Biophysical background of cerebral microembolization during phased radiofrequency ablation of atrial fibrillation: analysis of high resolution data from the ablation generator

Abstract

Background: Asymptomatic cerebral emboli (ACE) after catheter ablation of atrial fibrillation (AF) have been recently demonstrated on diffusion MRI. The highest incidence of new lesion on postablation diffusion MRI have been consistently found after ablation with the circular, multipolar pulmonary vein ablation catheter (PVAC) and phased refiofrequency (RF) energy. Our group has also demonstrated, that the total number of microembolic signals (MES) detected by transcranial Doppler (TCD) in the middle cerebral arteries (MCA) is significantly higher during phased RF as compared to during cryoballoon ablation. Limited data are available regarding the mechanism of MES generation during ablation with the PVAC. Therefore, we investigated different biophysical parameters of RF delivery with the phased RF technique in order to find potential correlation with the MES counts detected by TCD during the ablation. Methods and results: 48 consecutive patients (female:17, mean age:60, SD:10 years) undergoing PVI with phased RF for non-valvular, paroxysmal or persistent atrial fibrillation formed the study population. Preprocedural anticoagulation consisted of uninterrupted vitamin K antagonist to maintain an INR in the therapeutic range (2.0-3.0). Unfractionated Heparin was used intravenously to maintain the ACT level above 350 seconds during the ablation. MES counts were recorded with bilateral TCD monitoring of MCAs for each session of energy delivery and time stamped for correlation with the high resolution data on power, temperature and impedance collected from the GENius 14.4 generator. Significant correlation was demonstrated between MES count during RF delivery and the following biophysical parameters: average power, total energy delivered, the number of electrodes turned on for simultaneous RF delivery, temperature overshoot (temperature integral over 62°C) and improper contact caused by respiration (P<0.0001). Furthermore, ablations of the left sided PVs also resulted in higher MES counts as compared to RF applications on the right side (P=0.0003). The bipolar/unipolar ratio of energy delivery and the average temperature reached during an ablation session had no effect on MES generation. Conclusions: Several biophysical parameters of energy delivery demonstrated a significant correlation with a higher rate of microemboli production during PVI with the PVAC. Most of these parameters can be linked to a variable or poor electrode-tissue contact. Modifications in the catheter design and in the software regulation of power/temperature handling of the generator might improve the safety of this technology.