High‐density mapping of de novo focal atrial tachycardias using a new software: Protected low‐voltage areas by zones of conduction delay

Abstract Background The pathophysiological mechanism of focal atrial tachycardias (AT) remains obscure. Methods Fifteen patients (6 males, age 45 ± 18) with focal AT underwent high‐density activation mapping using a new software called extended early‐meets‐late (EEML). Results Irrespective of the arrhythmia mechanism, low bipolar voltage fractionated signals (0.14 ± 0.10 mV) were seen at the earliest activation site. The mean low‐voltage area (LVA) at the earliest activation site was 3.2 ± 1.0 cm2. EEML mapping revealed zones of conduction delay at the borders of LVAs. Conclusions LVAs protected by zones of slow conduction appears to play an important role in the initiation and maintenance of focal ATs.


| INTRODUC TI ON
Focal atrial tachycardias (AT) are characterized by a centrifugal atrial activation pattern irrespective of arrhythmia mechanism (microreentry, triggered activity, and enhanced automaticity). 1 However, the pathophysiology of focal ATs is not fully elucidated. This study aimed to investigate the substrate of focal ATs during high-density mapping using a novel software program.

| PATIENTS AND ME THODS
The study population consisted of 15 patients (10 males, age 45 ± 18 years) with an established diagnosis of focal atrial tachycardia based on electrophysiological criteria. 2 None of these patients exhibited a history of atrial fibrillation or had undergone previous ablation procedure. The mechanism of AT was established based on certain criteria. The diagnosis of microreentrant AT was set up when activation mapping covered >70% of the tachycardia cycle length (TCL). Triggered activity was diagnosed when AT was inducible and terminated with programmed stimulation, and <50% of the TCL was mapped at the earliest activation site. Enhanced automaticity was diagnosed when AT was not inducible with programmed stimulation, and induction was achieved by isoproterenol infusion. Similar to triggered activity, <50% of the TCL was mapped at the earliest activation site. Adenosine testing was performed in selected cases.
The electrophysiological study was performed in fasting state without sedation. High-density mapping of the atria was performed during AT to localize the site of origin using a three-dimensional nonfluoroscopic mapping system (CARTO 3 Version 6.0, Biosense-Webster). A new software of the mapping system [high-definition (HD) coloring] identifies areas of potential conduction delay/block. The upper threshold of the early-meets-late tool is the same as in the previous software versions and is commonly used for mapping of macro-or microreentrant tachycardias (default upper threshold of 75%). A lower threshold has been added in CARTO 3 Version 6.0 system, called extended early-meets-late (EEML), which divides interpolated areas with significant activation time differences. EEML calculates the local activation difference between adjacent points, and if this difference is higher than the percentage selected (default lower threshold of 25%) of the cycle length mapped a white line will be drawn between those adjacent points indicating conduction delay or block. The system will highlight anywhere on the map where one area has at least a 25% difference from one color interpolation to another. Of note, the EEML is not dependent on the actual activation values themselves, but on the relative differences

| RE SULTS
The clinical characteristics, echocardiographic data, medications, and procedural data of the study cohort are depicted in Table 1.
The underlying mechanism of the centrifugal ATs was triggered activity (n = 9), microreentry (n = 3), and enhanced automaticity (n = 3). The sites of origin included crista terminalis in 7, lateral right atrium in 2, ostium of the coronary sinus in 2, and left atrial septum in four patients. Adenosine testing when performed terminated triggered activity-related ATs (n = 4), while no termination was seen in microreentrant-related ATs (n = 2). Irrespective of the arrhythmia mechanism, very low-voltage bipolar signals (0.14 ± 0.10 mV) with or without fractionation were mapped at the earliest activation site.
The mean low-voltage area was 3.2 ± 1.0 cm 2 . Apart from these low-voltage areas at earliest activation site, the majority of patients (13/15) displayed normal electroanatomical bipolar voltage maps of the right or left atrium. All patients with microreentrant ATs demonstrated fractionated electrograms at the successful ablation site covering ≥70% of the TCL. EEML mapping revealed zones of conduction slowing/block (protected isthmus) around the earliest activation site in all subjects irrespective of arrhythmia mechanism. The zones of conduction slowing were located at the borders of the low-voltage areas of the earliest activation site. A preferential conduction channel was evident in isochronal mapping in all cases (Figures 1 and 2).
During a mean follow-up of 7 ± 4 months, 14 of 15 patients were free of arrhythmias.

| D ISCUSS I ON
In this high-density mapping study using a new software program, we demonstrated that focal ATs arise from low-voltage areas that are protected by zones of conduction delay.
Earlier mapping studies with low sampling points have introduced the concept of "diseased areas" with preferential conduction TA B L E 1 Clinical characteristics, echocardiographic data, medications, and procedural data of the study cohort (n = 15)  The very low-voltage areas of the foci possibly represent diseased fibrotic atrial tissue. Spach and Josephson reported reentrant circuits in humans that occur in very small areas (microreentry) because of nonuniform anisotropic conduction associated with microfibrosis. 7 The zones of slow conduction at the borders of AT foci might be either functional or anatomical. Functional conduction delay is related to specific electrophysiological properties of the atrial myocardium involving dispersion of excitability or refractoriness as well as anisotropic conduction. 8 Crista terminalis, a common location site of ATs, has been shown to display areas of prominent anisotropy with slow conduction properties. 9 A preferential conduction that AT activation wavefronts pass through has been additionally demonstrated along or across the CT. Anatomic obstacles and/ or anisotropic conduction is the most plausible explanation of these preferential conduction pathways. 9 In conclusion, using EEML software, this high-density mapping study demonstrated that focal ATs display a specific electroanatomical pattern involving a protected diseased atrial tissue by zones of slow conduction along with preferential conduction channels at the exit site. The thresholds of EEML software for the accurate detection of conduction blocks remain to be validated in further studies. An individual adjustment of the EEML thresholds may be applied for a more accurate visualization of the zones of conduction block.

CO N FLI C T O F I NTE R E S T
None to declare.