Management of Idiopathic Outflow Tract Ventricular Arrhythmias

Idiopathic ventricular arrhythmias (VAs) refer to ventricular tachycardia (VT) and premature ventricular contractions (PVCs) that occur in the absence of structural heart disease. While the majority of VT occurs in structurally abnormal hearts, 10% of VT is idiopathic. The incidence of idiopathic VAs is increasing, with approximately two-thirds of patients presenting with PVCs and one-third with VT.¹ This increase may be due to improved awareness and advancements in ambulatory monitoring and home ECG devices. The majority of idiopathic VAs originate in the outflow tract region, with earlier studies showing that 70–80% arise from the right ventricular outflow tract (RVOT).² However, there may be a recent trend towards more equal distribution between the RVOT and the left ventricular outflow tract (LVOT) given that successful sites of ablation are more frequently found in the LVOT, aortic root and coronary venous system.³ This shift is likely to be due to a combination of referral bias, increasing comfort with left-sided mapping/ablation, and recognition of intramural foci.

Outflow Tract Anatomy and Clinical Presentation

Anatomy

A detailed understanding of the complex anatomy of the outflow tract region is critical in interpreting ECGs of VT and PVCs, and for successful ablation. The RVOT courses superiorly in a leftward direction and crosses over the LVOT, and it becomes anterior and leftward to the LVOT (Figure 1 ).⁴ The pulmonic valve is superior and leftward to the aortic valve. Thus, the anterior aspect of the RVOT is the most leftward and highest outflow tract structure, and the septal RVOT is located posteriorly and inferiorly. Myocardial crescents are incorporated into the base of each pulmonary valvar sinus, and myocardial tissue is also found in the interleaflet triangle between the left and right coronary sinus.⁵ These myocardial crescents can be arrhythmogenic and are potential sites of ablation.

The LVOT corresponds to the elliptical opening of the LV and is composed of the aortic root anteriorly and the mitral annulus posteriorly and leftward.⁶ The aortomitral continuity (AMC) is a fibrous curtain of tissue that extends from the interleaflet triangle between the left coronary cusp (LCC) and non-coronary cusp (NCC) and the anterior mitral leaflet. The aortic valve occupies a central position in the heart and is in continuity with all four cardiac chambers. The right coronary aortic sinus of valsalva (RCAS) is in close proximity to the posteroseptal RVOT, and the distal RVOT lies anterior to the right coronary cusp (RCC)–LCC commissure. The non-coronary aortic sinus is posterior and adjacent to the interatrial septum. The membranous septum is below the RCC–NCC commissure, and the penetrating bundle of His is at this location.

The LV summit is the most superior, epicardial aspect of the LV (Figure 1C). It is a triangular region with the apex formed by the bifurcation of the left anterior descending and left circumflex arteries. The base is formed by an arc from the first septal perforator of the left anterior descending artery to the left circumflex artery.⁷ The great cardiac vein (GCV) can bisect this triangle into medial and lateral regions. The medial region is close to the bifurcation of the left main, and this region is covered with epicardial fat. This region is inaccessible for catheter ablation given the proximity of the coronary arteries and the insulating characteristics of fat. Communicating veins running off the GCV can supply the LV summit and can be targets for ablation.

Arrhythmia Mechanism and Clinical Manifestations

The primary mechanism for idiopathic outflow tract VAs is triggered activity. Lerman et al. demonstrated that most of these arrhythmias are due to catecholamine-induced delayed afterdepolarisations.⁸ Catecholamine stimulation of β-adrenergic receptors leads to an increase in cyclic adenosine monophosphate, which increases calcium current and subsequent calcium release from the sarcoplasmic reticulum. This leads to activation of the sodium–calcium exchanger and a net inward current, which in turn leads to delayed afterdepolarisations. This pathway can be interrupted by inhibiting the formation of cyclic adenosine monophosphate and/or calcium influx. Thus, these arrhythmias are often sensitive to β-blockers, calcium channel blockers, adenosine, and vagal manoeuvres. While triggered activity is the predominant mechanism, 12% of RVOT and 22% of LVOT VT in one study was adenosine insensitive, and other focal mechanisms such as enhanced automaticity can occur.⁹

Idiopathic outflow tract VAs present with three main phenotypes: repetitive monomorphic PVCs; runs of non-sustained VT; and exercise-induced sustained VT. Repetitive PVCs are the most common phenotype, occurring in 66.5% in one study of patients presenting with idiopathic VAs.¹ Exercise-induced VT is the least common phenotype. There can be overlap between the phenotypes, with exercise-induced VT present in 10% of patients whose clinical presentation is repetitive PVCs.¹⁰ Outflow tract arrhythmias often present between the third and sixth decades of life. Palpitations are the most common symptom, and fatigue, light-headedness and dyspnoea may also occur. Syncope is uncommon, and the risk of sudden death is very low. It is also important to recognise that patients may present with a PVC-induced cardiomyopathy, and this occurred in 4.7% in one study.¹ A PVC-induced cardiomyopathy can occur with a PVC burden as low as 10%, but the risk is highest with a burden of >20%.¹¹ This is a reversible cardiomyopathy, and ICD therapy for outflow tract VT in the setting of an isolated PVC-induced cardiomyopathy is not appropriate and may result in multiple ICD shocks (Figure 2).

Outflow tract VAs can be a manifestation of an underlying cardiomyopathy, such as arrhythmogenic right ventricular cardiomyopathy, and it is important to evaluate for structural heart disease with an echocardiogram. Cardiac MRI may also be reasonable to diagnose an occult cardiomyopathy, especially in older patients. In 255 patients with a mean age of 54.8 years, frequent PVCs and a normal echocardiogram, 13% had regional wall motion abnormalities and 11% had late gadolinium enhancement on cardiac MRI.¹²

Localisation by ECG

While invasive mapping is required to definitively pinpoint the location of the PVC or VT, the standard 12-lead ECG is useful in localising the PVC and for pre-procedural planning. Outflow tract VAs typically have a left bundle (LB) pattern in V₁ and an inferior frontal plane axis. V₁ is a right-sided and anterior lead. Thus, anterior free wall RVOT VAs have a near QS pattern. As the VA focus moves leftward (posteriorly), positive forces in V₁ become more prominent, with a right bundle (RB) pattern occurring with exits towards the mitral annulus (Figure 3).

ECG-based Separation of Right from Left Outflow Tract Sources

Differentiating RVOT and LVOT VAs can be challenging due to the close anatomical relationship of these structures. Lead I can be helpful in narrowing the site of origin. Rightward structures such as the posterior aspect of the RVOT, RCAS, para-Hisian region, and superior aspect of the tricuspid valve have a positive QRS in lead I.⁶ Leftward structures such as the anterior aspect of the RVOT, left coronary aortic sinus (LCAS), AMC, anterolateral mitral annulus and LV summit will produce a negative QRS in lead I. Further differentiation between RVOT and LVOT VAs requires close inspection of the precordial leads and the precordial transition. Given that the RVOT is more anterior than the LVOT, the precordial R–S transition (the lead where the QRS goes from predominantly negative to predominantly positive) for RVOT VAs is usually after V₃, whereas transition at V₂ or earlier is typical of LVOT origin. If the precordial transition is in V₃, then the site of origin can be in either the RVOT or LVOT. The first step is to compare the precordial transition between sinus rhythm and the outflow tract VA to account for baseline cardiac rotation. If the transition during the VA is later than sinus rhythm, this predicted an RVOT origin with 100% specificity in one study.¹³ A transition zone (TZ) index can also be calculated for the precordial transition between sinus rhythm and the VA, with a score assigned to the lead of transition.¹⁴ There are higher scores for later transition, and the lead score during sinus rhythm is subtracted from the lead score during the VA. If the TZ index is <0, then this predicts a left-sided origin with a sensitivity of 88% and specificity of 82%.

If the precordial transition for the outflow tract VA is in V₃ and is the same in sinus rhythm, other ECG criteria may be applied.¹⁵ The V₂ transition ratio can be applied to localise outflow tract VAs that exhibit precordial transition in V₃. The calculation for the ratio in V₂ is as follows: proportion of R wave in VA divided by the proportion of R wave in sinus; that is: R wave in VA/(QRS amplitude in VA) divided by the R wave in sinus/(QRS amplitude in sinus) (Figure 4A).¹³ A ratio of ≥0.6 predicted an LVOT origin with 91% accuracy. The V₂S/V₃R index was subsequently shown to be more accurate than the V₂ transition ratio and TZ index.¹⁶ This is calculated by measuring the amplitude of the S wave during the VA in V₂ and dividing it by the amplitude of the R wave in V₃ (Figure 4B). A cut-off of ≤1.5 predicted an LVOT origin with 89% sensitivity and 94% specificity and outperformed the other ECG criteria. Combining the TZ and V₂S/V₃R indices may be even more accurate in differentiating RVOT from LVOT VAs.¹⁷

Once the site of origin is determined to be in the RVOT or LVOT, other ECG characteristics can be used for localisation. For the RVOT, sites of origin can be in the free wall or septum, with septal locations more common than free wall locations.^18,19 Free wall sites have a wider QRS duration, notching in the inferior leads, and a later precordial transition (Figure 3). Septal sites have a narrow QRS, lack of notching, and earlier precordial transition. The majority of RVOT VAs originate just below the pulmonic valve, and lead I is useful to differentiate anterior and posterior locations as outlined above. In more recent cohorts, 11% of RVOT VAs may be found above the pulmonic valve in the pulmonary sinuses.²⁰ Suprapulmonic sites tend to have higher R wave amplitudes than subpulmonic sites.

In the LVOT, lead V₁ is useful for localising the site of origin. In the RCAS, V₁ typically has a QS or rS pattern in V₁, small broad R wave in V₂, and precordial transition in V₃ (Figure 3).²¹ The LCAS is more posterior, and VAs from this region have taller and broad R waves in V₁ with early precordial transition. There is often an ‘M’ or ‘W’ pattern in V₁ (Figure 3). Outflow tract VAs from the commissure between the LCC and RCC typically have an LB pattern in V₁ with a notch in the downstroke, and there is an abrupt precordial transition from V₂ (nearly all negative) to V₃ (nearly all positive; Figure 3).^22,23 VAs from the AMC have a qR pattern in V₁ and positive QRS in V₂.²¹ As the site of origin moves out to the anterior mitral annulus, V₁ is completely positive. While there are typically monophasic R waves in the precordial leads with VAs from the AMC, there is typically a small S wave in many precordial leads with VAs from the anterior mitral annulus.²⁴

Around 12% of outflow tract VAs may arise from the LV summit.²⁵ While LV summit VAs share many of the features seen in other outflow tract VAs, they may exhibit a QS complex in lead I, delayed time to reach the peak or nadir of QRS complexes, and early precordial transition in V₃.²⁶ They may also have ECG characteristics consistent with an epicardial origin such as a pseudodelta wave ≥34 ms and QRS duration >200 ms.²⁷ Hayashi et al. also described a characteristic V₂ pattern break (R wave in V₂ is smaller than V₁ and V₃) that corresponded to a focus adjacent to the interventricular sulcus and anterior interventricular vein (AIV; Figure 3).²⁸ LV summit VAs that arise from the basal/lateral portion (accessible region) may be amenable to epicardial ablation. ECG characteristics that predict origin in the accessible region include: aVL/aVR Q wave ratio >1.740; R/S ratio >2 in lead V₁; and absence of q waves in V1.^26,29

LVOT VAs may also have an intramural source in the region of the LV summit and septum, with one study reporting an intramural origin in more than 20% of all patients.³⁰ The ECG characteristics of intramural LVOT VAs are more similar to those of endocardial VAs, with a smaller maximum deflection index and aVL/aVR Q wave ratio compared with summit VAs. Intramural VAs may also exhibit multiple exits in the LVOT, RVOT and LV summit, and the ECG morphology may change during ablation due to block of the preferential exit (Figure 5).³¹

Treatment of Outflow Tract Ventricular Arrhythmias

Given that these are typically benign arrhythmias, the decision to initiate treatment is often dependent upon the frequency and severity of symptoms. Medical therapy or catheter ablation is reasonable for significant symptoms from VAs. Intervention is also reasonable for sustained VT or PVC-induced cardiomyopathy. While there is a risk for developing a PVC-induced cardiomyopathy with a high burden of PVCs as discussed above, many patients with a high burden of PVCs will not develop a cardiomyopathy. In one study, only 13/239 (5.4%) patients with a high PVC burden and normal LV function at baseline developed LV dysfunction, defined as an ejection fraction (EF) drop of >6% during follow-up.³² Thus, such patients may be managed conservatively with close surveillance of PVC burden and LV size and function.

Medical Therapy

First-line medical therapy for outflow tract VAs consists of β-blockers or calcium channel blockers with efficacy of around 25–50%.^33,34 If these are ineffective, Class I or III antiarrhythmic drugs (AADs) may be used. In one study, flecainide and mexiletine suppressed PVCs in 91% and 55% of patients, respectively.³⁵ Flecainide and propafenone were also used safely in a small cohort of patients with refractory PVC-induced cardiomyopathy, with suppression of PVCs and improvement in EF in the majority of patients.³⁶ However, response to AADs can be variable, with two studies showing efficacy rates for suppressing PVCs of <15% for propafenone and d-sotalol.^37,38

Catheter Ablation

Catheter ablation is a good treatment option for outflow tract VAs, with a success rate of around 70–90%.³⁹ The success rate is highest (80–90%) for VAs originating in the RVOT. Catheter ablation is typically considered with symptomatic VAs or PVC-induced cardiomyopathy when AADs are ineffective, not tolerated, or not the patient’s preference.⁴⁰ Catheter ablation may also be considered as a first-line therapeutic option for RVOT VAs or sustained monomorphic VT.⁴¹ In one randomised study of patients with RVOT PVCs, ablation was associated with a much lower recurrence rate compared with metoprolol or propafenone (19.4% versus 88.6%).³⁸

Preparation

AADs should be stopped for at least five half-lives to allow for maximum inducibility of the VA during the procedure. General anaesthesia and deep sedation should be avoided because this may suppress the VA. In one study of PVC ablation, activation mapping was able to be performed in only 26% of general anaesthesia cases compared with 77% of cases using local anaesthesia ± minimal sedation.⁴² Acute success was also lower with general anaesthesia, 50% versus 85%. Thus, general anaesthesia and deep sedation should be avoided. If necessary, mild sedation and/or the use of a short-acting agent, such as propofol, to allow for rapid weaning in the setting of VA suppression is reasonable. Therapeutic plasma lidocaine levels can occur with subcutaneous administration, and local anaesthesia has been reported to result in VA non-inducibility during ablation.^43,44 Therefore, lidocaine should be used judiciously. Intracardiac echocardiography (ICE) is also helpful to define the anatomical structures and relationships in this region (e.g. RCC, LCC, NCC, left main coronary artery, aortic annulus, pulmonic valve), and ICE can also be integrated into the 3D anatomical map (CARTO; Biosense Webster).

Mapping

Ideally, activation mapping is the preferred technique to identify the site of origin. This should be performed with the ablation catheter or a multi-electrode mapping catheter during spontaneous VA. If PVCs or VT are not spontaneous, these may be induced with ventricular stimulation and/or catecholamine infusion. Isoproterenol is typically the agent of choice, and this may be titrated up to a dose of 20 μg/min. VAs may be seen during washout rather than during the infusion. Programmed stimulation may be repeated during isoproterenol infusion to facilitate induction. Epinephrine may be used if isoproterenol is unsuccessful at inducing the arrhythmia. In one study, a 10 μg injection of epinephrine induced PVCs in 53% of patients without PVCs at baseline or with isoproterenol.⁴⁵ At the site of origin, the local bipolar electrogram (EGM) typically precedes the QRS onset of the VA by ≥20–30 ms, and there is a QS on the unipolar EGM (Figure 4C). A local activation time preceding the QRS by >30 ms is predictive of successful ablation.⁴⁶ Occasionally, an abnormal, late EGM may be seen during sinus rhythm that becomes early during the VA (Figure 4C). If activation mapping results in a diffuse area (≥3 mm) of early activation, then the site of origin is likely to be from another location (e.g. another chamber, intramural).⁴⁷ Additional mapping prior to ablation should be performed in this situation. Care must also be taken to differentiate catheter ectopy from the clinical VA, given that this can lead to an inaccurate activation map.

In cases in which the VA is non-inducible or the VA suppresses prior to completion of mapping, pace mapping may be used to guide ablation. A study recently compared outcomes with pace mapping-guided ablation to activation mapping-guided ablation in patients with predominant RVOT VA.⁴⁸ Pace maps were analysed using automated systems (PASO, CARTO, Biosense Webster; or Score map, EnSite Precision, Abbott Medical), and regions with a high (≥90%) pace map correlation were targeted for ablation. The median best pace map was a 96% match to the clinical VA. At 6-month follow-up there was no difference in outcomes, with 77% free of VA in the pace mapping group and 71% in the activation mapping group. While this approach appears to be effective for RVOT VA, pace mapping in the LVOT is less reliable due to variable conduction and exit from the LVOT, and may be associated with worse ablation outcomes.^49,50 Activation mapping remains the strategy of choice for LVOT VAs.

Mapping for intramural foci and LV summit VAs can be more complex and often requires mapping of the RVOT, LVOT and coronary venous system. In these cases, ventricular activation may not be significantly early (i.e. <20 ms), and there may be similar activation times in multiple chambers with broad areas of early activation.⁴⁶ EGM characteristics may also differ from endocardial sources. The local activation on a bipolar EGM occurs at the first sharp deflection, and this corresponds to the steepest slope (dV/dT) on the unipolar EGM. However, for intramural sources, activation at the intramural focus may correspond to the far-field component of the EGM, and the onset of the EGM may be most relevant in this situation. Sabzwari et al. compared EGM characteristics of RVOT VAs versus intramural VAs requiring ablation in the RVOT and LVOT.⁵¹ At successful ablation sites, the mean interval between the onset of the bipolar EGM and the steepest dV/dT on the unipolar EGM was longer for intramural VAs than for RVOT VAs. In addition, 43% of intramural foci did not exhibit a QS unipolar EGM.

The coronary venous system (GCV, AIV and branches) should also be mapped for intramural and/or LV summit VAs, given that the communicating veins supply the LV summit region (Figures 1 and 6). Baman et al. showed that an epicardial or intramural site of origin was found via mapping of the venous system in 14% of patients with outflow tract VAs.⁵² The ablation catheter is inserted into the GCV, and the sheath is advanced into the coronary sinus over the catheter. Cold (room temperature) saline may be injected through the ablation catheter near the GCV–AIV junction, with flow increased from 2 to 60 ml/min for 10 seconds. In one study, suppression of PVCs occurred in 11 of 26 patients, and suppression with cold saline was associated with 88% accuracy for an intramural focus.⁵³ The ablation catheter is then removed, and an angiogram is performed to outline the venous anatomy and branches (Figure 6).⁵⁴ The venous system may be mapped with a 2 Fr microelectrode catheter (EP star, Baylis Medical or MapIT, APT EP). In order to map the venous branches, an angioplasty wire is often required. A 6 Fr guide catheter, internal mammary artery or the JR4 (Judkins Right 4), can be advanced into the coronary sinus over a wire. A 0.014 inch balanced middleweight wire (BMW, Abbott) may then be advanced through the guide catheter. The wire is then covered with an angioplasty balloon, leaving the terminal 5 mm exposed for recording. Alternatively, a coated wire with an electrically active tip may be used (VisionWire, Biotronik). An alligator clip may be connected to the wire, and the other clip can be connected to a needle placed in the thigh to serve as the anode for unipolar recording. Activation mapping as well as pace mapping can be performed via the wire (Figure 6).

If mapping of the venous system does not result in a suitable target, then epicardial access and mapping may be performed if the ECG is suggestive of an exit in the accessible region of the LV summit (see prior discussion). This is often not required, given that epicardial access was needed in only 2 of 27 patients in the study by Baman et al.⁵² In addition, Nagashima et al. reported epicardial access in 13 patients with VA from the LV summit and distal great cardiac vein, and epicardial ablation was performed in only two patients.⁵⁵

Ablation

Right Ventricular Outflow Tract

For VAs from the RVOT, mapping may be performed with an ablation catheter or multi-electrode catheter as above. Given that 11% of RVOT VAs may originate in the pulmonary sinuses, the pulmonic root should also be mapped if very early sites are not identified in the RVOT.²⁰ If earliest activation is <20 ms, there is a broad area of earliest activation suggesting breakthrough from the left side, and/or a QS unipolar EGM cannot be identified, then additional mapping of the LV should be performed prior to ablation.

Ablation in the RVOT is generally performed via an open irrigated tip ablation catheter with normal saline at the site of earliest activation or the best pace map. Power may be applied at 20–35 W. Impedance should be monitored closely with a goal impedance drop of at least 10 Ω. Radiofrequency (RF) ablation is terminated if there is a sudden or large (>20%) impedance drop or rapidly expanding echogenicity seen on ICE, given that this may precede a steam pop.⁴⁶ If ablation is performed in the left pulmonary sinus, coronary angiography is recommended first due to proximity of the left main coronary artery.

Left Ventricular Outflow Tract

For left-sided mapping and ablation, a retrograde aortic or transseptal approach may be used, given that arrhythmias may be successfully ablated from above or below the valve.⁷ Retrograde aortic access allows for mapping and ablation in the aortic root. For mapping below the valve with this approach, a catheter inversion technique is useful in this region. The catheter is curled back on itself in the aorta, and the catheter is then advanced across the valve into the LV. The catheter tip is placed just below the aortic annulus, and the regions below the aortic sinuses can be reached by rotating the catheter clockwise and counterclockwise. Deflection can be released to achieve adequate contact force. However, it can be challenging to map the entire LVOT below the AV with this approach, and gaps in mapping of up 12 mm from the aortic annulus to the LVOT were found in one study.⁵⁶ There also may be an increased risk of thromboembolism with the retrograde approach.^57,58 Thus, a transseptal approach may also be used to reduce the risk of thromboembolism and for potentially better access to the LVOT. However, this approach does not allow for mapping and ablation in the aortic root, and it can be technically challenging to reach the LVOT because the catheter has to be placed in a ‘reverse S’ configuration.⁵⁶ A steerable sheath (Agilis, large curl, Abbott) can aid with catheter manipulation.

When ablating in the aortic root, the left coronary ostium for ablation in the LCAS and right coronary ostium for ablation in the RCAS should be defined, and ablation should be performed only ≥5 mm from the coronary ostium to avoid damage and potential myocardial infarction. Using 3D mapping and ICE alone to define the coronary anatomy has been shown to be safe without the need for coronary angiography (Figure 5).⁵⁹ In the LVOT, power is generally set at 30–40 W, and lesion duration may be extended to 60 seconds. Power and lesion duration may be uptitrated if necessary. If ablation is required in the GCV–AIV junction, power is started at 10–15 W and gradually uptitrated to 20–25 W due to high impedance. The high impedance cut-off may need to be increased to enable current delivery. Given that the coronary arteries are typically close to the GCV–AIV junction (Figure 1), coronary angiography prior to ablation is reasonable. In one study, the earliest site of activation in the distal GCV was ≤5 mm from a coronary artery in 74% of cases.⁵⁵ In eight of these patients, ablation was attempted 2–3 mm proximal to the target site, with VA suppression in three patients.

Ablation of Intramural Foci and Left Ventricular Summit VAs

LV summit VAs often have an intramural origin. Ablation of these foci can be challenging and may be associated with lower success rates due to difficulty in reaching the intramural substrate.³⁹ However, success rates for ablation of LV summit arrhythmias appear to be increasing over time. In one study, ablation success increased from 65% between 2009 and 2012 to 93% between 2016 and 2019.⁶⁰ There are several strategies and techniques that can improve the likelihood of procedural success.

Alteration of Ablation Parameters.

Half normal saline (NS) instead of NS may be used to achieve deeper lesions. The decreased ion concentration of half NS leads to increased current delivery to the myocardium with less current loss to the surrounding fluid, resulting in larger lesions.⁶¹ In a study of 94 patients with prior failed ablation in which more than half had VAs involving the interventricular septum or LV summit, the use of half NS resulted in acute and 6-month success in 83% of patients.⁶² Of note, steam pops occurred in 12.6% of patients. While there were no complications observed from the steam pops, impedance should be closely monitored during ablation, and starting at 30–35 W with gradual titration of power if necessary is reasonable. High power and prolonged lesion duration can also result in deeper lesions and ability to reach the intramural substrate. There have been case reports of using 50 W with lesion duration up to 3 minutes to ablate epicardial VT circuits from the endocardial surface.^63–65 Moving the dispersive patch electrode from the back to the anterior chest may also result in higher current density to the LV summit and larger lesions.⁶⁶

Anatomical Approach to Ablation

Intramural VAs in this region may originate from the anterior interventricular sulcus as described previously. The interleaflet triangle between the RCC and LCC commissure (R-L ILT) is the anatomical endocardial vantage point to the septal LV summit and is a target for ablation. Liao et al. showed that in 20 patients (65% with prior failed ablation) with VAs with LB in V₁, inferior axis, and abrupt V₃ transition, 80% were ablated from the R-L ILT.²³ Moreover, three patients with an intramural focus were all successfully ablated from the R-L ILT. In four patients with earliest activation in the GCV–AIV junction, three were ablated successfully from the R-L ILT, and one was ablated from the LCAS. Thus, ablation at the R-L ILT may be successful despite later activation at this site. Similarly, Kapa et al. also reported successful ablation of VAs with variable morphologies (RB with positive concordance in 78%, transition in V₂ in 14%, and transition in V₃ in 8%) in 51 patients just below the LCAS. Notably, in 39% of patients, the site of earliest activation was different to the site of successful ablation below the LCAS.⁶⁷ Given these findings, an anatomical approach to ablation has also been described when earliest activation is in the GCV. Ablation is initially performed at the endocardial site with the shortest distance from the epicardial site of earliest activation. If ablation at that site is unsuccessful, then ablation is performed at the site with earliest endocardial activation or at sites that have previously been reported to be successful, regardless of activation timing. At a single centre, ablation success increased significantly over time (93% between 2016 and 2019) for LV summit arrhythmias, and this coincided with the increasing use of the anatomical approach to ablation.⁶⁰

Ablation of Multiple Sites in the Outflow Tract

A subset of patients may present with multiple PVC morphologies due to multiple exits from an intramural source in the LV. The exit sites can be in the RVOT, aortic root, LVOT and GCV–AIV junction with corresponding ECG morphologies (Figure 5). Ablation at each of these sites may be required to achieve arrhythmia suppression. Di Biase et al. reported that among 116 patients with outflow tract VAs, 15 (13%) were found to have multiple sites of equally early activation.⁶⁸ In those 15 patients the mean pre-QRS activation time was shorter compared with patients with a single site of early activation (-26 ms versus -38 ms). Ablation of all early sites was possible in 14 of 15 patients (93%) and resulted in arrhythmia suppression. Thus, ablation may need to be performed from multiple sites to achieve success, and outlines the close anatomical relationships in the outflow tract (Figures 1 and 5).

Transcoronary Venous Ethanol Ablation.

As described previously, venous branches off the GCV and AIV junction supply the LV summit. The site of earliest activation may be in these branches, and these are targets for ablation. Tavares et al. reported outcomes of transcoronary venous ethanol ablation in 63 patients with refractory VAs, with 48 patients with VT arising from the LV summit.⁶⁹ Nearly 89% of patients (56/63) had veins suitable for ethanol ablation. For LV summit patients, the first septal branch was targeted the most (36 patients), followed by the LV annular vein (five patients; Figure 6). In 38 of 56 patients (68%), the VA terminated with ethanol ablation only. In 17 of 56 patients (30%), adjunctive RF ablation at endocardial sites adjacent to the infused vein was required to achieve success, and this demonstrates that venous mapping may also provide an endocardial target for ablation. Seventy-seven per cent of patients were free from arrhythmia at 1-year follow-up. Of note, there were two pericardial effusions due to venous dissections. There were no reported cases of conduction system damage or heart block.

To perform transcoronary venous ethanol ablation, an over-the-wire balloon (1.5–2 mm × 6 mm) is inserted into the branch of interest over an angioplasty wire. The balloon is inflated, and contrast is injected to define the vein. If the vein is large or has multiple collaterals, a double balloon approach may be needed.⁷⁰ One balloon is placed distally in the vein and inflated. A second balloon is placed proximally and inflated. Thus, when ethanol is injected, it is directed only to the targeted area. A 1 ml injection of ethanol is administered with monitoring of clinical response. Contrast is then administered to check for myocardial staining. Up to seven injections can be administered, but the mean amount of ethanol given in the prior study was 4 ml. It is important to note that in contrast to transcoronary arterial ethanol ablation, reflux of ethanol into the coronary sinus is of no clinical consequence. In some cases, the venous branch may be too small to allow for balloon inflation, and ethanol may be injected without inflating the balloon.

Adjunctive Ablation Techniques.

Simultaneous unipolar ablation involves using two ablation catheters with two generators and applying RF energy on opposite sides of the myocardium. This technique results in greater lesion depth compared with sequential unipolar ablation.⁷¹ In a study of 14 patients with intramural LVOT VAs, five required simultaneous endocardial and epicardial unipolar ablation to achieve success.⁷² Bipolar ablation uses one catheter connected to the generator and a second catheter connected to a spliced RF connector to a ground plate that serves as the ground catheter. Bipolar ablation results in higher current density and deeper lesions. In a study of 20 patients who failed unipolar ablation, bipolar ablation was performed from the GCV–AIV junction and the RVOT or LVOT.⁷³ Acute success was achieved in all patients and, notably, no steam pops were observed. Prior experimental studies reported a high incidence of steam pops, and the authors postulated that their protocol of gradual titration of power resulted in improved safety.⁷⁴ Multipolar ablation using two electrodes of a multi-electrode catheter in the GCV was also used in this case series and has been previously described.⁷⁵ This has the benefit of targeting the area with the best bipolar EGM, but the safety of this approach is yet to be established. Intramyocardial mapping and RF ablation via an intracoronary wire have also been described. Pulsed field ablation may result in deeper lesions, but its use in the treatment of VAs has not yet been well studied.⁴⁶

Conclusion

The outflow tract region is the most common site for idiopathic VAs. These arrhythmias can result in significant symptoms and cause a cardiomyopathy, and catheter ablation is often required to treat these VAs. A detailed understanding of ECG localisation and the complex anatomy of the region is paramount for pre-procedural planning. Thorough mapping of the entire region including the coronary venous system is needed to identify the site of origin. For intramural foci and the LV summit, anatomical ablation rather than ablation guided by activation, as well as advanced techniques such as transcoronary venous ethanol and bipolar ablation, may be required for procedural success.