Redo ablation in patients with recurrent persistent atrial fibrillation using a dedicated software ultra-high density mapping tool. A comparative study with a matched historical cohort

Background In redo ablations for persistent atrial fibrillation (AF) most groups perform left atrial ablation in addition to pulmonary vein (PV) re-isolation but there is no clear consensus as to how to select ablation targets, and overall outcomes remain suboptimal. A new dedicated software tool has been developed to analyze intracardiac electrograms (EGMs) during AF and detect potential AF drivers to be targetted. Purpose We have applied this tool in a consecutive series of patients and compared the results with those from a matched historical patients’ cohort. Methods Group 1: consecutive patients undergoing a redo ablation procedure for persistent AF using a dedicated software to localize potential ablation sites. An ultra–high-density electroanatomical map was created by successively acquiring EGMs from all atrial regions. Ablation targets were identified based on local cycle length, local spanning of the electrograms, and pattern spatiotemporal consistency (Figure 1), and subsequently underwent radiofrequency (RF) ablation. Group 2: matched cases selected from our historical cohort of redo ablations for persistent AF. In this group, conventional EGM-guided techniques to identify targets for RF ablation throughout the years included: complex fractionated EGMs, low voltage areas, non-PV triggers, areas with fast and regular activity. For each group 1 case a group 2 patient was randomly selected among historical controls that matched for: 1) similar age (±10 years), 2) same sex, 3) use of contact force-sensing catheters. To ensure comparable follow-up durations (since historical controls had longer follow-up), group 2 patients were censored at the same follow-up duration as their matched group 1 case. In both groups, the reconnection status of each PV was assessed, and any reconnected PVs were re-isolated. Results Group 1: 26 cases with at least 6 months follow-up. Group 2: 26 matched cases (from 55 eligible historical patients). Groups 1 and 2 were comparable (P>0.05) in relation to: age (63 vs 64), sex, hypertension (38% vs 50%) diabetes (4% vs 4%), cardiomyopathy (8% vs 23%), antiarrhythmic drugs (46% vs 39%), AF duration (6 vs 6 months), left atrial size (32 cm2 vs 33 cm2), number of previous ablations (1.7 vs 1.5). The mean follow-up was 11.8 ± 1.4 months. As shown in figure 2, arrhythmic recurrences (AF or flutter) were observed in 4 cases of group 1 (15.4%) vs 12 (46.1%) of group 2, P=0.009. AF occurred in 3 cases (11.5%) in group 1 vs 9 (34.6%) in group 2, p= 0.02, and flutter in 1 (3.8%) vs 3 (11.5%), p=0.15. Conclusions In patients undergoing redo RF ablation for recurrent persistent AF, the use of dedicated software to analyze ultra–high-density EGM maps and localize potential targets was associated with a significantly lower rate of AF and AF+flutter recurrence compared with conventional ablation approaches.