46 * Automated ablation lesion tagging with an integrated radiofrequency parameter index predicts sites of acute reconnection following pulmonary vein isolation

Abstract

Introduction: Delivering durable ablation lesions when performing pulmonary vein isolation (PVI) is now recognised as being of vital importance, with most recurrences in paroxysmal AF patients thought to be related to formation of gaps in PVI lines. Gaps have been shown to be related to a lower minimum force-time integral (contact force multiplied by ablation time) within a segment, but it is highly likely that power delivery also plays a significant role. The latest module for the CARTO 3 system utilises automated ablation lesion tagging (Visitag[TM]), with calculation of an integrated radiofrequency parameter index (IRFPI) for each lesion incorporating contact force, power and time. We hypothesised that lower minimum IRFPI (IRFPImin) within a PVI segment would predict acute PV reconnection. Methods: Patients with paroxysmal or persistent AF undergoing PVI were included. All cases were performed in a standard fashion, using image integration where available and with delivery of ablation in a wide area circumferential ablation (WACA) pattern. Standardised Visitag[TM] settings were used for lesion tagging (minimum time 10s, maximum range 2mm). All PVs were assessed for evidence of spontaneous acute PV reconnection using a circular catheter after a minimum waiting time of 20 minutes. If no spontaneous reconnection was apparent, 12-18mg of adenosine was given to unmask any dormant conduction. In either instance, the reconnected segment was documented. Ablation maps were reviewed offline to identify the IRFPImin for each WACA segment. Results: 52 patients (median age 61 [IQR 53-69] years, 69% male, 73% paroxysmal AF, LA diameter 41 [38.5-43] mm) were studied. Acute reconnection was identified in 23 of 52 (44%) patients and in 37 of 463 (8%) WACA segments. The median IRFPImin for acutely reconnected segments was significantly lower than that for segments without reconnection (274 [256-341] vs. 316.5 [286-359], P =0.003, Fig. 1). After an initial by-segment sub-analysis, posterior and inferior segments were found to have similar values and were grouped, as were anterior and roof segments. The IRPFImin for reconnected posterior/inferior segments was again significantly lower than for segments without reconnection (270 [255-286] vs. 301 [271-340], P =0.013) and the same was found for anterior/roof segments (295.5 [261-366.5] vs. 332 [295-381], P =0.023). The IRFPImin for non-reconnected anterior/roof segments (332 [295-381]) was significantly higher than that for non-reconnected posterior/inferior segments (301 [271-340], P <0.0001). Conclusion: This is the first report of the novel IRFPI. The prevalence of acute segment reconnection (spontaneous or adenosine-mediated) after PVI performed with automated lesion tagging is low. The IRFPImin in a WACA segment predicts acute reconnection within that segment. Higher IRFPImin values are required for anterior and roof segments than for posterior and inferior segments to prevent acute reconnection. ![Graphic][1] [1]: /embed/inline-graphic-1.gif