A new interpretation of nonpulmonary vein substrates of the left atrium in patients with atrial fibrillation

Abstract Background Substrate analysis of the left atrium in patients undergoing atrial fibrillation ablation has limitations when performed by means of simple bipolar acquisition. Objective To evaluate the incidence of low‐voltage zones (LVZs) through maps constructed by means of various catheters: multipolar (MC), omnipolar (OC), and circular catheters (CMCs) with the 3D electro‐anatomical systems (3d‐S) CARTO3 and EnSite Precision. Methods To assess LVZs, we acquired maps by means of CMC and MC in the voltage range 0.05‐0.5 mV in 70 consecutive patients in sinus rhythm. In the case of OC only, we made an intra‐patient comparison of bipolar maps constructed by means of the along and across, and HD‐Wave configurations of the EnSite 3d‐S in the ranges of 0.05‐0.5 and 0.5‐1.0 mV. On the basis of this comparison, we chose the range that best identified LVZs as a set of different colors (SDC) compatible with patchy fibrosis (qualitative analysis). Subsequently, we detected the voltage values corresponding to purple and gray points, close to SDC, and the value inside corresponding to blue, green, and red colors, and we evaluated the color change in other voltage ranges. Finally, we performed a quantitative analysis of LVZs by applying the qualitative characteristics described above. Results On the basis of our settings, for OC, the optimal range identifying LVZs was 0.3‐0.6 mV. OC revealed smaller LVZs than MC (P < .05 or P < .001), except in the lateral wall. No significant differences were observed between CMCs. Conclusions In our experience, OC does not present the limits of bipolar HD maps, though further studies are needed in order to confirm that 0.3‐0.6 mV is the optimal voltage range within which to identify LVZs.


| INTRODUC TI ON
Although pulmonary vein (PV) isolation (I) remains a cornerstone of any atrial fibrillation (AF) ablation, [1][2][3] other anatomical regions of the left atrium, that is, non-PV substrates, are involved in AF, especially in persistent AF. 4,5 High-density (HD) endocardial voltage mapping by means of multipolar (MC) catheters and 3d electro-anatomical systems (3d-S) has been increasingly used in clinical practice to identify both left ventricular 6 or left atrial anatomical areas of low-voltage electrical activity, 7,8 which is commonly considered a marker of atrial fibrosis. 9 Left atrial substrate modification by targeting low-voltage zones (LVZs) is an ablation strategy that, in addition to pulmonary vein isolation (PVI), tries to erase arrhythmogenic mechanisms harbored in such tissue. 7,10 However, this approach has limitations because LVZs are subject to various influences, such as the bipole orientation expressed by the angle of attack and the activation wave-front; this can mean that electrical signals may not be recorded even when they are present 11 . Recent reports have described experiences of the use of new catheters with omnipolar (OC) recording capacity, which do not seem to be affected by the negative influences described above with regard to bipolar HD maps. 12,13 The aim of the present study was to perform a post hoc analysis in patients undergoing PVI, in order to evaluate the incidence of non-PV substrates detected by different diagnostic catheter technologies, including MC, OC, and circular mapping catheters (CMCs), and by means of qualitative and quantitative analyses of LVZs on applying various voltage ranges.

| Study population
Between February 2019 and January 2020, we enrolled 70 consecutive patients undergoing PVI for paroxysmal AF (n = 40, 57.14%) or persistent AF (n = 30, 42.87%). All patients were strongly symptomatic for palpitations, fatigue, dyspnea, or chest pain, and refractory to antiarrhythmic drugs (AADs). Table 1 reports their baseline characteristics at the time of ablation. All patients provided written informed consent after being adequately informed of the risks and benefits. After PVI, we performed HD mapping in all patients during sinus rhythm: in 34 patients, we used the CARTO® mapping System (CARTO ® 3 V6, Biosense Webster, Inc) and the Lasso ® Nav Circular Catheter or the PentaRay ® Nav Multipolar Catheter (Biosense-Webster, Inc); in 36 patients, we used the EnSite Precision™ 3d system (Abbott) and the Inquiry™ AFocus II™ Circular Catheter or Advisor™ HD Grid-Mapping Catheter (Abbott).

| Procedural setup: Ablation settings
PVI was carried out through contact force-guided ablation by means of a 3.

| Voltage mapping
After PVI, the patients underwent HD mapping of the lesions around of the criteria on which PVI was based. All HD maps were based on the acquisition of as many voltage points as possible, and the Tissue Proximity Indicator filter of 3d-S CARTO was used to exclude map points deemed not to be in contact with the shell. Figure 1A presents LVZs. A large area of different colors (dashed area) is clearly visible; this is characterized by small islands of normal tissue (purple), separated by probable patchy fibrosis (blue/green/yellow/red) inside the remaining AW of gray color; this latter area displays no electrical activity and is unable to mediate reentry. C, AP projection. Persistent AF and nondilated LA. The entire AW presents LVZs and is green in color (compatible with dense fibrosis) with only some small islands of healthy tissue (purple) and scar tissue (red), which poorly represent patchy fibrosis. From our point of view, this substrate is unlikely to be able to mediate reentry. However, this map could also be influenced by the limitations of bipolar recordings. D, PA Projection. Paroxysmal AF and nondilated LA. The posterior wall (PW) presents LVZs that is qualitatively compatible with patchy fibrosis. E, PA Projection. Persistent AF and dilated LA. The IW presents normal tissue, and the PW a few small islands of low voltage which are not able to mediate reentry. F, AP Projection. The same patient as in D. The map shows a prevalence of green and red colors that are more compatible with dense fibrosis and less so with patchy fibrosis, indicating a substrate that is probably unable to mediate reentry creates voltage maps using the electrogram with the highest ampli-

| Postablation management and follow-up
The patients were followed up for a period of about 12 months.
The protocol included the prescription of AAD: for a 3-month blanking period in patients with paroxysmal AF and without in women).

| Patients
The two groups of patients (with paroxysmal and persistent AF) were homogeneous in terms of age, sex, diabetes, coronary artery disease, and body mass index (Table 1). Hypertension and left atrial dilation were more frequent in persistent than in paroxysmal AF patients (P =.04 and P <.001, respectively), while left ventricular ejection fraction >50% and normal left atrium were more frequent in paroxysmal than in persistent AF (P = .01 and P <.001, respectively;  for the green points, and 0.33 ± 0.02 for the red points.

| Comparison of LVZs: CMC AFocus versus Lasso Nav and OC versus MC
As no LVZs were found in the left atrial roof of any patient, this anatomical area was excluded from the analysis.    Note: Values are expressed as numbers and percentages.

| D ISCUSS I ON
Abbreviations: LVZs, low-voltage zones; SD, standard deviation; VP, voltage points. a scar threshold of 1.5 mV (a value far higher than the standard 0.05 mV for conventional bipolar mapping) corresponded better to electrophysiologist-determined scar than the area determined from bipolar signals. 22  probably constitutes the basis of a truly tailored ablation. Indeed, our approach, which was both qualitative and quantitative, helped us to identify potential patchy fibrosis; thus, it was probably able to define a substrate that could represent local non-PV conductive alterations that influence anisotropy. This substrate probably plays a critical role in giving rise to AF mechanisms, as described by some authors. 26,27 We used the OC to search for the range that Finally, the results from our quantitative analysis of the substrate identified by means of OC and the optimal range were similar to those reported by Masuda et al. 28 These authors compared along and across CMC bipolar HD maps with OC HD maps, and found a significantly lower percentage of LVZs in the majority of left atrial anatomical areas studied by means of OC than in those evaluated by means of bipolar HD maps. In our experience, the total number of voltage points acquired were significantly higher with OC than with MC, thus allowing us to construct maps with higher density. was similar among the different catheters used; this was probably because of the fact that more complex mapping is required in patients with persistent AF, in whom OC were preferentially used, while CMC mapping yielded fewer voltage point.

| Limitations
This study has some limitations.

| CON CLUS ION
According to our experience, the combined use of OC and 3d-S EnSite Precision selectively identifies substrates that are potentially responsible for the mechanisms of AF by detecting LVZs that are not affected by the orientation of the catheter with respect to the activating wave-front. This approach therefore overcomes one of the limitations of using HD maps derived from bipolar catheters. In our study, the 0.3-0.6 mV range identified potential fibrotic substrates that could play a role in the mechanism of sustained AF better than the standard range of 0.05-0.5 mV. However, further studies are needed in order to determine whether 0.3-0.6 mV is the optimal range within which to identify LVZs as expression of patchy fibrosis by means of HD Grid and EnSite Precision.