Local impedance tissue characterization to implement pulmonary veins isolation success in AF patients

FUNDING ACKNOWLEDGEMENTS: Type of funding sources: None. BACKGROUND: Novel radiofrequency (RF) ablation technology permits local impedance (LI) assessment, enabling increased knowledge of the underlying tissue’s properties. In addition, LI can provide details on catheter tip contact and extent/effectiveness of radiofrequency (RF) delivery. To date, the extent to which left atrial (LA) tissue characterization by LI can guide RF ablation has not yet been explored. PURPOSE: We aimed to assess LI capabilities of predicting RF ablation success in consecutive AF cases with different underlying cardiac tissue characteristics. METHODS: Two-hundred twelve consecutive patients undergoing de novo AF ablation at 16 European centers were included in the CHARISMA registry. A novel ablation catheter (Stablepoint) with dedicated algorithm (DirectSense) was used to measure LI at the distal electrode. Each ablation point was characterized in terms of RF delivery time, baseline LI and LI drop during ablation according to different atrial substrates: normal-voltage (NV, >0.5mV), intermediate-voltage (IV, 0.1-0.5mV) and low-voltage tissue (LV, <0.1mV). First pass isolation (FPI) was defined as successful PVI at or before completion of the first encircling lesion set regardless of visual gaps. Ablation endpoint was pulmonary vein isolation (PVI) as assessed by entrance and exit block. Data are reported as mean±SD. RESULTS: Atrial substrate was analyzed at 11405 (82%) sites with complete high-quality data. Ablation spots were more frequently deployed in NV areas (n=6714, 58.9%) than in areas of IV (n=4065, 35.6%) or LV (n=625, 5.5%). Both baseline LI (163.8±19Ω) and LI drop (23.0±9Ω) were higher in NV areas than in IV areas (baseline LI: 159.4±18Ω, p<0.0001; LI drop: 20.5±9Ω, p<0.0001) and LV areas (baseline LI: 157.8±18Ω, p<0.0001; LI drop: 19.5±9Ω, p<0.0001). RF delivery time was shorter in the NV (9±5s) than both in IV and LV areas (10±4s, p<0.0001 for IV; 10±4s, p<0.0001 for LV, respectively). The LI drop that best predicted successful ablation spots was >20Ω for NV areas (Sensitivity=62.5%, Specificity=91.5%, PPV=99.6%, Area under the receiver-operating characteristic curve (AUC)=0.7896, p<0.0001) and >18Ω for IV and LV areas (Sensitivity=57.7%, Specificity=94.6%, PPV=99.8%, AUC=0.8008, p<0.0001). No complication occurred. The acute procedural success was 100%, with all PVs successfully isolated in all study patients. CONCLUSION: LI drop during ablation significantly differs according to atrial substrates. These findings suggest that a tailored AF ablation strategy taking into account the underling LA substrate may be useful.