Mapping of ventricular tachycardia in patients with structural heart disease

Abstract

Catheter ablation is now recognized as one of the potent options to reduce the number of implantable cardioverter defibrillator (ICD) shock therapies by modifying arrhythmic substrate in patients affected by ventricular tachycardia (VT). However, operators often face some difficulties during mapping and ablation procedure of VT; unstable VT makes it difficult to create activation map, the thick ventricular myocardium yields complex tachycardia circuit, and endocardial radiofrequency (RF) energy application hardly creates transmural lesion. Thus some novel approaches are required to improve outcome of VT ablation. Substrate mapping aiming low voltage or abnormal fragmented potential could identify the critical slow conduction area during sinus rhythm and define the target of RF delivery. CT and MRI image integrated to electroanatomical mapping (EAM) system also could help to find diseased myocardium.

In this article focused on VT mapping procedure, conventional and novel mapping technique as well as some technical tips should be reviewed comprehensively.

1 Introduction

Implantable cardioverter defibrillator (ICD) therapy has improved the prognosis of patients with lethal ventricular arrhythmias [1–3]. The ICD terminates episodes of ventricular tachycardia (VT) and saves patients from sudden cardiac death; however, it does not change the arrhythmogenic substrate. Previous studies have shown that patients who receive shock therapy have a poorer prognosis [4], which suggests the necessity of supplemental preventive therapy to reduce the ICD shock frequency. The indication for antiarrhythmic drugs is strictly limited because of their negative inotropy, proarrhythmic effect, and poor compliance [5]. Catheter ablation is now recognized as the most potent non-pharmacological option to modify the reentrant arrhythmogenic circuit and in some clinical studies has actually been proved to reduce the number of ICD shocks [6,7]. However, poor inducibility of the index VT, concomitant non-clinical VT, and hemodynamic and electrical instability often make it difficult to perform activation mapping during VT. Moreover, radiofrequency energy applied endocardially is sometimes insufficient to create transmural lesions across thick ventricular myocardium. Thus, to achieve success in VT ablation, some novel approaches are required in addition to conventional mapping and ablation techniques. When activation mapping is not available, identification of the substrate that forms the critical part of the reentrant circuit during sinus rhythm enables mapping in a hemodynamically stable condition, and permits a strategy for ablation even in non-inducible patients. The bipolar/unipolar scar identified by the electroanatomical mapping (EAM) system, late enhancement detected by CT/MRI, and late or fractionated local activity could be potential targets for ablation. In a number of cases, the VT circuit could be located in the epicardium or even in an intramural layer. Thus, it is important to identify the substrate optimally at the beginning of the procedure and to prepare an appropriate approach (transaortic, transseptal, or epicardial). With better technology, such as contact force monitoring, high-definition mapping catheters, and new signal processing algorithms, our understanding of VT substrate mapping will improve in the near future. In this article, the present strategy of procedures for the mapping and ablation of VT is reviewed.

2 Preparation of the procedure 2.1 12-Lead ECG

It is highly advisable to obtain a 12-lead recording of spontaneous or “clinical” VT before the procedure, not only because it can be used as a reference for induced VT or pace mapping during the procedure, but also because it provides rough information about the origin of the tachycardia. Generally, a right bundle-branch block pattern in lead V₁ suggests that the tachycardia originates from the left ventricle, while tachycardia from the right ventricle or intraventricular septum presents as a left bundle-branch block pattern. The electrical axis of the QRS complex reflects the direction of the activation wavefront in the frontal plane; i.e., positive QRS in inferior leads (II/III/aVF) indicates that the VT comes from the antero-superior part, whereas a negative QRS complex represents an inferoposterior origin. On the other hand, positive concordance in the precordial leads means that the VT origin is located in the basal area, while negative concordance suggests the VT origin is located apically [8,9]. The transitional zone and the change of polarity (positive to negative or negative to positive) provide information about propagation in the horizontal plane.

Epicardial VT usually reveals a wide QRS duration with a slow upstroke from onset to peak [10,11], because it takes a relatively long time before the activation wavefront captures the His–Purkinje system with rapid conduction located in the endocardium. A careful estimation of the VT morphology should be taken into account when judging the necessity for an epicardial approach in the actual procedure.

Regardless of whether they are ischemic or non-ischemic, a considerable number of patients have multiple VT. Detailed analysis of each VT morphology gives an outline of the VT circuit (i.e., whether each VT circuit shares a common isthmus or is totally independent) and thus provides important information for plotting treatment strategy.

2.2 Anatomy acquisition and scar identification by MRI or CT

In the absence of any contraindication, such as renal dysfunction or electrical devices, it is highly recommended to obtain enhanced MRI or CT images as far as possible before the mapping procedure. The CT or MRI image can be merged with the electroanatomical mapping system and improves the accuracy of the map. Anatomical information, such as the size of the ventricle, the location of coronary arteries, and the distribution of epicardial fat, provides important information for an adequate mapping strategy or identifies potential risks [12,13]. Moreover, the abnormal damaged myocardium, which can play a critical role in the VT, can be detected by late enhancement [14–16].

2.3 General preparation

Generally, patients who have VT associated with structural heart disease have impaired cardiac function and poor tolerability for the treatment procedure. For successful VT ablation, it is essential to design a procedural strategy that is based on the patient's background, the characteristics of the tachycardia, possible procedural risks, and any other conditions. As even tolerable VT sometimes loses its hemodynamic stability easily under triggers such as drug administration and radiofrequency energy delivery, it is mandatory to make a standby mechanical support system (PCPS, ECMO, or IABP) available in any VT case [17–19], as well as estimating the cardiac capacity thoroughly using echocardiography, coronary angiography, and other possible modalities before the procedure.

For an epicardial approach or hemodynamically unstable VT, intensive care—including general anesthesia and mechanical support in collaboration with an anesthetist or cardiac surgeon—is often required [20]. On the other hand, the inducibility and sustainability of VT are decreased under general anesthesia, which makes the mapping procedure during VT difficult; thus, substrate mapping might be a major tool for identifying the target of ablation.

2.4 Access and catheter settings

To avoid a risk of excess bleeding followed by possible iatrogenic trauma, all access should be performed before the administration of heparin. Left ventricular endocardium can be accessed by two approaches: retrograde transaortic and antegrade transseptal. Which approach is more preferable depends on the location of the target site in the ventricle.

Technically, we usually use the 8.5 Fr armored long sheath (Super Arrow-Flex, Arrow International Inc., Reading, PA, USA) for the retrograde transaortic approach, which can straighten the femoral artery and provide easier delivery of the catheter to the descending aorta. For the transseptal approach, an SL0 sheath is used for the initial puncture, and is then replaced by a manually controlled steerable sheath (Agilis, St. Jude Medical, Minneapolis, MI, USA). The tip of the sheath should be placed at the level of the mitral annulus with the mapping catheter tip protruding at least 3 cm from the sheath.

The subxiphoidal pericardial puncture is currently performed using the technique originally proposed by Sosa et al. [21,22].

For an inferior epicardial approach, starting from the border of the subxiphoid, a Tuohy needle is advanced toward the middle inferior wall of the right ventricle, generally at 45° to the skin's surface and at the same angle to the middle line of the sternum. In some cases, a selective approach to the anterior wall of the right ventricle may be useful, in which case the angle between the needle and skin surface should be smaller than the conventional 45°. Catheters positioned in the right ventricle apex and coronary sinus will be good landmarks. Continuous fluoroscopic monitoring in right and left anterior oblique views is recommended during the advancement of the needle tip toward the cardiac silhouette. A small amount of contrast medium will make it easier to see tenting of the pericardium. Monitoring the pressure of the tip of the needle is also a useful technique: once the needle goes into the pericardial space, the pressure usually drops steeply and the oscillating pressure curve clearly indicates penetration of the needle tip into the ventricle. With the conventional Seldinger technique, a 9-F sheath introducer is positioned in the pericardial space so that the mapping catheter may be inserted, while also allowing fluid to be drained through the side port. To avoid the risk of losing access, we prefer to use a long sheath (23 cm) instead of the conventional 11 cm sheath. Another recommendation is to verify the position of the sheath by injecting a small amount of contrast media into the pericardial space before removing the wire.

3 Initial induction of ventricular tachycardia

Commonly, VT induction by programmed ventricular stimulation (PVS) is attempted at the beginning of the procedure, except in some cases, such as when clinical VT has already been documented with reproducibility, the patient exhibits incessant or spontaneous VT, and the ablation target is clearly identified. Although the relationship between the inducibility of VT and clinical VT has not been fully elucidated, induced VT provides some useful information [23]. The presence or absence of VT inducibility, the hemodynamic stability or sustainability of the induced VT, and the VT morphologies help to determine the procedure strategy, required support (including extracorporeal circulation or surgical approach), and endpoint.

It is advisable to perform initial VT induction with the patient in a waking state, as general anesthesia markedly modifies the inducibility of VT by PVS. If VT is not induced easily, repeated PVS (with maximum 4 extrastimuli or burst pacing) is delivered at a different basic cycle length or different site (right ventricular apex, right ventricular outflow tract, or left ventricle). If necessary, isoproterenol may be administered.

4 Conventional mapping and ablation of ventricular tachycardia 4.1 Possible target of VT ablation

For the initiation and maintenance of VT, triggers are usually required in addition to an anatomic substrate. The electrophysiological condition of triggering is not always consistent and may be affected by the status of the patient, such as heart failure, myocardial ischemia, abnormal blood levels of electrolytes, or autonomic imbalance.

In coronary artery disease, most VTs arise from a reentrant circuit containing a typical slow-conduction zone, usually protected and located within the infarct scar or between the scar and an anatomical obstacle such as the mitral annulus.

On the other hand, VTs in patients with non-ischemic cardiomyopathy (NICM) are more heterogenous [24–26]. The abnormal conduction properties of NICM are caused by deletion of myocytes, replacement by interstitial fibrosis, and myocardial disarray rather than necrosis. Ischemic scar is located predominantly in endocardial myocardium, whereas the VT substrate of NICM is detected in endocardial, epicardial, or even in intramural myocardium [27]. The substrate in NICM VT is often more diffuse, more atypical, with a wider spectrum than ischemic VT, and occasionally may even be difficult to identify. Moreover, the potential mechanism of NICM VT is triggered activity in addition to reentry [28,29]. For these reasons, VTs in NICM patients are less monomorphic and are infrequently induced during EP studies.

It is often difficult to determine the reentrant circuit in patients with polymorphic VT. In some cases, polymorphic VTs may be initiated by a specific ectopic beat with unique morphology, which can be also a target of ablation.

4.2 Activation mapping during VT

When spontaneous or induced VT is hemodynamically and electrophysiologically stable, the exit from the slow-conduction area can be identified by the earliest activation site of the activation map obtained during VT, using an electroanatomical mapping system such as CARTO3 (Biosense Webster, CA) or EnSite (St. Jude Medical, MN). The onset or peak of the QRS complex on the surface ECG, or local potentials of the electrode catheter positioned at the RV apex can be used as reference.

A presystolic potential is often observed near the exit site of the slow-conduction area during VT, and is especially common among patients who have ischemic VT with scar. Typically, as the tip of the recording catheter moves proximally, this potential precedes gradually, being seen during the mid-diastolic phase in the isthmus and just after the QRS on entry to the slow-conduction zone [30]. It is sometimes difficult to detect these specific potentials, especially when only a little viable myocardium remains. On the other hand, they are recorded even in innocent bystander sites in the slow-conduction area. Overdrive pacing from the site with diastolic or presystolic potentials during VT is useful in order to characterize the position of the catheter in the slow-conduction area; if the catheter tip is located in the critical isthmus, the stimulus-QRS interval is identical to the interval between the local potential and the QRS (EGM-QRS). If the catheter is positioned in a bystander site, the stimulus-QRS interval should be longer than the local EGM-QRS.

4.3 Entrainment mapping during VT

Overdrive pacing intervention during VT from a different site is often utilized to investigate the mechanism of the tachycardia and to determine the position and distance between the pacing site and the tachycardia circuit [31,32].

Pacing during VT at a slightly higher rate produces a fused QRS morphology composed of the orthodromic wavefront from the VT itself and the antidromic wavefront due to pacing. When VT is induced by increased automaticity, pacing does not affect the tachycardia rate; thus, the degree of fusion differs beat by beat. Meanwhile, when the VT is reentrant, a paced impulse that enters and passes through the reentrant circuit breaks out from the exit to form an orthodromic wavefront and then collides against an antidromic wavefront formed by the next pacing impulse. As the paced impulse constantly resets the tachycardia, the degree of fusion is fixed and reveals consistent QRS morphology (constant fusion). At the end of pacing, the last paced impulse enters/emerges from the circuit and activates the whole ventricle without collision against the next impulse. Thus, the last pacing beat produces only the orthodromic wavefront, resulting in the same QRS morphology as clinical VT. A higher pacing rate makes the fusion occur more proximally to the circuit, i.e., it yields a greater area captured by pacing in an antidromic manner. As a result, with an increasing pacing rate, the morphology of the fused QRS remains constant, and gradually becomes dominated by a paced QRS morphology rather than the clinical VT (progressive fusion). The same phenomenon is observed in the intracardiac electrogram. The change in activation timing, polarity, and morphology of a specific local potential caused by a change in the pacing rate means that the manner of capture of the local corresponding myocardium has shifted from orthodromic to antidromic. Termination of the VT presents a different activation pattern, since the next pacing impulse can no longer fuse with the orthodromic wavefront.

The interval between the final pacing spike and next local activation (PPI: post pacing interval) presents as: PPI=conduction time from pacing site to reentrant circuit+cycle length of VT+conduction time from reentrant circuit to pacing site.

If the myocardium is paced within the reentrant VT circuit, the conduction time between the pacing site and the circuit should be zero; thus, PPI is equal to cycle length. Furthermore, when myocardium in the slow-conduction zone is paced, the antidromically captured area is limited to within slow-conduction area, without any contribution to form the surface QRS; thus, the surface QRS morphology remains identical to that of the clinical VT.

4.4 Pace mapping

In case previously documented clinical VT is not induced or unstable, pace mapping can be used. The origin of VT is investigated by comparing the paced QRS configuration obtained from different parts of the ventricle with that of the clinical VT. When the paced QRS morphology is completely matched (12/12 leads) to the clinical VT, it is considered “ideal.” When 10–11/12 leads match, the paced QRS is regarded as “good.” The stimulus-QRS interval also provides useful information; a site with stimulus-QRS latency more than 80 ms is more likely to be an ablation target [33].

It is recommended to pace using the same cycle length as that of the clinical VT. Broad capture by high-output pacing or bipolar pacing can alter the QRS morphology. Even within the true critical VT isthmus area, pacing from the site near the entrance of the isthmus with a long stimulus-QRS interval can also yield a different QRS morphology, because an antidromic wavefront can propagate through normal myocardium, while the wavefront spreads in one direction from the site of origin to the remainder of the heart during clinical VT. The absence of functional block during pace mapping under sinus rhythm, which is commonly observed during clinical VT, can also produce a different QRS configuration.

In contrast, pacing from a non-critical bystander site in a slow-conduction area can occasionally produce a QRS morphology identical to that of the clinical VT, as the surface QRS starts when the local activation exits the area and propagates into normal myocardium.

Therefore, the spatial resolution of pace mapping is not as high as that of activation mapping during VT. Although pace mapping is a useful method for the localization of a small VT focus in a structurally normal heart, such as outflow VT, its advantage might be smaller in VT patients with structural heart disease. Especially when the VT originates from intramural myocardium or epicardium, far from the endocardium, the reliability of pace mapping is relatively poor.

5 Substrate mapping during sinus rhythm

As mentioned above, conventional mapping, including identification of the earliest activation site and critical slow-conduction isthmus by activation mapping or entrainment during spontaneous/induced VT, has been widely utilized in the mapping of supraventricular tachycardias. Essentially, these mappings are performed during VT and are not available in a patient with collapsing VT. Moreover, the inducibility of VT is not high and, even in a patient who has had recurrent episodes, clinical VT cannot always be induced. These technical issues of conventional mapping have been associated with the low success rate of mapping/ablation procedures for VT. Recently developed electroanatomical mapping systems have introduced the new concept of substrate mapping, which provides information about abnormal myocardium and makes it possible to deliver radiofrequency (RF) energy to the target substrate of the VT circuit during sinus rhythm.

5.1 Voltage mapping

Voltage mapping is based on the concept that abnormal myocardium delivers a low voltage and can form a critical slow-conduction area that can be detected using reconstructed data of the voltage amplitude for each point in the ventricle, obtained using an EAM system (Fig. 1). Pathologically, slow conduction is caused by anisotropy due to missing or fibrous myocardium. In such areas, which lack a sufficient volume of intact myocardium, the total voltage should be decreased. Commonly, the area constructed by points with a bipolar voltage amplitude <0.5 mV is defined as dense scar, while the area between 0.5 and 1.5 mV is designated as a low-voltage area. Based on the voltage map, conduction channels are often identified as regions located within dense scar or between the scar and an anatomical obstacle (Fig. 2) [34,35].

Fig. 1. (A) Bipolar voltage map of the left endocardium (left panel) and epicardium (right panel) of a 66-year-old male with an inferior myocardial infarction. Transmural infarct scar is observed in the basal inferolateral and mid-inferior wall. (B) Local electrograms in the infarct scar (indicated as yellow points). The latency of late activity is greater in epicardium compared to endocardium. In fact, it required epicardial radiofrequency delivery in addition to endocardial ablation in order to obtain non-inducibility of ventricular tachycardia.

Fig. 2. (A) Bipolar voltage map of the epicardium (left panel) and left ventricular endocardium (right panel) of a 61-year-old male patient with non-ischemic cardiomyopathy. In the epicardial map, dense scar with signal voltage less than 0.5mV appears as a red area and the low voltage (signal voltage between 0.5 and 1.5mV) areas in yellow and green are located in the basal inferior and mid-inferior wall, whereas the endocardial voltage map is normal. Small red dots indicate ablation points. (B) Local activation time map during sinus rhythm. The late potential area is located between two inferior scars. (C) Local electrogram taken at the yellow point in the late potential area. A discrete isolated potential is observed. Total abolition of these potentials resulted in non-inducibility of ventricular tachycardia.

In typical myocardial infarction-related VT, the exit is usually located in the scar border zone. Thus, it is important to take as many points as possible along the scar to create a precise map with maximum resolution. A pacing maneuver is helpful to identify and characterize the conduction properties within the scar or border zone; the failure to capture a paced impulse provides information about fixed conduction block. The paced QRS morphology and stimulus-QRS interval also help to identify the conducting channel.

5.2 Abnormal potential mapping: late potentials and local abnormal ventricular activity

Within a diseased area, the local electrogram is generally characterized by low voltage, high frequency, and a long-lasting “fragmented” potential [36] caused by local conduction block or anisotropies among the myocardial bundle. Sometimes, an isolated discrete delayed potential is also observed. Typically, along the border of a myocardial infarct scar, a large and dull signal representing far-field signals of normal myocardium is followed by small, continuous, or isolated high-frequency signals. This small, fragmented potential is defined as the late potential if it is recorded after the end of the surface QRS. It can be a useful item for identifying conduction delay as a target of RF ablation. The latency or duration is likely to be associated with a slow-conduction isthmus, and in this context [24,37], total abolition of late potentials might be an endpoint of catheter ablation (Fig. 3). Previous studies have shown that this ablation strategy provides a favorable success rate [38].

Fig. 3. (A) Local activation time (LAT) map (upper panel) obtained from a 34-year-old male post-myocarditis. Local late activity (lower panel) is located in the epicardial apical-anterior area (indicated as purple). (B) Mid-diastolic potential seen in the same area during clinical ventricular tachycardia. (C) Relatively good correspondence of QRS morphology obtained by pacing from the same point. (D) LAT remap after epicardial ablation. Total abolition of the late potential resulted in non-inducibility of clinical ventricular tachycardia.

Although late potentials are one established target for ablation, there remains an issue. The detection of late potentials is affected by such factors as the morphology of the QRS complex, the location of the slow-conduction area, and the direction of the activation wavefront. For example, when the slow-conduction area is located in the septal wall, the abnormal fragmented potential is buried within the QRS. The presence of right bundle-branch block also can mimic negative late potentials in the left ventricle. In such cases, the abnormal fragmented potential is often recorded within the QRS or normal far-field activity and is never detected as a “late” potential. The potential of the conduction system sometimes forms multiple components and can be another factor that makes it difficult to distinguish the pathological fragmentation from the normal electrogram in the septal wall. Jaïs et al. reported the utility of pacing when such an abnormal high-frequency potential is observed within the QRS complex [39]. If the abnormal component can be separated by pacing from different sites, it suggests the presence of multiple conduction properties, possibly including critical slow conduction. Although the detection and confirmation of such fragmented activity by pacing is time-consuming, it can be useful to avoid overlooking it. Investigation is continuing into methods of separating these potentials from normal far field potentials directly, using signal processing techniques such as frequency analysis.

How specific these abnormal potentials are for a critical isthmus is an important question that remains to be solved. Total abolition of these potentials can eliminate all possible VTs, which can be critical in the future; however, it might involve a great area that includes innocent bystander sites. To improve the diagnostic accuracy or to clarify the pathological meaning of the potentials, additional features of the local electrogram, including the response to pacing, are recommended.

6 Technical issues 6.1 Importance of contact force monitoring during substrate mapping

The importance for accurate mapping of appropriate contact between the myocardial tissue and the catheter tip is widely recognized [40]. In ablation procedures, since the tip temperature feedback is practically meaningless for controlling lesion formation using irrigated catheters, the importance of contact force (CF) monitoring should be also emphasized.

As previous studies using non-irrigated ablation catheters have shown that CF is a predictor of successful lesion creation and complications during RF application, the importance of appropriate CF has been widely recognized since the early days of catheter mapping and ablation procedures. In the absence of a direct quantitative indicator of CF, many attempts to identify surrogate parameters of “good contact” have been made in the past, including direct visualization of the tip of the catheter with fluoroscopy or echocardiography, or indirect quantitative parameters, such as ST-segment elevation on the unipolar electrogram and changes in temperature or impedance during ablation. However, since the relationship between these surrogate indicators and actual CF has not been fully elucidated, no data on the adequate CF to be used in clinical settings will be available until we see improvements in real-time force sensing technology.

To evaluate the importance of CF monitoring during a VT mapping procedure, we measured CF using SmartTouch in 27 chambers (6 endocardial maps of the right ventricle, 13 endocardial maps of the left ventricle and 8 epicardial maps) of 17 patients [41]. Of all 8892 points reviewed, 5926 that had stable recordings across at least 2 s were subjected to further analysis. The frequency of points taken with poor contact (without positive force during diastole) reached 26%, despite all the operators’ efforts to obtain constant contact based on classical surrogate indicators. There was a categorical change in electrophysiological parameters such as bipolar and unipolar signal amplitude or local impedance with relatively low CF. Furthermore, points with late potentials were more frequent in the good contact group, which was consistent across the characteristics of the myocardial tissue (normal, low voltage area, and scar). Our data suggest that sufficient CF is needed to create a precise substrate map, especially for the recognition of abnormal late activity, as well as to deliver sufficient RF energy (Fig. 4).

Fig. 4. (A) Unipolar voltage map of points taken with insufficient (upper panel) and sufficient (lower) contact force. The unipolar scar seems to be bigger with insufficient contact force. (B) Frequency of late potentials in normal tissue, low voltage area, and dense scar. Late potentials were detected more frequently with good contact, independently of myocardial tissue characteristics. (C) A case in which increased contact force was associated with effective ablation. During the initial 30s, radiofrequency energy was delivered with a mean contact force (white line in lower panel) of 15g to the endocardial basal anterolateral wall without any effect. Four seconds after the contact force increased to 40g, the ventricular tachycardia terminated.

6.2 Usefulness of steerable sheath in endocardial and epicardial mapping

Generally, perspective and precise maps, as well as stable contact between the ablation catheter and myocardial tissue, are indispensable for all successful procedures. This is especially true in substrate mapping and VT ablation procedures, which require high resolution to identify an abnormal substrate and intense RF energy to create a transmural lesion. However, the possibility of reaching the whole myocardial surface and gaining a fine contact with the desired location is often hampered by insufficient back-up force via a standard sheath. A manually controlled steerable sheath provides a greater range of orientation and stable contact [42]. Previous observational and randomized prospective studies have already clarified the utility of this steerable sheath in procedures for cavotricuspid isthmus ablation and pulmonary vein isolation in patients with atrial flutter and fibrillation. This advantage of the steerable sheath should be maximized to improve the efficacy of VT procedures.

7 Conclusion

Three-dimensional electroanatomical mapping systems have provided a great deal of fundamental information about the VT substrate, which has extended the indications for VT ablation to unstable VT, which had been unmappable with conventional techniques. An electrophysiologically abnormal area with low voltage or fragmented potentials is now established as a potential target for ablation. However, its sensitivity and specificity in identifying a true critical isthmus need to be improved. Whether to ablate all abnormal substrates that can produce non-clinical VT is another remaining question. In the future, not only anatomical but also functional information, such as quantified wall motion or metabolism, may be integrated into EAM systems to provide a precise appreciation of the VT substrate.

Conflict of interest

The author (Hiroya Mizuno) receives study grant from Biosense Webstar Inc.