs were used. The mean left atrial volume index (LAVI) was 40 ± 14 mL/m 2 . The A-H interval during sinus rhythm was 94 ± 16 ms. The number of mapping points was 29 ± 9 points in the RA and 25 ± 10 points in the LA. The minimal St-V interval was 162 ± 45 ms in the RA and 161 ± 42 ms in the LA.

Location of the input site of the antegrade fast pathway in the RA
As shown in Fig. 2 and  (Fig. 4).

Location of the input site of the antegrade fast pathway in the LA
As shown in Fig. 2 and Table 3, the MinSt-V[LA] was observed at site M1 in 10 (23%), site M2 in 22 (50%), and site M3 in 12 (27%). There were no significant differences among these three groups with regard to the age, gender, type of arrhythmia, LAVI, A-H interval during sinus rhythm, minimal St-V interval, and site of the MinSt-V[RA]. Interestingly, only sites F and S were recognized in the LA type. The minimal distance between the His bundle electrogram recording sites and MinSt-V(LA) was not correlated with the A-H interval during sinus rhythm (Fig. 4).

Type of the input site of the antegrade fast pathway by combining the right and left atria
As shown in Table 3, the AFpI was classified as an RA type in 11 (25%), LA type in 8 (18%), and BA type in 25 (57%). There were no significant differences among these three groups with regard to the age, gender, type of arrhythmia, LAVI

Major findings
In the present study, using the 3-D bi-atrial St-V map, it was possible to classify the input sites of the AFp in both the RA and LA.

St-V map as an alternative to an St-H map
In the previous studies, the St-H map was the standard map used for the determination of the AFpI [1][2][3][4][5]. In this study, an St-V map was used instead of the St-H map because a His catheter was not used in the majority of the patients who had atrial fibrillation (77%) and the St-V map was equivalent to the St-H map. As shown in Fig. 1, the St-V interval (ai + b + c) was equal to the interval that consisted of the St-H interval (ai + b) and interval between the onset of the His bundle electrogram and peak of the QRS wave in lead II (c). Here, the interval between the onset of the His bundle electrogram and peak of the QRS wave in lead II (c) was equivalent to the sum of the H-V interval and

Left antegrade fast pathway input to the atrioventricular node
To the best of our knowledge, this study was the first to investigate the detailed input site of the AFp in the LA. The 3-D bi-atrial St-V mapping with the CARTO system in this study revealed that an AFpI[LA] was present in 75% of the patients and was the only AFpI in 18%. In the previous reports, an AFpI[LA] was suggested in cases with a left posteroseptal accessory pathway ablation [11,12] or left-sided fast pathway ablation for typical atrioventricular nodal reentrant tachycardia (AVNRT) [13]. Moreover, the low success rate (46-69%) of a right-sided fast pathway ablation for AVNRT reported by Mitrani

Right antegrade fast pathway input to the atrioventricular node
The 3-D bi-atrial St-V mapping with the CARTO system in this study revealed the presence of an AFpI[RA] in 82% of the patients and only an AFpI in 25%. An inferiorly dislocated AFp (IDF) was recognized in 41% (18/44) of the patients. The incidence of an IDF in this study was much higher than that reported by Delise et al. (10%) [2], but was similar to that reported in the other Japanese studies (33-56%) [3,4]. These differences might be related to the ethnic differences and therefore, further studies will be needed.

Study limitations
(1) This study was conducted at a single center and included only a small number of patients. (2) Antiarrhythmic drugs were continued in many patients (66%) because the majority of the patients studied had AF. Therefore, the St-V interval and other electrophysiological parameters could have been affected by those drugs. (3) Two different diseases, AF and WPW, were included in this study, and all AF patients were studied under deep sedation and all WPW patients under local anesthesia. Therefore, the difference in the disease and effect of the sedation could have affected the electrophysiological parameters. Further studies are needed to evaluate the effects of those. (4) In this study, we did not compare the effect of isoprenaline loading. The possibility that the input site of the AFp could change according to the isoprenaline loading remained. (5) We fixed the pacing cycle length at 600 ms in order to evaluate the AFp, to overdrive the sinus rate, and to avoid Wenckebach periodicity. However, keeping a uniform pacing rate might have led us to overlook other concealed AV pathways. (6) In this study, the St-V map was not compared with the St-H map. Although further studies will be needed to confirm this, it is reasonable to assume that the two maps would be equivalent.

Conclusion
The 3-D bi-atrial St-V map could classify the input site of the AFp in both the RA and LA. An AFpI[LA] was frequently present and was recognized in 75% of the patients. Except for the 3D bi-atrial St-V mapping, there were no significant clinical or electrophysiological indicators that could reveal the precise input site of the AFp.