In this issue of The Journal of Cardiovascular Electrophysiology , Hsu et al. evaluated the safety and efficacy of delivering PFA left atrial lesions in an in- vivo porcine model [22]. They delivered PFA with a novel 10-electrode irrigated circular catheter and pulse generator that is integrated with the CARTO mapping system (Biosense Webster, Irvine, California). PFA was delivered in 8 swine and included PFA delivered at (1) a PV ostium, (2) directly inside a PV to assess for stenosis, (3) over the phrenic nerve in the right atrium to assess for phrenic nerve injury, (4) lateral mitral valve annulus to assess for valvular damage, and (5) in the aorta adjacent to the esophagus to assess for esophageal injury. The swine were survived for 30 days and were remapped prior to being euthanized for histologic examination.

The pulse field waveform was delivered in a bipolar configuration alternating between different catheter electrodes in trains of biphasic pulses with a total application duration of approximately 250 msec. The authors report that the pulse generator was set to 1800V. However, it is not clear if this represents the amplitude for each electrode pair or if this is divided between multiple electrodes simultaneously. The pulse width, duty cycle, and the voltage amplitude of the waveform between each bipolar electrode pair are not described.

We congratulate the authors on this study, which adds to the growing body of evidence supporting the safety and efficacy of PFA. Hsu et al. showed that PFA did not cause collateral tissue damage acutely or at 30 days despite delivery of numerous PFA lesions to vulnerable areas. There was no pulmonary vein stenosis, phrenic nerve injury, mitral valvular damage, esophageal injury, or reduction in left ventricular function.

PFA lesions delivered at PV ostia demonstrated excellent efficacy with PV isolation acutely and durable isolation at 30 days in all swine and histology showing circumferential, transmural necrosis and repair.

Regarding safety, one of the potential issues with electroporation is the phenomenon of arcing, where energy above a given threshold results in rapid gas accumulation that results in a shock wave that can cause significant barotrauma [23]. The arcing threshold may be different for a given waveform and catheter design and should be carefully evaluated for each PFA system. In the current study by Hsu et al, there was no occurrence of thrombus and/or charring on the catheter tip, pericardial effusion and/or cardiac tamponade, steam pop events, or mural thrombus (on intracardiac echocardiography [ICE] during the procedure nor during gross pathology), no incidence of clinically significant mechanical tissue injury was noted from gross pathology, and no incidence of clinically significant thrombo-emboli was found in upstream and downstream organs nor within the heart. Silent cerebral infarctions (SCI) have been described in up to 67% of patients undergoing CA for AF and are defined as the presence of asymptomatic cerebral lesions detected with imaging (i.e., magnetic resonance imaging) studies (24). As PFA is associated with micro-bubble formation (probably due to electrolysis), there is a concern of SCI associated with these micro-bubbles. Nonetheless, a canine model failed to demonstrate SCI after PFA was administered in the ascending aorta (25). Moreover, Reddy et al failed to find SCI in the 13/81 patients who underwent cerebral MRI after PFA in the IMPULSE/PEFCAT trial (26). However,

using a lattice catheter, postprocedural brain MRI revealed diffusion-weighted imaging+/fluid-attenuated inversion recovery- and diffusion-weighted imaging+/fluid-attenuated inversion recovery+ asymptomatic lesions in observed in 9.8% and 5.9% of patients, respectively (27). Further studies and trials are needed to clarify this important dilemma.

The authors delivered supratherapeutic PFA lesions directly adjacent to the phrenic nerve from the endocardium. No acute injury was seen, nor was any injury found during gross pathology. This data is encouraging and may allow ablation of cardiac tissue that previously would have been prohibitively risky with RF, such as treating atrial tachycardias originating from the Crista terminalis, which frequently are near the right phrenic nerve. Currently, complex and risky procedures including epicardial access with deviation/protection of the phrenic nerve using air, saline, deflectable sheaths and balloons are the only option for these patients.

There were no incidents of PV narrowing on the same day or 30 days post PFA application when

the vein was targeted for isolation via ablation delivered to the ostium, nor when ablation was performed directly inside of the vein, as demonstrated by X-ray, ICE, and by flow velocity data. This corroborates the work by Howard et al, hypothesized that pulsed field ablation (PFA) would reduce PV stenosis risk and collateral injury compared with irrigated radiofrequency ablation (IRF). A more precise method of evaluating PV stenosis was developed that includes the use of 3-dimensional modeling based on computed tomography angiography with triplicate measures of cross-sectional area at the distal PV and ostial ablation sites. PV measurements were made pre-ablation and at 2-, 4-, 8-, and 12-weeks post-ablation to provide a detailed time course of the progression of PV stenosis from radiofrequency ablations while showing negligible changes in dimensions in pulsed field ablation treated sites [28].

There was no evidence of esophageal injury despite delivery of supratherapeutic PFA lesions adjacent to the esophagus from the aorta. The esophagus in all animals showed no injury to the mucosa, muscularis or serosa on histology after surviving the animal for 30 days. This is consistent with prior in vivo studies showing no esophageal injury despite administering high doses of PFA adjacent to the esophagus and in human trials showing no evidence of esophageal lesions or enhancement on EGD or chest CMR (26, 29).