Substrate‐Based Ablation Guided by Combined Fractionation and Frequency Analysis for Multiple and Unmappable Ventricular Tachycardias

Abstract

Stepwise evaluation of fractionation and peak frequency maps on substrate mapping identified VT‐related substrate regions in a patient with multiple ventricular tachycardias. This complementary substrate‐based approach may help guide ablation when complete circuit characterization is challenging.

For ventricular tachycardia (VT) ablation, activation or entrainment mapping can identify the critical site when the VT is hemodynamically stable. However, mapping is difficult in patients with hemodynamically unstable VTs or multiple VT morphologies, requiring substrate‐based ablation. Although abnormal electrograms and late potentials are frequently identified within low‐voltage areas, these findings do not always identify the critical isthmus. Irie et al. demonstrated that successful ablation sites may be located in areas of slow conduction rather than in the latest activated regions during sinus rhythm [1]. Fractionated electrograms are associated with VT channels but are insufficient to identify the VT circuit by themselves [2]. Peak‐frequency (PF) analysis has been used to evaluate local electrograms; however, PF alone may not clearly distinguish the critical isthmus from bystander scar [3, 4]. We report a patient with multiple VT morphologies in whom combined fractionation and PF analysis helped identify VT‐related substrate.

A 58‐year‐old man with dilated‐phase hypertrophic cardiomyopathy (left ventricular ejection fraction, 30%) was admitted for catheter ablation because of a VT storm. Although clinical VT was not inducible, four VT morphologies were induced (Figure 1). VT2 and VT3 could be mapped, but VT3 terminated before completion of activation mapping, whereas VT1 and VT4 were hemodynamically unstable. Therefore, substrate evaluation during baseline rhythm was used to determine the ablation strategy except for VT2. A left ventricular voltage map obtained during right ventricular pacing demonstrated low‐voltage areas (≤ 0.1 mV) at the apex and anterior wall (Figure 2). Fractionated electrograms and delayed potentials were observed within these regions. Activation mapping during VT2 demonstrated a circuit within the apical low‐voltage area with diastolic potentials. Radiofrequency application at the presumed critical site terminated VT2 after 52 s (Figure 3). For the unmappable VTs, abnormal electrograms were analyzed using the Fraction map and PF map of the EnSite system. In the Fraction map, small components in local electrograms were detected using a sensitivity of 0.04 mV, minimum width of 5 ms, and minimum refractory period of 6 ms. Detected components were expressed as the fractionation number, and points exceeding the selected threshold were displayed on the map. When the mapping window included all delayed potentials from QRS onset, areas showing marked QRS fragmentation could be visualized.

FIGURE 1.

ECGs during sinus rhythm and VT. The cycle length (CL) of clinical VT was 330 ms. Lead I showed a Qr pattern, leads II and III an RR pattern, lead V1 an rS pattern, and lead V6 an RR pattern. Four types of VT were induced during the procedure, and all differed from the clinical VT. VT1 and VT2 had a superior axis, while VT3, VT4, and the clinical VT had an inferior axis. From the precordial leads, VT1 and VT2 were suspected to originate from the apex. VT3, VT4, and the clinical VT showed an R wave in V1. VT2 and VT3 could be mapped. VT3 terminated during mapping and was incomplete. VT1 and VT4 were hemodynamically unstable, so activation mapping was not performed. CL, cycle length; SR, sinus rhythm.

FIGURE 2.

Substrate map. A substrate map of the left ventricle was created during right ventricular pacing. The left panel shows a voltage map obtained using near‐field detection. The right panel shows Isochronal late activation mapping (ILAM). White circles (1 cm in diameter) indicate deceleration zones. Low‐voltage areas were seen in the apex and on the anterior‐lateral wall. Voltage was relatively preserved in the septum. Delayed potentials were seen in the apical low‐voltage area (light blue tags). Two good pace maps for VT1 were identified nearby (white tags). A good pace map for VT2 was also found (yellow tag).

FIGURE 3.

VT2 activation map. VT2 rotated within the low‐voltage area of the apex. The white dashed lines show the conduction direction estimated from the activation map during VT2. Mid‐diastolic potentials were recorded around the circuit (green tags). The VT2 circuit likely enters near the yellow tag and exits near the VT1 good pace map point (white tag). The local electrogram at the yellow tag showed a fragmented diastolic potential. Radiofrequency energy delivered at this site terminated VT after 52 s.

Electrograms within the VT2 circuit demonstrated higher fractionation numbers and higher PF than those in surrounding myocardium and mostly corresponded to the VT isthmus identified by activation mapping (Figure 4A). The fractionation threshold was gradually adjusted from 1 to 9 to identify continuous abnormal electrograms. A fractionation number of 5 best highlighted the VT‐related region based on the activation map and signal continuity (Figure 4B). The PF cutoff was varied from 200 to 600 Hz in 50‐Hz increments, and a cutoff of 450 Hz most clearly delineated the VT‐related region (Figure 4C). Combining both maps with stepwise threshold adjustment improved identification of VT‐related substrate. In addition to the apical substrate, areas with a fractionation number ≥ 5 and PF ≥ 450 Hz were considered involved in the other VT circuits based on pace mapping and partial activation mapping. Empirical ablation targeting these areas rendered all VTs non‐inducible (Figure 4D). No VT recurrence has been observed during three years of follow‐up.

FIGURE 4.

(A) Local electrograms on and around the VT2 isthmus. A voltage map during right ventricular pacing is shown. The white dashed lines show the conduction direction estimated from the activation map during VT2. Local electrograms, fractionation number, and peak frequency (PF) are shown at the entrance, isthmus, exit, and outer loop. Electrograms within the VT circuit had relatively higher fractionation number and higher PF than surrounding areas. Ablation near the entrance terminated the VT. PF, peak frequency. (B) Emphasis map (voltage map+fractionation map). This is an emphasis map that combines the voltage and fractionation maps. The white dashed lines show the conduction direction estimated from the activation map during VT2. Yellow tags mark the VT entrance, and white tags mark the exit. We changed the fractionation threshold stepwise from 1 to 9. Mapping points that meet the fractionation threshold are shown as small fluorescent yellow dots. The figures show fractionation thresholds of 4, 5, and 6. The VT‐related region was most clearly delineated at a fractionation threshold of 5. (C) Emphasis map (voltage map + fraction map + peak frequency map). This emphasis map combines the voltage map, peak frequency (PF) map, and fraction map. The white dashed lines show the conduction direction estimated from the activation map during VT2. Yellow tags mark the VT entrance, and white tags mark the exit. We fixed the fractionation threshold at 5 and changed the PF cutoff in 50 Hz steps from 200 Hz to 600 Hz. The highlighted areas represent local potentials with a PF higher than the cutoff. The figure shows the PF cutoffs at 350 Hz, 400 Hz, 450 Hz, and 500 Hz. The VT‐related region was most clearly delineated at a PF cutoff of 450 Hz. PF, peak frequency. (D) Final ablation sites. VT1 and VT2 were ablated in the apical region. VT3 was ablated across the anterior to lateral wall based on pace mapping and fractionated electrograms. Areas with a peak frequency ≥ 450 Hz and fractionation number ≥ 5 were targeted. These areas likely covered all VT circuits. No VT was inducible after ablation. There was no recurrence during three years of follow‐up.

Importantly, the thresholds were not fixed but were adjusted stepwise to identify areas visually distinguishable from the surrounding tissue. For VT2, the highlighted region mostly corresponded to the VT isthmus estimated by activation mapping. By continuously adjusting these thresholds, the VT‐related region became more clearly identifiable. These findings suggest that VT substrate may not be defined by a single threshold but rather represent continuous electrophysiological properties.

In this case, the VT‐related region was located near an isochronal late activation mapping (ILAM) deceleration zone (Figure 2), and additional analysis identified a region suggestive of a rotational activation pattern (RAP). These findings suggest that the substrate identified by fractionation and PF analysis may overlap with regions detected by established functional substrate mapping techniques. However, ILAM and RAP may be influenced by factors such as pacing site and annotation settings. Therefore, we also evaluated the substrate using different electrogram characteristics, including fractionation and PF. Rather than replacing established functional substrate mapping techniques, this approach may serve as a complementary tool for assessing the substrate from multiple perspectives. Multiple VT morphologies were induced in this case. Although these VTs were not assumed to originate from an identical circuit, they may have arisen from related scar regions and shared some substrate characteristics. Electrogram characteristics identified in the mapped VT were therefore used to assess other potentially arrhythmogenic regions within the same patient. The aim was not to identify a universal cutoff value, but to recognize relatively abnormal electrogram characteristics within the individual substrate. Unlike the conventional approach of identifying a circuit for each VT, a strategy that focuses on the substrate itself may be especially useful in cases with multiple or unmappable VTs. This approach may be applicable when the clinical VT is not inducible, when VTs are hemodynamically unstable, or when multiple circuits are present.

There are several limitations. First, the cutoff values for fractionation number and PF were selected from a single mapped VT in this patient and should not be interpreted as universally applicable thresholds. They were used only to highlight relatively abnormal electrogram characteristics within the individual substrate and may vary according to substrate, mapping conditions, and electrogram characteristics. Second, although combined fractionation and frequency analysis may provide information regarding the three‐dimensional extent of the substrate, epicardial mapping was not performed. Therefore, the presence of epicardial substrate and the need for epicardial mapping should be investigated. Third, the results may vary between operators because the quality of substrate mapping differs according to mapping conditions and electrogram acquisition. Finally, this report describes a single patient, and further studies are required to validate the usefulness and reproducibility of this approach.

The combined use of the Fraction map, which visualizes the distribution of fractionated electrograms, and the PF map, which evaluates electrogram frequency, may complement conventional substrate mapping and help identify ablation targets in patients with multiple or unmappable VTs.

Author Contributions

M.K. designed the study, collected and analyzed the data, prepared the figures, and wrote the manuscript. K.H. contributed to the study design, reviewed the manuscript, and provided clinical input. K.A. supervised the study and critically revised the manuscript. All authors read and approved the final manuscript.

Funding

The authors have nothing to report.

Ethics Statement

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.