Background: There is an increasing interest in endoscopic laser balloon systems (ELBS) to achieve pulmonary vein isolation (PVI). The third generation ELBS (X3, Heartlight) presents novel structural changes compared to its predecessor. These include a novel feature (RAPID), which potentially allows a full 360 degrees single lesion for PVI. Nevertheless, there is a lack of data on the X3 system.

Objective: To analyse the safety, procedural times, and acute success rate of PVI in a cohort of patients who underwent ablation of atrial fibrillation using the third-generation ELBS.

Methods: Between June 2019 and November 2022, we prospectively enrolled all consecutive patients who underwent PVI using the X3 system at two Centres. Four operators performed all the procedures. For each pulmonary vein (PV), the operator always attempted to obtain a complete circumferential lesion using only the RAPID feature. If this was not feasible, the operator treated the remaining visual gaps with point-by-point energy delivery. At the end of PVI, an acute check for a bidirectional block was performed at the operator’s discretion.

Results: We enrolled 163 consecutive patients (mean age 61.9 ± 11.4, female patients 30.6%, persistent atrial fibrillation 30.6%). 655 pulmonary veins were identified and successfully visually isolated. 41 (25.1%) patients presented an anatomical variation of the PVs, of whom 23 had a supernumerary right PV, and 19 had either a left or a right common trunk. Of 655 PVs, 424 (64%) were isolated using the RAPID feature for at least 80% of their circumference, 389 (59.4%) using it for at least 90% of their circumference, and finally, 289 (44%) were isolated using only RAPID. In 35 (21.5%) patients, the operators could perform the entire procedure with uninterrupted use of the RAPID feature. After PVI, we performed an acute check for a bidirectional block in 73 patients (287 PVs), and 5 (6 PVs) still presented a gap in the ablation lines. Mean total procedural and total fluoroscopic times were 95.7 ± 44.5 min and 13.9 ± 8.4, respectively. We observed 5 (3%) periprocedural complications. Three episodes of transient phrenic nerve palsy that promptly resolved during follow-up (longest episode two months after discharge). One episode of pericardial tamponade, with subsequent need for pericardiocentesis. One periprocedural transient ischaemic attack (TIA). During energy delivery, eight pinhole balloon ruptures occurred.

Conclusion: To our knowledge, this is the largest cohort of patients treated with third-generation ELBS. In our two-centre experience, the X3 system proved effective and safe, and the RAPID feature showed good applicability.