Background: Different technologies and ablation strategies have been developed to enhance the durability of pulmonary vein isolation (PVI). Recently, radiofrequency (RF) ablation with high power short duration (HPSD) technique has been introduced, guided by the ablation index (AI). Furthermore, the new generation of the visually guided laser balloon (VGLB; X3 Heartlight) allows to perform a continuous ablation line with an implemented rotating motor (rapid mode). Previous reports showed that the detection of dormant pulmonary vein (PV) conduction using adenosine provocation test (APT) may reveal the differences in lesion efficacy between different ablation techniques. Here we compared these two new ablation strategies for acute efficacy with APT.

Methods: 70 (age 62±11 years; 57% male) patients with atrial fibrillation (AF) were enrolled into HPSD or VGLB arm in a 1:1 randomized manner. In the HPSD arm, circumferential PVI was performed for ipsilateral PVs with irrigated RF ablation (power controlled mode with 50 W, 15 ml/min irrigation rate, target contact force 10-20 g, interlesion distance 4-6 mm, duration of ablation per lesion determined with AI targeting 400 at the posterior and 500 at the anterior wall) and 3D-mapping system (Carto 3). In the VGLB arm, each PV was isolated separately. Further, we strived for a continuous ablation line around the PV ostia using the automated rapid mode with 13 W ablation power. In regions with poor balloon-tissue contact we changed to manually performed point-by-point ablation with reduced energy settings (5.5-10 W for 20-30 s). Intraluminal temperature of the esophagus was monitored with a temperature probe only in the VGLB arm. 20 minutes after each successful PV isolation, we performed an APT and looked for PV reconnection.

Results: 137 (100%) PVs in the HPSD arm and 131 (98.5%) in the VGLB arm were successfully isolated (p=0.24). The overall procedure time was similar in both groups (155±39 in HPSD vs. 175±58 min in VGLB, p=0.191), however, fluoroscopy time, LA dwelling time and duration from the first to the last ablation were longer in the VGLB arm (23±8 vs. 12±4 min, p<0.001; 157±53 vs. 130±33 min, p=0.013; 92±43 vs. 72±40 min, p=0.01, for VGLB and HPSD, respectively). The usage of automated rapid mode correlated with a decrease of procedure duration (p<0.001, R²=0.52) and 43 PV (32%) in the VGLB arm were isolated with the first ablation circle using rapid mode only. In the HPSD arm, 60% (42 of 70) of the ipsilateral PVs and 74,4% (99 of 133) of the individual PVs in the VGLB arm were isolated by first encirclement.
In the HPSD arm, 10 PVs in 8 patients and in the VGLB arm 6 PVs in 5 patients had a reconnection under APT (HPSD 7% vs. 5% VGLB; p=0.34).
One patient in the VGLB arm suffered from esophageal erosion without a fistula formation, which was clipped and healed up without further complications. Another patient suffered from reversible phrenic nerve palsy. In two cases, transient ST-elevation occurred after insertion of the VGLB, probably due to air embolism. In the HPSD arm, one patient suffered from postinterventional pericarditis.

Conclusion: Acute lesion efficacy of PVI with RF ablation using HPSD technique and the new Heartlight X3 VGLB is very high and do not differ from each other. Studies with greater patient population are needed to examine if the high acute efficacy of these techniques could influence the clinical outcome of PVI.