Safety profile of multielectrode-phased radiofrequency pulmonary vein ablation catheter and irrigated radiofrequency catheter

Silent cerebral lesions with the multielectrode-phased radiofrequency (RF) pulmonary vein ablation catheter (PVAC^®) have recently been investigated. However, comparative data on safety in relation to irrigated RF ablation are missing.

One hundred and fifty consecutive patients (58 ± 12 years, 56 female) underwent first pulmonary vein isolation (PVI) for atrial fibrillation (61% paroxysmal) using PVAC^® (PVAC). Procedure data as well as in-hospital complications were compared with 300 matched patients who underwent PVI using irrigated RF (iRF). Procedure duration (148 ± 63 vs. 208 ± 70 min; P < 0.001), RF duration (24 ± 10 vs. 49 ± 25 min; P < 0.001), and fluoroscopy time (21 ± 10 vs. 35 ± 13 min; P < 0.001) were significantly shorter using PVAC. Major complication rates [major bleeding, transitoric ischaemic attack (TIA), and pericardial tamponade] were not significantly different between groups (PVAC, n = 3; 2% vs. iRF n = 17; 6%). Overall complication rate, including minor events, was similar in both groups [n = 21 (14%) vs. n = 48 (16%)]. Most of these were bleeding complications due to vascular access [n = 8 (5.3%) vs. n = 22 (7.3%)], which required surgical intervention in five patients [n = 1 (0.7%) vs. n = 4 (1.3%)]. Pericardial effusion [n = 4 (2.7%) vs. n = 19 (6.3%); pericardial tamponade requiring drainage n = 0 vs. n = 6] occurred more frequently using iRF. Two patients in each group developed a TIA (1.3% vs. 0.6%). Of note, four of five thromboembolic events in the PVAC group (two TIAs and three transient ST elevations during ablation) occurred when all 10 electrodes were used for ablation.

Pulmonary vein isolation using PVAC as a ‘one-shot-system’ has a comparable complication rate but a different risk profile. Pericardial effusion and tamponade occurred more frequently using iRF, whereas thromboembolic events were more prevalent using PVAC. Occurrence of clinically relevant thromboembolic events might be reduced by avoidance of electrode 1and 10 interaction and uninterrupted anticoagulation, whereas contact force sensing for iRF might minimize pericardial effusion.