https://orcid.org/0000-0003-3753-7883
Abstract
An incomplete understanding of the mechanism of atrial tachycardia (AT) is a major determinant of ablation failure. We systematically evaluated the mechanisms of AT using ultra-high-resolution mapping in a large cohort of patients.
We included 107 consecutive patients (mean age: 65.7 ± 9.2 years, males: 81 patients) with documented endocardial gap-related AT after left atrial ablation for persistent atrial fibrillation (AF). We analysed the mechanism of 134 AT (94 macro-re-entries and 40 localized re-entries) using high-resolution activation mapping in combination with high-density voltage and entrainment mapping. Voltage in the conducting channels may be extremely low, even <0.1 mV (0.14 ± 0.095 mV, 51 of 134 AT, 41%), and almost always <0.5 mV (0.03–0.5 mV, 133 of 134 AT, 99.3%). The use of multipolar Orion, HDGrid, and Pentaray catheters improved our accuracy in delineating ultra-low-voltage areas critical for maintenance of the circuit of endocardial gap-related AT. Conventional ablation catheters often do not detect any signal (noise level) even using adequate contact force, and only multipolar catheters of small electrodes and shorter interelectrode space can detect clear fractionated low-amplitude and high frequency signals, critical for re-entry maintenance. We performed a diagnosis in 112 out of 134 AT (83.6%) using only activation mapping and in 134 out of 134 AT (100%) using the combination of activation and entrainment mapping.
High-resolution activation mapping in combination with high-density voltage and entrainment mapping is the ideal strategy to delineate the critical part of the circuit in endocardial gap-related re-entrant AT after AF ablation.
High-resolution activation mapping in combination with high-density voltage mapping and entrainment mapping is the ideal strategy for ablation and to delineate the critical part of the circuit in endocardial gap-related re-entrant atrial tachycardia (AT) after atrial fibrillation ablation.
Voltage in the conducting channels may be extremely low, even <0.1 mV, and almost always <0.5 mV.
Multipolar catheters improve our accuracy in delineating ultra-low-voltage areas, critical for the maintenance of the circuit of endocardial gap-related AT.
Termination sites are linked with negative PPI-TCL using high output pacing (10–15 mA, 2 ms) and PPI-TCL < 20 ms using low output pacing (3–7 mA, 1 ms).
Introduction
Complex macro-re-entrant or localized re-entrant atrial arrhythmias may occur spontaneously or can result from iatrogenic substrate creation after ablation, especially for atrial fibrillation (AF).1–3 Detailed knowledge of the precise mechanisms underlying atrial tachycardia (AT) remains challenging, especially in patients with atrial scarring from previous atrial ablation or surgery or incomplete ablation of AF driver. Ultra-high-density mapping may achieve accurate characterization of circuits/mechanisms in AT.4 Advances in mapping technologies include introduction of multielectrode mapping catheters, designed to acquire a larger number of electrograms (EGMs) per beat, producing higher density maps with shorter mapping times.4,5 In addition, electrodes have become progressively smaller and more closely spaced reducing the sampled tissue size and increasing the mapping resolution (details are provided in Supplementary material online). In this study, we sought to evaluate the electrophysiological characteristics of the termination sites of tachycardia using multipolar catheters, measuring the amplitude and duration of the fragmented bipolar signals in the critical part of the isthmus of endocardial gap-related AT.
Methods
Consecutive patients with endocardial gap-related AT post-AF ablation were included in our study. All patients gave written informed consent. Data collection and analysis were done under a protocol approved by the institutional review committee.
Ablation procedure
We prospectively enrolled consecutive patients who underwent radiofrequency catheter ablation for endocardial gap related AT post-AF ablation in three different high-volume centres between January 2016 and November 2018. All patients received uninterrupted anticoagulation for at least 4 weeks before the procedure. Antiarrhythmic drugs were discontinued >5 half-lives prior to ablation with the exception of amiodarone. Catheter ablation procedures were performed under conscious sedation using midazolam and morphine. In all patients, transesophageal echocardiography or an enhanced CT scan was performed before the procedure to exclude intracardiac thrombus. During the procedure, intracardiac EGMs were continuously recorded and stored at a sweep speed of 100 mm/s on a computer-based digital system (Labsystem Pro, BARD, Boston Scientific, MA, USA). After placement of a deflectable decapolar catheter within the coronary sinus (CS) (Extreme, Sorin/Dynamic XT, Boston Scientific, MA, USA), a transseptal puncture was performed under fluoroscopic guidance; one transseptal sheath (BRK Needle, Agilis NxT steerable sheath, St Jude Medical) was introduced into the left atrial (LA). Before inserting the mapping multipolar catheter, heparin (100/IU/kg) was administered to achieve an activated clotting time (ACT) of >300 s. Subsequent boluses of heparin were administered every 20–30 min as needed (target ACT >300 s). Details regarding high-density activation, voltage, and entrainment mapping using multipolar catheters (Orion, HDGrid, and Pentaray) are provided in Supplementary material online.
The AT was classified according to our previous publication6 (see Supplementary material online).
Exclusion criteria
We excluded all AT for which we were unable to perform activation mapping and identify the mechanism of the arrhythmia. If we could not map >10% of the TCL, we did not analyse the electrophysiological characteristics of the circuit. Biatrial AT were excluded, together with AT displaying irregular TCL (beat-to-beat variations >20 ms).7 We also excluded AT using epicardial connections as an active part of the circuit (missing >10% of the TCL), CS or Marshall Bundle-related AT,8 roof dependent AT using septopulmonary bundle, and biatrial tachycardias using the Bachmann bundle or interatrial connections.9,10 We excluded all patients where we had only activation mapping without entrainment mapping and patients where the AT changed to another AT or terminated to SR during entrainment.
Catheter ablation target selection
A 3.5-mm tip open-irrigated catheter (ThermoCool SF, Biosense Webster, Diamond Bar, CA, USA) was used for procedures using Rhythmia and Carto, and a pressure sensing 3.5 mm irrigated tip ablation catheter with 2‐2‐2 mm interelectrode spacing (Tacticath, Abbott Technologies, Minneapolis, MN, USA) was used for Ensite Precision. Maximum RF energy delivery was 30–40 W with a temperature limit of 45°C; irrigation was 10–17 mL/min. Sheaths were irrigated with a solution of heparinized normal saline (1 U/mL) at a rate of 1 mL/min.
After completion of mapping and confirmation of the re-entrant mechanism, ablation was performed at the narrowest accessible critical isthmus based on voltage, activation, and entrainment mapping. Our diagnosis was considered correct if ablation led to AT termination (to SR or to another AT). We targeted the slowest conduction zone or the narrowest part of the active circuit between scars or anatomical structures. We started our ablation in regions displaying low voltage <0.5 mV based on our previous voltage mapping using multipolar catheters in agreement with activation and entrainment mapping (PPI-TCL < 20 ms). For macro re-entrant tachycardias, the narrowest pathway between two non-conducting obstacles of the re-entry course was targeted to interrupt the circuit by linear ablation. Localized re-entries were targeted by focal ablation. Interruption of AT during RF delivery without ectopy was used as confirmation of a critical site for re-entry. Ablation was continued after restoration of SR if necessary, to complete lines or connect ablation sites to non-conducting structures. The goal of RF ablation was creation of linear lesions within the zone of slow conduction, bridging the nearest anatomic–surgical barriers, or extension of the existing fixed lines of conduction block to reach large anatomic barriers (often either one of the PVs or mitral annulus) or scar regions due to previous ablation or substrate due to atrial cardiomyopathy.11 PV isolation was always performed. Our endpoint was to render the AT non-inducible during LA or right atrial (RA) burst-pacing and isoproterenol administration at the end of the procedure. Inducibility was tested in all patients exclusively by burst pacing from two different pacing sites (LA and RA), down to 220 ms (or the local refractory period, if >220 ms) and repeated 3 times. If the clinical AT converted to a different AT or in the event of induction of another AT, the ablation was continued aiming at termination to sinus rhythm. Inducibility was tested in all patients exclusively by burst pacing from two different pacing sites (LA and RA), down to 220 ms (or the local refractory period, if >220 ms) and repeated 3 times. If the clinical AT converted to a different AT or in the event of induction of another AT, the ablation was continued aiming at termination to sinus rhythm. Validation of ablation lines was performed using differential pacing or activation mapping during LAA or CS pacing.
Follow-up
Patients remained in hospital for 24 h after ablation. All patients underwent duplex sonography of the access site the day after the procedure to rule out possible vascular complications. Echocardiography to rule out pericardial effusion was performed at the end of the procedure and the next day. No antiarrhythmic drugs were continued 3 months after ablation. Oral anticoagulants were administered 3 h after catheter ablation and continued for 3 months or longer, based on CHA2DS2-VASc score. Definitions for bleeding and thromboembolic complications were used as previously described.12 Patients were followed up at 1, 3, 6, and 12 months, and every 6 months thereafter. A detailed history, electrocardiogram, and Holter monitoring were performed in symptomatic patients. Any documented atrial arrhythmia lasting >30 s was considered a recurrence.
Statistical analysis
Continuous variables were presented as mean values ± standard deviation, while categorical ones were presented as absolute and relative frequencies (percentages). Continues variables were tested for normal distribution using the Kolmogorov–Smirnov test. Continuous variables with and without normal distribution were compared using Student’s t-test/one-way ANOVA or the Mann–Whitney U-test/Kruskal–Wallis, respectively. Pearson’s χ2 or Fisher’s exact test was used to test for any associations between two categorical variables. Analyses were done with SPSS (version 22.0, SPSS Inc., Chicago, IL, USA) and all reported P-values are two-tailed with a significance level of 0.05.
Results
We included 107 out of 169 consecutive patients (mean age: 65.7 ± 9.2, males: 81 patients) with documented AT after LA ablation for persistent AF. We excluded 62 patients from our study based on the aforementioned exclusion criteria. The baseline characteristics of the patients are shown in Table 1. All patients had undergone at least one previous ablation procedure including PVI, either as a standard-alone procedure or combined with ablation of AT. The mean follow-up was 1.6 ± 0.5 years. Thirty-seven (34.6%) out of 107 patients had undergone only PVI, 41(38.3%) had undergone a stepwise EGM-guided ablation approach and 29 patients (27.1%) had undergone ECG-I based ablation targeting rotational activities. We analysed 134 AT in these 107 patients. All were re-entries with a clear endocardial gap due to previous ablation for persistent AF.
Clinical characteristics of 107 consecutive patients undergoing gap-related AT post-AF ablation
| Clinical variables | Mean± SD or n (%) |
|---|---|
| Age (years) | 65.7 ± 9.2 |
| Sex (males), n (%) | 81 (75.7) |
| CHADS2VASC score | 2.2 ± 1.3 |
| Hypertension, n (%) | 68 (63.6) |
| Diabetes mellitus, n (%) | 26 (24.3) |
| CAD, n (%) | 15 (14) |
| HF, n (%) | 59 (55.1) |
| LVEF (%) | 51 ± 5.4 |
| LAD (mm) | 42.5 ± 1.3 |
| β-Blockers | 96 (89.7%) |
| Amiodarone | 38 (35.5%) |
| Number of previous procedures | 2.3 ± 1.2 |
| Index procedure |
|
| Lines completed in previous AF ablation procedure |
|
| Post-cardiac surgery, n (%) | 5 (4.7) |
| Clinical variables | Mean± SD or n (%) |
|---|---|
| Age (years) | 65.7 ± 9.2 |
| Sex (males), n (%) | 81 (75.7) |
| CHADS2VASC score | 2.2 ± 1.3 |
| Hypertension, n (%) | 68 (63.6) |
| Diabetes mellitus, n (%) | 26 (24.3) |
| CAD, n (%) | 15 (14) |
| HF, n (%) | 59 (55.1) |
| LVEF (%) | 51 ± 5.4 |
| LAD (mm) | 42.5 ± 1.3 |
| β-Blockers | 96 (89.7%) |
| Amiodarone | 38 (35.5%) |
| Number of previous procedures | 2.3 ± 1.2 |
| Index procedure |
|
| Lines completed in previous AF ablation procedure |
|
| Post-cardiac surgery, n (%) | 5 (4.7) |
AT, atrial tachycardia; AF, atrial fibrillation; CHA2DS2VaSc, congestive heart failure, hypertension, age ≥75 (doubled), diabetes mellitus, prior stroke or transient ischaemic attack (doubled), vascular disease, age 65–74, female; CAD, coronary artery disease; HF, heart failure; LVEF, left ventricular ejection fraction; LAD, left atrial diameter.
Clinical characteristics of 107 consecutive patients undergoing gap-related AT post-AF ablation
| Clinical variables | Mean± SD or n (%) |
|---|---|
| Age (years) | 65.7 ± 9.2 |
| Sex (males), n (%) | 81 (75.7) |
| CHADS2VASC score | 2.2 ± 1.3 |
| Hypertension, n (%) | 68 (63.6) |
| Diabetes mellitus, n (%) | 26 (24.3) |
| CAD, n (%) | 15 (14) |
| HF, n (%) | 59 (55.1) |
| LVEF (%) | 51 ± 5.4 |
| LAD (mm) | 42.5 ± 1.3 |
| β-Blockers | 96 (89.7%) |
| Amiodarone | 38 (35.5%) |
| Number of previous procedures | 2.3 ± 1.2 |
| Index procedure |
|
| Lines completed in previous AF ablation procedure |
|
| Post-cardiac surgery, n (%) | 5 (4.7) |
| Clinical variables | Mean± SD or n (%) |
|---|---|
| Age (years) | 65.7 ± 9.2 |
| Sex (males), n (%) | 81 (75.7) |
| CHADS2VASC score | 2.2 ± 1.3 |
| Hypertension, n (%) | 68 (63.6) |
| Diabetes mellitus, n (%) | 26 (24.3) |
| CAD, n (%) | 15 (14) |
| HF, n (%) | 59 (55.1) |
| LVEF (%) | 51 ± 5.4 |
| LAD (mm) | 42.5 ± 1.3 |
| β-Blockers | 96 (89.7%) |
| Amiodarone | 38 (35.5%) |
| Number of previous procedures | 2.3 ± 1.2 |
| Index procedure |
|
| Lines completed in previous AF ablation procedure |
|
| Post-cardiac surgery, n (%) | 5 (4.7) |
AT, atrial tachycardia; AF, atrial fibrillation; CHA2DS2VaSc, congestive heart failure, hypertension, age ≥75 (doubled), diabetes mellitus, prior stroke or transient ischaemic attack (doubled), vascular disease, age 65–74, female; CAD, coronary artery disease; HF, heart failure; LVEF, left ventricular ejection fraction; LAD, left atrial diameter.
We excluded 62 patients (91 AT) from our study: (i) 41 AT (45%) for which we were unable to perform either activation or entrainment mapping and identify the mechanism of the arrhythmia or AT changed to another AT or terminated to SR and (ii) 4 biatrial AT (4.4%), (iii) 32 (35.1%) AT using epicardial connections, and (iv) 14 (15.4%) focal AT.
We analysed the electrophysiological characteristics of the termination sites of 94 (70.1% of total AT) macro-reentries using anatomical obstacles and 40 (29.9% of total AT) localized re-entries using scar regions formed either because of previous ablation or structural atrial cardiomyopathy.2 Regarding macro-reentries, we mapped and analysed 43 (45.8%) peri-mitral AT, 35 (37.2%) roof-dependent AT, 2 (2.1%) CTI-dependent AT, and 14 (14.9%) PV-related AT (Figure 1). Of the localized re-entries, we identified the mechanism as follows: 16 (40%) anterior localized re-entries (Figure 2), 5 posterior circuits (12.5%), 2 (5%) roof-anterior, 7 (17.5%) using the left septum, 5 (12.5%) inferior part of LA, 3 (7.5%) anterior-based LAA, 1 (2.5%) posterior RA due to scar from atrial cardiomyopathy or previous ablation or incomplete ablation of AF driver, and 1 (2.5%) localized re-entry in the anterior part of RA close to RAA due to scar from previous ablation.
Mapping using Pentaray mapping catheter (Carto System). PV-related re-entry AT using the RIPV in a patient presenting to our electrophysiological lab due to endocardial gap-related AT after PVI wide circumferential ablation for AF. Combining activation mapping, voltage, mapping, and entrainment mapping allowed us to identify the circuit delineating the macrorentry using the RPV, which had the posterior part of RIPV as an entrance, used the carina of the RPVs, and exited in the anterior part of the RIPV. We successfully terminated the arrhythmia in an ULVA exhibiting amplitude <0.1 mV in the anterior part of the RIPV close to the carina of the RPVs. The bipolar EGM at the termination site showed a high-frequency multiphasic fragmented electrogram of ultra-low amplitude of 0.08 mV with long duration of 95 ms. PPI-TCL was—5 ms using high output pacing (8–15 mA, duration 2 s) and 0 ms using low output pacing (3–7 mA, duration 1 s) during entrainment manoeuvers at the critical isthmus of the circuit. AT, atrial fibrillation; AF, atrial fibrillation; EGM, electrogram; ULVA, ultra-low-voltage area.
Mapping using Orion Catheter (Rhythmia mapping system). Localized re-entry in the anterior part of the LA due to endocardial gap-related AT post-AF ablation using the stepwise approach. Combining activation mapping, voltage mapping, and entrainment mapping allowed identification of the circuit and successful termination in an ultra-low voltage areas ULVA exhibiting amplitude <0.1 mV. The bipolar EGM at the termination site showed a high-frequency multiphasic fragmented electrogram of ultra-low amplitude of 0.03 mV with long duration of 92 ms. PPI-TCL was −8 ms using high output pacing (8–15 mA, duration 2 s) and 0 ms using low output pacing (3–7 mA, duration 1 s) during entrainment at the critical isthmus of the circuit. AT, atrial fibrillation; AF, atrial fibrillation; EGM, electrogram; ULVA, ultra-low-voltage area.
Regarding the amplitude of the fragmented bipolar signals at the termination site, multipolar catheters showed significantly higher amplitudes compared to the ablation catheter (0.14 ± 0.095 vs. 0.11 ± 0.095 mV, P < 0.001) (Table 2). Multipolar catheters showed significantly longer fragmented bipolar EGM durations at the termination site compared to the ablation catheter (101.2 ± 30.1 vs. 94.6 ± 25.4 ms, P < 0.001). There was no statistically significant difference in the amplitude of the fragmented signal at the termination sites using different multipolar catheters (Orion vs. HDGrid 0.140 ± 0.095 vs. 0.14 ± 0.88 mV P = 0.59; Orion vs. Pentarray 0.140 ± 0.095 vs. 0.16 ± 0.13 mV, P = 0.85; and HD Grid vs. Pentaray 0.14 ± 0.088 vs. 0.16 ± 0.13 mV, P = 0.93). There was no significant difference in the longest duration of fragmented bipolar signals at the termination sites between multipolar catheters (Orion catheter vs. HD Grid 102.8 ± 29.8 vs. 99.1 ± 32.8 ms, P = 0.40; Orion vs Pentaray 102.8 ± 29.8 vs. 86.5 ± 18.7 ms, P = 0.19; and HD Grid vs. Pentaray 99.1 ± 32.8 vs. 86.5 ± 18.7 ms, P = 0.32). Specifically, in macro-reentrant AT the mean amplitude of the longest fragmented bipolar signal at the termination site were 0.13 ± 0.08 vs. 0.1 ± 0.07 mV, P = 0.002 for multipolar and conventional catheters, respectively, and mean durations were 99.2 ± 30.3 vs. 92.7 ± 25.5 ms, P = 0.07. In localized re-entries, the mean amplitude of the longest fragmented bipolar signal at the termination site was 0.16 ± 0.12 vs. 0.13 ± 0.14 mV, P = 0.06 using multipolar and conventional ablation catheters respectively, and mean durations were 105.8 ± 29.6 vs. 98.9 ± 24.9 ms, P = 0.29.
Electrophysiological characteristics of endocardial gap-related AT post-AF ablation
| Multipolar catheters | Orion 97, 72.4% | HDGrid 31, 23.1% | Pentaray 6, 4.5% | P-value | |
|---|---|---|---|---|---|
| AT CL (ms) | 291 ± 73 | 284.6 ± 76.8 | 307.3 ± 58.7 | 309.7 ± 68.4 | 0.03a |
| Number of points | 19 704 ± 9343 | 18 546 ± 9137 | 25 054 ± 7615 | 10 781 ± 8475 | <0.01a,b,c |
| Number of macro-reentries | 94 (45 peri-mitral, 31 roof, 14 PV-related, 2 CTI, 2 double loop) | 67 (37 peri-mitral, 26 roof, 4 PVs) | 24 (7 peri-mitral, 5 roof, 2 CTI, 9 PVs, 1 double loop) | 3 (1 peri-mitral, 1 PVs, 1 double loop) | – |
| Number of localized re-entries | 40 | 30 | 7 | 3 | – |
| Amplitude of fragmented potential at termination site (mV) | 0.14 ± 0.095 | 0.14 ± 0.1 | 0.14 ± 0.09 | 0.16 ± 0.13 | 0.86 |
| Duration of fragmented potential at termination site (ms) | 101.2 ± 30.1 | 102.8 ± 29.8 | 99.1 ± 32.8 | 86.5 ± 18.7 | 0.31 |
| PPI-TCL at termination site using high (10–15 mA , 2 s) vs. low output pacing (3–7 mA , 1 s) (ms) | High output: −6.08 ± 3.4 Low output: 2.64 ± 3.79 | High output: −5.51 ± 3.19 Low output: 2.73 ± 3.73 | High output: −7.74 ± 3.62 Low output: 2.23 ± 4.19 | High output: −7.67 ± 3.14 Low output: 2.5 ± 4.32 | High output: 0.002* Low output: 0.18 |
| Mean ablation time for termination (s) | 34.6 ± 53.6 | 42.9 ± 60.9 | 13.5 ± 8.0 | 10.7 ± 6.5 | 0.02** |
| Multipolar catheters | Orion 97, 72.4% | HDGrid 31, 23.1% | Pentaray 6, 4.5% | P-value | |
|---|---|---|---|---|---|
| AT CL (ms) | 291 ± 73 | 284.6 ± 76.8 | 307.3 ± 58.7 | 309.7 ± 68.4 | 0.03a |
| Number of points | 19 704 ± 9343 | 18 546 ± 9137 | 25 054 ± 7615 | 10 781 ± 8475 | <0.01a,b,c |
| Number of macro-reentries | 94 (45 peri-mitral, 31 roof, 14 PV-related, 2 CTI, 2 double loop) | 67 (37 peri-mitral, 26 roof, 4 PVs) | 24 (7 peri-mitral, 5 roof, 2 CTI, 9 PVs, 1 double loop) | 3 (1 peri-mitral, 1 PVs, 1 double loop) | – |
| Number of localized re-entries | 40 | 30 | 7 | 3 | – |
| Amplitude of fragmented potential at termination site (mV) | 0.14 ± 0.095 | 0.14 ± 0.1 | 0.14 ± 0.09 | 0.16 ± 0.13 | 0.86 |
| Duration of fragmented potential at termination site (ms) | 101.2 ± 30.1 | 102.8 ± 29.8 | 99.1 ± 32.8 | 86.5 ± 18.7 | 0.31 |
| PPI-TCL at termination site using high (10–15 mA , 2 s) vs. low output pacing (3–7 mA , 1 s) (ms) | High output: −6.08 ± 3.4 Low output: 2.64 ± 3.79 | High output: −5.51 ± 3.19 Low output: 2.73 ± 3.73 | High output: −7.74 ± 3.62 Low output: 2.23 ± 4.19 | High output: −7.67 ± 3.14 Low output: 2.5 ± 4.32 | High output: 0.002* Low output: 0.18 |
| Mean ablation time for termination (s) | 34.6 ± 53.6 | 42.9 ± 60.9 | 13.5 ± 8.0 | 10.7 ± 6.5 | 0.02** |
All continuous variables are displayed as mean ± SD. Categorical data are displayed as counts.
AF, atrial fibrillation; AT, atrial tachycardia.
A statistically significant difference was observed in the mean cycle length (P = 0.01), mean number of points (P < 0.01).
A statistically significant difference in the mean number of points was observed between Orion and Pentaray catheters (P = 0.04).
Statistically significant difference in the mean number of points was observed between HDGrid and Pentaray catheters (P < 0.001).
Statistically significant difference betweem Orion and HDGrid catheters.
Electrophysiological characteristics of endocardial gap-related AT post-AF ablation
| Multipolar catheters | Orion 97, 72.4% | HDGrid 31, 23.1% | Pentaray 6, 4.5% | P-value | |
|---|---|---|---|---|---|
| AT CL (ms) | 291 ± 73 | 284.6 ± 76.8 | 307.3 ± 58.7 | 309.7 ± 68.4 | 0.03a |
| Number of points | 19 704 ± 9343 | 18 546 ± 9137 | 25 054 ± 7615 | 10 781 ± 8475 | <0.01a,b,c |
| Number of macro-reentries | 94 (45 peri-mitral, 31 roof, 14 PV-related, 2 CTI, 2 double loop) | 67 (37 peri-mitral, 26 roof, 4 PVs) | 24 (7 peri-mitral, 5 roof, 2 CTI, 9 PVs, 1 double loop) | 3 (1 peri-mitral, 1 PVs, 1 double loop) | – |
| Number of localized re-entries | 40 | 30 | 7 | 3 | – |
| Amplitude of fragmented potential at termination site (mV) | 0.14 ± 0.095 | 0.14 ± 0.1 | 0.14 ± 0.09 | 0.16 ± 0.13 | 0.86 |
| Duration of fragmented potential at termination site (ms) | 101.2 ± 30.1 | 102.8 ± 29.8 | 99.1 ± 32.8 | 86.5 ± 18.7 | 0.31 |
| PPI-TCL at termination site using high (10–15 mA , 2 s) vs. low output pacing (3–7 mA , 1 s) (ms) | High output: −6.08 ± 3.4 Low output: 2.64 ± 3.79 | High output: −5.51 ± 3.19 Low output: 2.73 ± 3.73 | High output: −7.74 ± 3.62 Low output: 2.23 ± 4.19 | High output: −7.67 ± 3.14 Low output: 2.5 ± 4.32 | High output: 0.002* Low output: 0.18 |
| Mean ablation time for termination (s) | 34.6 ± 53.6 | 42.9 ± 60.9 | 13.5 ± 8.0 | 10.7 ± 6.5 | 0.02** |
| Multipolar catheters | Orion 97, 72.4% | HDGrid 31, 23.1% | Pentaray 6, 4.5% | P-value | |
|---|---|---|---|---|---|
| AT CL (ms) | 291 ± 73 | 284.6 ± 76.8 | 307.3 ± 58.7 | 309.7 ± 68.4 | 0.03a |
| Number of points | 19 704 ± 9343 | 18 546 ± 9137 | 25 054 ± 7615 | 10 781 ± 8475 | <0.01a,b,c |
| Number of macro-reentries | 94 (45 peri-mitral, 31 roof, 14 PV-related, 2 CTI, 2 double loop) | 67 (37 peri-mitral, 26 roof, 4 PVs) | 24 (7 peri-mitral, 5 roof, 2 CTI, 9 PVs, 1 double loop) | 3 (1 peri-mitral, 1 PVs, 1 double loop) | – |
| Number of localized re-entries | 40 | 30 | 7 | 3 | – |
| Amplitude of fragmented potential at termination site (mV) | 0.14 ± 0.095 | 0.14 ± 0.1 | 0.14 ± 0.09 | 0.16 ± 0.13 | 0.86 |
| Duration of fragmented potential at termination site (ms) | 101.2 ± 30.1 | 102.8 ± 29.8 | 99.1 ± 32.8 | 86.5 ± 18.7 | 0.31 |
| PPI-TCL at termination site using high (10–15 mA , 2 s) vs. low output pacing (3–7 mA , 1 s) (ms) | High output: −6.08 ± 3.4 Low output: 2.64 ± 3.79 | High output: −5.51 ± 3.19 Low output: 2.73 ± 3.73 | High output: −7.74 ± 3.62 Low output: 2.23 ± 4.19 | High output: −7.67 ± 3.14 Low output: 2.5 ± 4.32 | High output: 0.002* Low output: 0.18 |
| Mean ablation time for termination (s) | 34.6 ± 53.6 | 42.9 ± 60.9 | 13.5 ± 8.0 | 10.7 ± 6.5 | 0.02** |
All continuous variables are displayed as mean ± SD. Categorical data are displayed as counts.
AF, atrial fibrillation; AT, atrial tachycardia.
A statistically significant difference was observed in the mean cycle length (P = 0.01), mean number of points (P < 0.01).
A statistically significant difference in the mean number of points was observed between Orion and Pentaray catheters (P = 0.04).
Statistically significant difference in the mean number of points was observed between HDGrid and Pentaray catheters (P < 0.001).
Statistically significant difference betweem Orion and HDGrid catheters.
Entrainment at termination sites exhibiting low-voltage amplitude, at a pacing CL 10–20 ms shorter than TCL, gave the following results: mean PPI-TCL −6.08 ± 3.4 ms using high output (HOP, 8–15 mA, duration 2 s) and mean PPI-TCL 2.64 ± 3.79 ms using low-output (LOP, 3–7 mA, duration 1 s) (P < 0.01). We were able to entrain at the termination site in all 107 AT (100%), confirming the presence of viable tissue in the critical isthmus of the re-entry circuit.
Mapping using HDGrid mapping catheter (Ensite Precision). (A) A counterclockwise mitral isthmus dependent AT due to an endocardial gap-related AT post-AF ablation using EGM-guided (ECG-i) ablation targeting rotational activities. Combining activation mapping, voltage mapping, and entrainment mapping allowed us to identify the circuit and successfully terminate the arrhythmia in a low-voltage region with amplitude <0.2 mV. The bipolar EGM using a HDGrid mapping catheter at the termination site recorded a high-frequency multiphasic fragmented electrogram of ultra-low amplitude of 0.166 mV with long duration of 152 ms. PPI-TCL was—6 ms using high output pacing (8–15 mA, duration 2 s) and 2 ms using low output pacing (3–7 mA, duration 1 s) during entrainment at the critical isthmus of the circuit. (B) A conventional ablation catheter (Tacticath, Abbott Technologies, 2‐2‐2 mm interelectrode spacing) did not detect any signal (bipolar EGM amplitude <0.03 mV, at the limit of the definition of noise) in the same region despite contact force of 20 g. AT, atrial fibrillation; AF, atrial fibrillation; EGM, electrogram.
Interestingly, 133 of 134 AT (99.3%) were terminated in low-voltage areas (LVA) with amplitude <0.5 mV measured using multipolar catheters and also 133 of 134 AT (99.3%) were terminated in LVA with amplitude <0.5 mV measured using a conventional ablation catheter. Fifty-five out of 134 AT (41%) were terminated in ultra-LVA (ULVA) with an amplitude <0.1 mV measured using multipolar catheters and 79 (59%) were terminated in ULVA with an amplitude <0.1 mV measured using a conventional ablation catheter (Figure 3). Thirty-one AT (29%) were terminated in regions with an amplitude ≤0.03 mV at the limit of the definition of noise using conventional ablation catheters. Only multipolar catheters could delineate clear fractionated low-amplitude and high frequency signals, critical for the maintenance of re-entry.
Mean ablation time for arrhythmia termination was 34.6 ± 53.6 s; 42.9 ± 60.9 s for the cases mapped using Orion (97 AT), 13.5 ± 8.0 s for HDGrid (31 AT), and 10.7 ± 6.5 s for Pentaray (6 AT). Mean procedure time was 214 ± 90 min, mean total ablation time was 35 ± 57 min, and mean fluoroscopy time 28 ± 20 min.
The positive predictive value of activation mapping identifying of gap-related AT post-AF-ablation and distinguishing the active of the passive part of the circuit was 83.6%: 84.5% for Rhythmia, 77.4% for HD Grid, and 100% for Pentaray.
The positive predictive value of activation mapping identifying of gap-related AT post-AF-ablation and distinguishing the active of the passive part of the circuit was 83.6%: 84.5% for Rhythmia, 77.4% for HD Grid, and 100% for Pentaray. As we have already mentioned, the total number of AT of our research protocol was 225. Of them, 41 (18.2%) AT were unmappable. We performed a diagnosis using only activation mapping in 152 out of 225 AT (67.6%). When we combined activation mapping and entrainment mapping we identified the critical isthmus of the clinical AT in 184 out of 225 AT (81.8%). We improved our accuracy analysing only endocardial gap-related AT after AF ablation, excluding 50 AT (41 unmappable, 4 biatrial, 32 AT using epicardial connections, and 14 focal AT). By excluding 50 AT with the reasons that are mentioned in the methods section and the 41 unmappable AT, the total number of AT was 134. We performed a diagnosis in 112 out of 134 AT (83.6%) using only activation mapping and in 134 out of 134 AT (100%) using the combination of activation and entrainment mapping (Figure 4).
![Graphical presentation of AT diagnosis [(A) total AT and (B) endocardial gap-related AT] using only activation mapping or combining activation and entrainment mapping. AT, atrial fibrillation](https://oup.silverchair-cdn.com/oup/backfile/Content_public/Journal/europace/23/7/10.1093_europace_euaa394/3/m_euaa394f4.jpeg?Expires=1747860884&Signature=oQV5urI9Y-ezxZ8UmZtcblulOQ3UL5vOlT-Cti84DhtGOSd8nxg9m6oxLhVEhSCOigALfSuEkzQHkj~m183~3JaLcsG7zo~vJd~uCCondgxVhvtCcWFZG2M-1-1ZABH4U0IPN6tfnuWIwAi7Lp2WcAbcRdYMn~EkI4cqhhTnpwkRFI0aJTAZJAf-Ea-uNW0JRMQcaJ2qW0rqFsu3IYdtGEy4gcpJ8L1xb9QmrW2iEHr2yDbgQri7j3IrwszNNXlK6IplYIPj4tqqHaxwebTUoGnrX3AlZpTmbdUNAjmUOGNVyn7PX7IW6JLneyUwLoIadeC21ZIcKoBHmIhBdROIXg__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
Graphical presentation of AT diagnosis [(A) total AT and (B) endocardial gap-related AT] using only activation mapping or combining activation and entrainment mapping. AT, atrial fibrillation
Complications
There were no major periprocedural complications. One patient (0.9%) had a femoral arterio-venous fistula not requiring intervention and two patients (1.8%) had pericardial effusion <1 cm not requiring drainage.
Long-term follow-up
Of the 107 patients, 17 (15.9%) presented with atrial arrhythmia recurrence in a mean follow-up of 16.4 ± 3.2 months. Of the 62 patients, 15 (24.1%) excluded from our study (unmappable AT, biatrial AT, AT using epicardial connections, and focal AT) exhibited an atrial arrhythmia (AT/AF) recurrence in a mean follow-up of 16.4 ± 3.2 months. Seven patients (6.5%) experienced mitral isthmus macroreentrant AT using epicardial connections either of the Marshall Bundle or CS musculature, successfully terminated in the redo procedure after alcohol injection inside the Vein of Marshall and RF ablation inside the CS. Five patients (4.7%) presented with roof dependent AT successfully ablated and terminated at the endocardial roof gap. Three patients (2.8%) presented with localized re-entries: one (0.9%) anterior localized re-entry in the LA, one (0.9%) localized re-entry in the infero-posterior part of LA, and one (0.9%) localized re-entry at the RA lateral wall. All three localized re-entries were successfully ablated after identifying the critical isthmus using the same protocol of activation, entrainment, and voltage mapping. Two patients (1.9%) presented with episodes of AF where we decided not to perform a redo procedure and continue with medical therapy. Of the 62 excluded patients, 15 (24.1%) (unmappable AT, biatrial AT, AT using epicardial connections, and focal AT) exhibited an atrial arrhythmia (AT/AF) recurrence in a mean follow-up of 16.4 ± 3.2 months.
Discussion
The main findings of our study are
High-resolution activation mapping combined with high-density voltage and entrainment mapping is the ideal strategy for ablation delineating the critical part of the circuit in endocardial gap-related re-entrant AT after AF ablation.
Voltage in the conducting channels may be extremely low, even <0.1 mV (0.14±0.095 mV, 51 out of 134 AT, 41%), and almost always (133 out of 134 AT, 99.3%) <0.5 mV (0.03–0.5 mV).
Multipolar catheters improved our accuracy in delineating ULVA critical for the re-entry maintenance of endocardial gap-related AT. Multipolar catheters showed significantly higher amplitude values of the fragmented bipolar signals at the termination site, compared with the ablation catheter. Conventional ablation catheters often do not detect signals, and only multipolar catheters of small electrodes and shorter interelectrode space can detect clear fractionation of low-amplitude and high frequency signals that are critical for re-entry maintenance.
Termination sites were linked with negative PPI-TCL (−6.08±3.4 ms) using HOP and PPI-TCL<20 ms using LOP during entrainment, confirming that they correspond to viable tissue in LVA critical for maintenance of the circuit.
A small cluster of ablation lesions targeting these LVA (<0.5 mV) or even ULVA (<0.1 mV) sites exhibiting PPI-TCL<20 ms using LOP and negative PPI-TCL using HOP during entrainment in this critical part of the isthmus terminates the arrhythmia.
The main limitation of previous studies has been the use of 3.5 mm irrigated tip ablation catheters for mapping complex AT after extensive ablation, which increases far-field recording and reduces accuracy, in comparison with using multipolar catheters with short interspaced electrode. Abnormal EGMs can be detected only by catheters with small electrodes and shorter interelectrode distance, which allow for higher near-field resolution, thereby allowing detailed characterization of the atrial substrate, including multicomponent fragmented signals in ULVA <0.1 mV.13 The ability to collect at least one EGM per mm2 identifies certain critical regions of the isthmus. Previous studies comparing multielectrode mapping catheters with linear ablation catheters in a model of healed infarction, in agreement with our study, found that multielectrode catheters were able to identify surviving myocardial bundles that were not recognized by ablation catheters.14,15
Detection of viable atrial tissue in low-voltage regions closely related with atrial scar
In our study, we were able to characterize the electrophysiological characteristics at the termination sites of endocardial gap-related AT post-AF ablation. Multipolar catheters improved activation and voltage mapping, thus delineating LVA critical for the re-entry circuit. We confirmed the existence of viable tissue in these critical ULVA (i) with entrainment mapping and (ii) by terminating the arrhythmias in these regions. The termination sites were almost always (133 out of 134 AT, 99.3%) in LVA (0.03–0.5 mV). Voltage in the conducting channels may be extremely low, even <0.1 mV (0.14 ± 0.095 mV, 51 out of 134 AT, 41%), and almost always <0.5 mV. The use of multipolar catheters allowed identification of these ULVA critical for the maintenance of the circuit. It is important while mapping a scar-related AT, to detect viable tissue within the scar, as it may represent the critical isthmus.16 The lower threshold of recordable physiological electric signals will be represented by the amplitude of electronic noise, which should be minimized in order to improve the signal/noise ratio.
The morphology and amplitude of the recorded EGM depends on myocardial properties, wavefront direction, conducting medium, catheter-tissue contact and orientation, catheter electrodes size, composition, shape, and inter-electrode spacing. Relative voltage preservation within denser regions of scar is a hallmark of central conducting channels that may form anatomically constrained isthmuses during AT. Combining high-density voltage, activation, and entrainment mapping using multipolar catheters, we could identify the critical isthmus in gap-related AT. A small cluster of ablations at these ULVA exhibiting PPI-TCL < 20 ms using LOP and negative PPI-TCL using HOP during entrainment in this critical part of the isthmus terminated the arrhythmia. Mean ablation time for the arrhythmia termination was 34.6 ± 53.6 s, showing the efficacy of our strategy. Atrial tachycardia termination is a desirable intermediate endpoint and important in proving that the target isthmus is critical for maintaining the re-entry circuit. However, completion of an adequate line of bidirectional block connecting the critical isthmus with anatomical obstacles or dense scar regions is the optimal final endpoint in order not to have further recurrence in long-term follow-up.
Extensive discussion regarding the advantages of multipolar catheters (Orion, HDGrid, and Pentaray) is provided in Supplementary material online.
The value of entrainment mapping
Entrainment mapping confirmed the results of activation mapping, identifying active circuits and differentiating them from dead-end channels. Activation mapping of AT has been facilitated by the use of electroanatomic mapping systems. The advent of high-density mapping has allowed generation of electroanatomic maps with a high-resolution level.2,17 However, these highly detailed activation maps are difficult to interpret because they include both active and passive circuits. In this setting, entrainment mapping is a powerful tool to identify which parts of the atria are directly involved in the re-entrant circuit.18–20 In agreement with our study, Johner et al.18 recently demonstrated that: (i) PPI-TCL < 0 ms is more frequently located within limited width critical isthmuses compared with sites with PPI-TCL = 0–30 ms, (ii) sites with PPI-TCL < 0 ms exhibit slower local conduction velocity and lower EGM voltages than sites with PPI-TCL = 0–30 ms, and (iii) sites with PPI-TCL > 30 ms are rarely found within the re-entry circuit or within isthmuses.
Although slow conduction around, but removed from, the entrained site is expected to prolong PPI-TCL when pacing outside the tachycardia circuit, slow conduction should have the opposite or no effect when pacing within the critical isthmus or close to the site of slow conduction, by amplifying the effect of far-field capture on the duration of the return cycle. Entrainment using HOP results in synchronous capture of an area of tissue around the tip of the catheter. The resulting return cycle is shorter than the TCL because the area of slow conduction of the critical isthmus and active part of the circuit is effectively bypassed. The slower the conduction velocity in this region, the more time is removed from the TCL, the more negative the PPI-TCL. Entrainment with LOP likely results in a smaller area of capture, thus a longer return cycle.
Our study shows that PPI-TCL < 0 ms sites are not only more commonly located in the tachycardia circuit but also markers of narrow isthmuses and slow conduction. The key concept that arises from this finding is that PPI-TCL is not only a function of the distance between the entrained site and the leading wavefront, but is also dependent on local electroanatomical properties of the entrained site. These properties suggest that PPI-TCL < 0 ms using HOP can be used as a parameter to identify clinically relevant critical isthmuses and potential ablation targets better than standard entrainment criteria. In endocardial gap-related AT, sites exhibiting PPI-TCL < 0 ms using HOP and PPI-TCL < 20 ms using LOP are markers of narrow critical isthmuses with slower conduction velocity, whereas a significant proportion of sites with PPI-TCL = 0–30 ms are not always in close proximity to the re-entry circuit.
Detailed activation maps are difficult to interpret because they include the activation of both active and passive circuits. In this setting, entrainment mapping is a powerful tool to identify which parts of the atria are actively involved in the re-entry circuit2,20 (Figure 5). The positive predictive value of activation mapping identifying of gap-related AT post-AF-ablation and distinguishing the active of the passive part of the circuit was 83.6%.
Combining activation and entrainment mapping. Clockwise perimitral re-entry gap-related AT after PVI wide circumferential ablation and electrogram-based ablation (Complex fractionated atrial EGM plus lines) for AF. Combining activation mapping, voltage mapping, and entrainment mapping (PPI-TCL around mitral isthmus and mid-CS<20 ms) allowed us to identify the active part of the circuit delineating the macro-reentry using mitral isthmus. Activation mapping in combination with entrainment mapping delineated the passive part of the AT turning around LPVs (PPI-TCL LPVs>30 ms). We successfully terminated the arrhythmia in an LVA exhibiting amplitude <0.5 mV in the posterior part of the mitral isthmus. The bipolar EGM at the termination site showed a high-frequency multiphasic fragmented electrogram of low amplitude of 0.31 mV with duration of 75 ms. PPI-TCL was—11 ms using high output pacing (8–15 mA, duration 2 s) and 0 ms using low output pacing (3–7 mA, duration 1 s) during entrainment manoeuvers at the critical isthmus of the circuit. AT, atrial fibrillation; AF, atrial fibrillation; EGM, electrogram.
Clinical implications
The main finding of our study is that in 29% of AT (n = 31) was impossible to precisely identify the critical site of arrhythmia conduction using only a standard ablation catheter, owing to the absence of a clearly detectable local signal. In these AT, termination occurred in regions with amplitude ≤0.03 mV, at the limit of the definition of noise using a conventional ablation catheter. Multipolar catheters with small closely spaced electrodes could detect clear fractionated low-amplitude and high frequency signals, critical for re-entry maintenance. The presence of double potentials during AT may represent only areas of functional conduction block. Atrial tachycardia termination is a desirable intermediate Endpoint and important in proving that the target isthmus was critical for maintaining the re-entry circuit. However, the optimal endpoint is completion of an adequate line of bidirectional block connecting the critical isthmus with anatomical obstacles or dense scar regions either from previous ablation or atrial cardiomyopathy. Most recurrences of flutter are due to either incomplete assessment of block or true re-connection. Differential pacing from both sides of the line in combination with voltage and activation mapping using high-resolution mapping should form the endpoints in order to confirm bidirectional block of the lines and result in durable lesions with long-term successful results (Figure 6).
An activation map during LAA pacing using the Orion catheter to determine whether there is conduction or block across the posterior mitral isthmus line after ablation of a perimitral AT. Using 0.05 mV as a cut-off to distinguish atrial scar from viable tissue, we cannot be sure of a possible endocardial connection in the posterior mitral isthmus line. However, using 0.03 mV as a cut-off allows us to detect and analyse bipolar EGMs of ultra-low-voltage (0.025 mV) and identify the critical endocardial connection-gap in the posterior mitral isthmus. Differential pacing from both sides of the line in combination with voltage and activation mapping using high-resolution mapping were our endpoints to confirm bidirectional block of the lines and result in durable lesions with long-term successful results. AT, atrial fibrillation; EGM, electrogram.
Limitations
The main limitation of our study is that voltage mapping, EGM morphology, and amplitude are influenced by multiple factors. The morphology and amplitude of the recorded EGM depends on myocardial properties, wavefront direction, conducting medium, catheter-tissue contact and orientation, catheter electrodes size, composition, shape, and inter-electrode spacing. Nevertheless, previous studies have shown that mapping with small closely spaced electrodes can improve mapping resolution, which is of critical importance within areas of low voltage.13 In the case of reduced conduction velocity (LVA), the difference between the times at which the wave arrives at the electrodes increases; however, at the same time, the magnitude of signal overlap between them also increases, which conversely results in a larger degree of cancellation and an even further reduced voltage amplitude.21–23 In our study, multipolar catheters produced higher voltages of the virtual bipolar EGMs in comparison with conventional ablation catheters, allowing us to map ULVA critical for gap-related AT. Universal voltage cut-off values may not be appropriate for bipolar voltage-guided substrate mapping. We need further studies evaluating the effect of activation wavefront and CL on bipolar voltage amplitude, identifying the cut-off values for each specific catheter.
Conclusion
High-resolution activation mapping combined with high-density voltage and entrainment mapping is the ideal strategy for ablation delineating the critical part of the circuit in endocardial gap-related re-entrant AT after AF ablation. Using multipolar catheters with small interlectrode spacing improves the accuracy in delineating ULVA critical for the re-entry circuit.
Supplementary material
Supplementary material is available at Europace online.
Conflict of interest: none declared.
Data availability
The data underlying this article cannot be shared publicly due to the privacy of individuals that participated in the study. The data will be shared on reasonable request to the corresponding author.
