This was an independent, single-center, prospective study that was designed to evaluate the feasibility, efficacy, and safety of a hybrid thoracotomy approach with epicardial substrate mapping and ablation in patients with BrS. Herein, we report the results of the first six patients enrolled in the UNCOVER(BrS) study (NCT05643209), which was conducted at Maria Cecilia Hospital (GVM Care & Research; Cotignola, Italy). The study was approved by the local ethics committee, and written informed consent was obtained from all patients.

Patients diagnosed with a spontaneous type-1 Brugada ECG pattern, induced after an ajmaline infusion, and who had undergone an ICD implantation were screened for symptomatic events. After ICD implantation, patients with syncope events and/or documented ventricular arrhythmia were considered symptomatic. Eligible patients were selected for ablation with a 3D high-density mapping system in accordance with current guidelines and medical decision-making, and they were enrolled in the study after providing informed consent. The exclusion criteria included the inability to provide written informed consent, the inability to comply with follow-up visits, a life expectancy of less than 12 months based on clinical assessment, prior cardiac ablation within 90 days before enrollment, and pregnancy.

The primary endpoint was the survival-free rate of symptomatic events related to episodes of ventricular arrhythmia at the 6- and 12-month follow-ups after the procedure. Symptomatic events were defined as documented episodes of ventricular arrhythmia and/or episodes of arrhythmic syncope.

The secondary endpoint was to identify the most appropriate mapping configuration to describe the conduction properties of the arrhythmogenic substrate in BrS. Data from late potentials and voltage maps, acquired using different settings (BiEGM vs. OTEGM), conduction vectors, and wave speed maps, were analyzed to provide a detailed electrical characterization of abnormal myocardial tissue.

The hybrid mini-thoracotomy approach for epicardial ablation in patients with BrS has been previously described (19, 20). Briefly, pericardial access was obtained by a cardiac surgery team via left anterior mini-thoracotomy, with the patient in a supine position under general anesthesia. A 3–4-cm skin incision was made either over the third left intercostal space or periareolarly. The subcutaneous tissue and muscle layers were then dissected using electrocautery until the upper edge of the rib was reached. After dividing the external and internal intercostal muscles, ventilation of the left lung was temporarily suspended, and a tissue retractor was used to displace the lung, allowing direct visualization and access to the pericardial space above the RVOT. Epicardial mapping and ablation were performed by the electrophysiology team. The catheters were introduced through the thoracotomy access. The open surgical field enables direct visualization of the anterior RV wall, allowing the operator to manipulate the catheters manually using their fingers or forceps (Figure 1).

Initial mapping was performed using a contact force catheter (TactiCath™ or TactiFlex™ Sensor Enabled™, Abbott, Chicago, IL, USA) in Voxel mode in the mapping system (EnSite™ X, Abbott, Chicago, IL, USA) to delineate the epicardial surface boundaries. Next, high-density mapping was performed using a multipolar grid catheter (Advisor™ HD Grid Sensor Enabled™, Abbott, Chicago, IL, USA) during sinus rhythm to acquire voltage and potential duration bipolar maps of the anterior epicardial substrate of the RVOT. Automatic software annotation (last deflection) was used and manually verified.

Following an ajmaline infusion (1 mg/kg over 5 min), a second high-density map was acquired to identify the BrS substrate. Abnormal signals were evaluated based on voltage amplitude, fractionation, signal duration, and morphology. Substrate analysis aimed to identify conduction block zones and areas of electrical dispersion. The identified pathological area was tagged and delimited for ablation.

Substrate homogenization was performed using a contact-force ablation catheter with a point-by-point technique, applying radiofrequency energy at a power limit of 40 W and a maximum tip temperature of 43°C. The target inter-lesion distance was 6 mm, with each lesion delivered for 20 s under stable contact. RF applications were delivered over the entire pathological area previously identified and delineated in the mapping system.

Post ablation, ajmaline was re-administered to confirm complete substrate elimination. Any residual abnormal potential was targeted with additional RF applications until complete substrate abolition was achieved.

The patients remained hospitalized for at least 3 days post-procedure for monitoring purposes. Before discharge, clinical (vital signs and therapy adjustments) and instrumental evaluations (12-lead ECG and echocardiography) were performed. Follow-up visits were scheduled at 3, 6, and 12 months, assessing symptoms, adverse events, vital parameters, 12-lead ECGs, echocardiograms, 24 h Holter monitoring, and ICD interrogation. The ajmaline test was repeated at 6 and 12 months to evaluate for any recurrence of the Brugada ECG pattern. No additional invasive electrophysiology studies were performed during follow-up.

Continuous variables are presented as a median and interquartile range due to the small sample size. Continuous variables were compared using the Mann–Whitney U test and the Wilcoxon signed-rank test. Categorical variables are expressed as a frequency or percentage. Categorical variables were compared using the chi-squared test. P-values  < 0.05 were considered statistically significant. The statistical analysis was performed using SPSS V.22.0 (IBM).

A total of 228 patients were screened for BrS. Of these, 62 (27%) underwent ICD implantation (Class Ic¹) and were monitored for symptomatic episodes. Eight patients experienced at least one symptomatic event, including syncope and/or ventricular arrhythmia. Six patients consented to and underwent catheter ablation (Figure 2; Supplementary Table S1).

Two patients received subcutaneous ICDs (S-ICDs) and four patients received transvenous ICDs. The median age at ICD implantation was 53 years (range: 34–70). ICD implantation followed a comprehensive evaluation for a characteristic type 1 BrS ECG pattern, which was either spontaneous or unmasked during an ajmaline infusion, and risk stratification for arrhythmia.

Programmed ventricular stimulation was performed as part of the risk stratification process. Inducible VA with two extra stimuli was observed in three patients (patients #3, #4, and #5). One patient (patient #5) carried a pathogenic SCN5A gene mutation. Two patients (patients #1 and #2) had a family history of BrS and had experienced documented episodes of polymorphic ventricular tachycardia during 7-day Holter ECG monitoring, although arrhythmia was not inducible during PVS.

All patients exhibited a type 1 ECG pattern during the ajmaline infusion, with a prolonged dST-Tiso interval >300 ms [median: 320 ms (IQR: 310–340 ms)].

All patients underwent epicardial mapping and ablation via a mini-thoracotomy approach. The median skin-to-skin procedure duration, including surgical time, was 200 min (IQR:157–260) (Figure 3). Surgical time was consistent across cases [median: 50 min (IQR:50–55)], whereas the electrophysiological operating time progressively decreased [median: 150 min (IQR:115–225)].

All patients underwent direct epicardial mapping under baseline conditions, after an ajmaline infusion, and following RF application.

In the baseline conditions, patchy low-voltage areas (<1 mV) were identified in 50% of patients with a median extension area of 4.0 cm² (IQR: 2.2–7.6). Late potentials (outside the QRS complex) were observed in an area of 0.5 cm² (IQR: 0–3). The isochronal late activation mapping (ILAM) showed no deceleration zones (defined as >3 color bands per linear cm), and no conduction blocks were present. The fragmentation maps revealed no signals with >2 components (Figure 4C), and the wave speed maps demonstrated normal conduction velocities (>1 mm/ms) across the epicardial surface (Figure 4A).

Following the ajmaline infusion, all patients exhibited one or more areas with late, fragmented, long-lasting multicomponent electrograms with a median extension of 10.3 cm² (IQR: 3.7–12.5). Signal fragmentation increased by up to five deflection components in the fragmentation maps (Figure 4D). Moreover, while under the effect of the drug, conduction block lines emerged, replacing the previously linear propagation pattern. Deceleration zones appeared on the ILAM maps, and the wave speed maps revealed localized zones of slowed conduction (<1 mm/ms; rainbow-colored zones in Figure 4B).

Pathological substrate identification was based on an OTEGM LAT map analysis, targeting regions with long-lasting fragmented signals, deceleration zones, and block lines post-ajmaline. The median area targeted for RF ablation was 6.6 cm² (IQR: 4.2–8.9). Figure 5 illustrates the ablation target zones (dark lines) overlaid on LAT maps showing prolonged electrograms.

Point-by-point RF applications were delivered at 40 W for a median duration of 14 min (IQR:10–17) to achieve full pathological substrate elimination (voltage reduction <1 mV) over an epicardial area of 11 cm² (IQR: 8–15) (Figure 5). Due to direct catheter manipulation close to the ablation tip, a median contact force of 19 g (IQR: 13–30) was achieved (Supplementary Table S2).

Following RF ablation, complete epicardial remapping was performed after an ajmaline infusion to confirm pathological substrate abolition. In cases where residual pathological OTEGM signals were detected, additional RF applications were delivered until complete abolition was achieved. The ablation endpoint was defined as the absence of prolonged fragmented electrograms across the RVOT’s epicardial surface.

Post-procedural remapping was performed in all cases using a standard bipolar configuration with the Advisor™ HD Grid, collecting BiEGMs along and across the splines to emulate conventional linear catheter mapping. The bipolar maps were less dense than the OT maps, with fewer acquired and displayed points (BiEGM: 734 vs. OTEGM: 1503, p < 0.094).

Standard bipolar mapping resulted in a poorer substrate delineation, requiring more human intervention due to false annotations. For several electrocardiograms, the BiEGMs failed to detect the latest and most fragmented signal components, which were clearly identified with OTEGMs (Figure 6). In addition, the OTEGM conduction vectors facilitated the identification of conduction blocks, while the wave speed maps highlighted zones of signal wavefront deceleration, data that were not available when using BiEGMs.

Complete substrate homogenization was achieved in all patients. After the post-ablation ajmaline infusions, no BrS ECG patterns reappeared. All procedures were completed without an acute adverse event.

A high impedance alert was the main cause of premature RF delivery interruption, occurring a median of 4.5 times per patient (IQR: 3.5–5.5). These interruptions were attributed to a transient loss of tip-to-tissue contact or discontinuation of saline irrigation in the open surgical field. RF delivery was resumed in all cases without clinical consequences.

The patients were discharged after a median hospitalization of 7.5 days [IQR: 6–9] with no major adverse events. Four patients experienced minor and fully reversible complications (pleural effusion and transient pericarditis) during their hospital stay.

All patients completed a structured follow-up protocol with in-person evaluations at 3, 6, and 12 months post-ablation. At the 12-month visit, all patients remained in a stable sinus rhythm without spontaneous recurrence of the type I Brugada ECG pattern on standard 12-lead electrocardiography. Serial device interrogations throughout the follow-up period confirmed the absence of ventricular arrhythmias or ICD therapies. Repeat ajmaline provocation testing performed during follow-up demonstrated sustained elimination of the Brugada ECG pattern in all cases. No additional invasive electrophysiological studies were performed to reassess VT/VF inducibility.

To the best of our knowledge, this is the largest case series describing a novel hybrid mini-thoracotomy approach for substrate mapping and ablation in patients with BrS. Herein, we report on six selected patients who underwent an electrophysiologically guided surgical epicardial procedure, enabling direct visualization of the RVOT’s epicardial surface without the use of fluoroscopy or other video-assistance devices. Our preliminary findings suggest the following: 1.

The approach is technically feasible for identifying and eliminating the pathological BrS substrate. After an initial learning curve, the duration of the procedure is comparable to that of other techniques of epicardial ablation.

The hybrid mini-thoracotomy approach for epicardial ablation in Brugada syndrome has previously only been described in case reports (19, 20). The UNCOVER(BrS) study provides the first structured evidence of its feasibility, safety, and efficacy in a small but well-characterized cohort.

A rapid learning curve was observed, with procedural refinements addressing common technical challenges such as EGM noise and catheter instability (21). Procedural times, which vary widely in the literature, are reported in detail to support reproducibility (18, 21).

The procedure achieved complete substrate homogenization and ECG normalization in all patients. While the long-term impact of different ablation strategies remains to be determined, our point-by-point ablation approach resulted in electrical silence of <0.1 mV across the majority of the treated areas. This contrasts with other studies that report that RF application via the dragging technique does not significantly affect voltage amplitude or local activation times (15) or studies that lacked information on the substrate after ablation (16, 18). Our findings align with recent data showing a significant reduction in signal amplitude following ablation (22).

Recently, Namadee et al. demonstrated in a randomized clinical trial that epicardial substrate ablation significantly reduces VF recurrence in patients with symptomatic BrS who have an ICD, supporting its potential role as a first-line therapy to prevent recurrent VF in patients with symptomatic BrS (23). Although the long-term efficacy of the mini-thoracotomy ablation approach is still being investigated, early follow-up data are promising, with no episodes of arrhythmia reported in the study population. The technique also demonstrated a favorable safety profile with minor procedural complications. Despite the smaller sample size and shorter follow-up, our findings align with recent evidence (23). If these results are confirmed in larger cohorts with longer follow-up periods, the mini-thoracotomy approach may represent a viable alternative to the percutaneous technique.

The characterization of the BrS epicardial substrate is still controversial and lacks standardization. In this study, we proposed a feasible and reproducible method for substrate assessment. OTEGMs provided additional information that was not available with standard bipolar EGMs (orientation-independent signals, maximum signal voltage, and conduction velocity) and showed improved substrate delineation performance with lower annotation errors and higher point density.

Our findings contribute to addressing the current gap in detailed electrophysiological information on surgical techniques (18, 21). The extent of the substrate still needs to be clarified. We observed significant inter-patient variability in substrate morphology and extent with the ajmaline infusion unmasking highly fragmented long-lasting EGMs, the formation of chaotic conduction with conduction block, and zones of delayed conduction. While some studies have described well-defined anatomical substrates without conduction blocks (15), others report heterogeneous activation and conduction delay (22).

The use of potential duration maps, introduced by Pappone et al., appears promising but has yet to be widely replicated (15, 17). Standardized and detailed substrate characterization is essential to improve procedural outcomes and guide ablation strategies in BrS.

This study was conducted at a single, highly experienced center and included only a small cohort of symptomatic patients, with limited and ongoing follow-up. These factors limit the generalizability of the findings to the broader BrS population. Anatomical variability, substrate heterogeneity, and differences in institutional expertise may significantly influence procedural outcomes.

The hybrid surgical approach is technically demanding, requiring specialized expertise and equipment that may not be widely available. A major limitation is the absence of comparative or randomized data directly comparing surgical and percutaneous epicardial ablation strategies. This lack of evidence makes it difficult to accurately assess the relative benefits, risks, and long-term outcomes of the surgical approach.

These are preliminary results and should not be considered definitive proof of efficacy; caution is warranted to avoid off-label application of the technique. Finally, the follow-up period remains short and ongoing.