Electrophysiological study and catheter ablation
The electrophysiological study (EPS) and ablation procedures were performed as previously described.11 Catheter ablation was performed under conscious sedation using dexmedetomidine, thiopental, and fentanyl. A multipolar catheter with internal cardioversion (BeeAT, Japan-Life-Line, Tokyo, Japan) was placed in the coronary sinus through the internal jugular vein, and a deflectable multipolar catheter was placed in the right ventricle or RA through the right femoral vein. Transseptal puncture was performed with a radiofrequency needle (Baylis Medical, Montreal, QC, Canada) under intracardiac echocardiography guidance, and two sheaths were introduced into the LA. A deflectable sheath (Agilis, Abbott, St. Paul, MN, USA, or Vizigo, Biosense Webster Inc., Irvine, CA, USA) and a long sheath (Swartz SL0 introducer, Abbott) were inserted into the LA. If AF persisted, defibrillation was performed. During RA pacing (from the high or lateral RA), a high-resolution LA activation map was created using a mapping catheter (Pentaray, Biosense Webster). Points were acquired using the CARTO ConfiDENSE module (Biosense Webster). The preprocedural computed tomography image of the LA was integrated with a 3D electroanatomical map. After LA activation mapping was created, bilateral extensive PVI was performed using a contact force-sensing radiofrequency catheter (TermoCool SmartTouch SF, Biosense Webster) with a high-power short-duration (50 w) protocol. The contact force was controlled at 10–20 g, and the application time was titrated using the lesion formation index (Ablation Index (AI), Biosense Webster). The ablation lesions were tagged in a contiguous fashion within a 6-mm interval (VISITAG, 6-mm diameter setting). The AI was targeted to 500–550 except for 400 along esophageal lesions on the posterior wall with the limitation of maintaining an esophageal temperature of <41 oC. The PVI lines were designed by operators without considering the ECs site. To confirm the completion of PVI, the bidirectional block of paced or self-activated PV electrograms was assessed using a 10 or 20 multipolar circular catheter after ablation of all PVs. If PVI could not be achieved using the originally designed isolation line, the earliest activation sites inside the isolation line were ablated. If acute PV reconnections were observed during the waiting period, the earliest activation sites inside the isolation line were ablated.