This was a single-center retrospective study including consecutive patients with SHD undergoing VT ablation from 2012 to 2023. Only those with scar-related VT of either ICM or dilated cardiomyopathy/non-dilated left ventricular cardiomyopathy etiology were considered. Alternative cardiomyopathies, including Chagas disease, sarcoidosis, right ventricular arrhythmogenic cardiomyopathy, hypertrophic cardiomyopathy, and acute myocarditis were excluded. In all cases, VT was documented using cardiac implantable electronic devices, 12-lead electrocardiogram (ECG), or external defibrillator monitoring. The definitions of ischemic and non-ischemic cardiomyopathy were according to current international guidelines.15,17 Electrical storm was defined as the occurrence of ≥3 episodes of sustained VT or ventricular fibrillation during a 24-hour period. This study was according to the amended Declaration of Helsinki. All patients signed an informed consent, and the registry had been previously approved.30

All patients underwent the procedure under conscious sedation or general anesthesia according to the operator and anesthesiologist's judgment. Whenever clinically appropriate, anti-arrhythmic drugs (AADs) were withheld at least 48 hours before the VT ablation. Anticoagulants were stopped 24–48 hours prior depending on renal function and/or INR level (if under vitamin K antagonist) and resumed within the first 24 hours in the absence of any hemorrhagic complications. The choice of the ablation access (endocardial, epicardial or combined) was left to the operator's discretion and decided according to etiology, previous ablation site, and imaging information regarding scar location.

A decapolar steerable catheter was placed in the coronary sinus via the femoral vein and a quadripolar non-steerable catheter was placed in the right ventricular apex. Left ventricular (LV) endocardial mapping was performed either via retrograde aortic or transseptal route according to the operator's discretion. After LV access was obtained, systemic anticoagulation with unfractionated heparin was administered, targeting a minimum activation clotting time of 300 s during endocardial ablation. Anticoagulation was reversed with protamine sulfate (1 mg/100 U) at the end of the procedure or before percutaneous epicardial access, when needed. If clinical or preprocedural imaging features suggested an epicardial circuit, access was obtained through the subxiphoid space under fluoroscopic guidance as previously described by Sosa et al.31

Electroanatomical maps were obtained with the CARTO® (Biosense Webster, Diamond Bar, CA, USA), EnSite NavX® (Abbot, St Paul, MN, USA), or Rhythmia® (Boston Scientific, Marlborough, Massachusetts, USA) systems and included activation and voltage mapping acquired in sinus rhythm, ventricular pacing or during hemodynamically stable VT. Mapping catheters used included a 20-electrode PentaRay® catheter (Biosense Webster), a 16-electrode HDGrid® catheter (Abbot), or a 64-electrode Intellamap Orion® basket catheter (Boston Scientific), respectively. Catheters were supported by an Agilis® steerable sheath (Abbot) during anterograde mapping.

An attempt to induce the clinical VT was performed at the beginning of each procedure according to the clinical characteristics of the patients and the operator's discretion. Clinical VT was defined as an inducible VT with a similar cycle length (CL) of stored ICD electrograms and/or the 12-lead ECG morphology and CL when available. When hemodynamically tolerated, activation and entrainment mapping were used to define critical components of the circuit. Substrate mapping was performed during normal sinus rhythm or pacing. Normal tissue was defined by a bipolar voltage threshold greater than 1.5 mV, while dense scar as <0.5 mV (in extensive areas of low voltage, a cut-off of 0.3 mV – or as low as 0.1 mV – was used to better discriminate slow conduction zones). Local abnormal ventricular activities (LAVA) were defined as sharp high-frequency ventricular potentials, distinct from the far-field ventricular electrogram, occurring during or after the ventricular electrogram.32

Radiofrequency (RF) was delivered with a 4 mm irrigated and contact force monitoring ablation catheter in power control mode, with power set to 30–50 W and irrigation set to 17–30 ml/min. In one patient with ICM, pulsed-field ablation (FARAPULSE®: FARAWAVE catheter and FARADRIVE steerable sheath; Boston Scientific) was also utilized. In the presence of sustained and tolerated monomorphic VT, isthmus was the first target for ablation. If clinical VT was not inducible or mappable, substrate modification was performed in sites in which abnormal electrograms were identified.

The procedure was considered acutely successful if no VT was induced with a standard stimulation protocol of up to three extrastimuli at a cycle length of down to 200 ms or ventricular effective refractory period from two different sites, excluding induction of VF or polymorphic VT.30 Inducibility testing was withheld if patients were hemodynamically unstable, or if repeated cardioversions were required for hemodynamically unstable VT during the procedure.

The primary outcome of interest was VT-free survival. VT recurrence included any ICD-appropriate therapy (anti-tachycardia pacing or shock), or a documented sustained VT not detected by the ICD. The secondary endpoint was all-cause mortality.

The follow-up comprised regular outpatient visits according to the assistant physician's discretion. Outcomes were collected from patients’ electronic medical records, ICD home monitoring software, and a structured telephonic interview whenever necessary. All patients without an ICD underwent implantation before discharge. After the ablation, all ICD were programmed with a detection zone slower than the clinical VT. Vital status was determined for each patient using the Portuguese National Registry.

Categorical variables were reported as numbers and percentages. Continuous variables were described as means and standard deviations for normally distributed variables and medians and interquartile ranges for non-normally distributed variables. Clinical characteristics of the subgroups of interest were compared using the χ2-test and Fisher's exact test (when applicable) for dichotomous variables; and the unpaired Student's t-test or Mann–Whitney U test (when applicable) for continuous variables. A propensity score (PS) was applied to match each patient of the ICM cohort with a similar subject in the NICM group, in a 1:1 fashion. The following baseline variables, known to impact the results and safety of the procedure, were included the propensity score model: age, sex, left ventricular ejection fraction (LVEF), New York Heart Association (NYHA) class, electrical storm (ES) at presentation, and previous endocardial VT ablation.33–36 Patients were matched on the logit of the propensity score using calipers of width equal to 0.1 of the standard deviation. Anti-arrhythmic drugs were not included in the main analysis as time points for their collection were different in each group. Kaplan–Meier survival curves were plotted to compare groups, and the log-rank test was used to assess for significant differences in time to endpoint between groups. Cox proportional hazards regression was applied for univariable and multivariable analysis to investigate the association between clinical variables with VT recurrence. Variables with a p<0.05 or theoretically relevant (domain knowledge) were included in the multivariable model. A multivariate Cox regression model for the outcome of VT recurrence in the entire non-matched population was performed as a sensitivity analysis. A two-tailed p<0.05 was considered statistically significant for all tests. All analyses were performed using SPSS v26.0 (IBM Corporation, Armonk, New York) and STATA v13.0 (StataCorp, College Station, Texas).

Of the included 248 patients, the PS model yielded two groups of 71 patients each, well matched for baseline variables. Briefly, the mean age of the included population was 63±10 years, 92% (n=131) were male, mean left ventricular ejection fraction (LVEF) was 35±10% and 36% (n=51) of patients had electrical storm (ES) at presentation. A total of 23% (n=33) had history of previous endocardial ablation. The baseline characteristics are presented in Table 1.

The characteristics related to the ablation procedure are depicted in Table 2. Patients with NICM were more likely to undergo either epicardial or combined VT ablation (43.7% vs. 9.9%, p<0.001). In the 114 patients in which programmed ventricular stimulation was performed at the end of the procedure, non-inducibility of any VT was achieved in 76% (n=87). Non-inducibility was numerically higher in the ICM cohort (50/71 patients vs. 37/71), despite not meeting statistical significance (p=0.076).

A total of 92 patients (65%) were hospitalized at the time of the ablation, whereas the remaining had their procedures scheduled in an ambulatory setting. The median duration of hospital stay was two (interquartile range (IQR) of 2–7) days. Overall, the procedural complication rate was low (7.0%, n=10) and similar between groups (p>0.50) – Supplementary Table 1.

During a median follow-up of 2.3 (IQR 1.3–3.8) years, 39% (n=55) of patients had VT recurrence, 4% (n=6) underwent heart transplantation or long-term left ventricular assist device (LVAD), and 20% (n=28) died. A total of two patients died during the same hospital admission, none related to procedure-related complications.

Patients with NICM had a significantly higher rate of VT recurrence (26.9%/year vs. 14.2%/year, log-rank p=0.037; HR 1.77 [95% CI 1.03–3.05], p=0.039), although there were no statistically significant differences regarding the outcome of all-cause mortality between groups (9.6%/year vs. 6.2%/year, log-rank p=0.245; HR 1.56 [95% CI 0.76–3.31], p=0.249) – Figure 1.

Figure 1.

Kaplan–Meier curves for overall VT-free survival (panels A and B) and all-cause mortality (panels C and D), plotted according to VT etiology.

(0,25MB).

In the multivariate analysis, including baseline clinical characteristics, NICM remained an independent predictor of time-to-VT recurrence (adjusted HR 2.34 [95% CI 1.32–4.14], p=0.004), even after adjusting for the presence of ES at presentation, heart failure (HF) functional class, LVEF and chronic kidney disease (CKD). NYHA class III or IV (adjusted HR 2.11 [95% CI 1.11–4.04], p=0.024), and CKD (adjusted HR 2.23 [95% CI 1.25–3.96], p=0.006) were the other independent predictors of VT recurrence (Table 3).

Regarding the procedural characteristics, non-inducibility of any VT immediately after ablation (i.e., acute ablation success) was associated with significantly lower rates of VT recurrence during follow-up (15.8%/year vs. 33.0%/year, log-rank p=0.03 – Supplementary Figure 1), despite no differences in all-cause mortality (8.2%/year vs. 7.6%/year, log-rank p=0.949).

The baseline characteristics of the entire non-matched cohort before PS are depicted in Supplementary Table 2. The predictive value of NICM etiology for time-to-VT recurrence was similar in a sensitivity multivariate model including the entire cohort before matching: aHR 1.87 (95% CI 1.19–2.93), p=0.006 (Supplementary Table 3). In this model, ES (HR 1.77 [95% CI 1.15–2.70], p=0.009), reduced LVEF <35% (HR 1.86 [95% CI 1.17–2.95], p=0.009) and CKD (HR 1.64 [95% CI 1.06–2.54], p=0.027) were the other independent predictors of recurrence.