Nanosecond pulsed field ablation for paroxysmal atrial fibrillation: feasibility, safety, and skeletal muscle activation

Pulsed field ablation (PFA) treats atrial fibrillation through irreversible electroporation rather than thermal injury. 1 Nanosecond pulsed field ablation (nsPFA), characterized by pulse durations of <1 µs, has shown favourable lesion durability and 1-year outcomes in paroxysmal atrial fibrillation (AF). 2 , 3 Experimental studies suggest that nanosecond pulses produce distinct electrophysiological and cellular effects while maintaining electroporation efficacy. 4 , 5 These mechanisms may contribute to the limited skeletal muscle activation observed during nsPFA delivery. However, clinical evidence remains limited. We therefore evaluated the feasibility, early safety, and skeletal muscle activation of a novel nsPFA system under local anaesthesia without sedation. This prospective, multicentre, single-arm pilot study enrolled patients aged 18–75 years with symptomatic, drug-refractory, or drug-intolerant paroxysmal AF documented for ≥30 s. Key exclusions were prior AF ablation, left atrial thrombus, left atrial diameter ≥50 mm, left ventricular ejection fraction ≤40%, and recent cerebrovascular events. The SYE-PFA-1A system comprised a pulse generator and shape-configurable circular catheter ( *Figure 1A* ). The study received institutional review board approval and was registered at ChiCTR (ChiCTR2400090647). Each pulmonary vein received two application pairs in the hat configuration with a 180° rotation and three in the flower configuration with 60° rotations. Pulmonary vein isolation was confirmed by electrogram recordings and electroanatomic voltage mapping. The rhythm endpoint was freedom from documented AF, atrial flutter, or atrial tachycardia lasting ≥30 s between the end of the 3-month blanking period and the 6-month follow-up. Rhythm was assessed by 12-lead electrocardiography and 24-h Holter monitoring at 1, 3, and 6 months, with additional symptom-triggered electrocardiography as needed. Antiarrhythmic drugs were allowed during the blanking period and discontinued thereafter unless clinically indicated. Safety events were adjudicated by an independent adverse-event committee. Pulmonary vein stenosis, atrio-oesophageal fistula, and neurological or vascular complications were assessed clinically throughout follow-up. Skeletal muscle activation was assessed under local anaesthesia to minimize anaesthetic effects on neuromuscular responses. A triaxial accelerometer was attached to the anterior abdominal wall approximately 1 cm below the xiphoid process and continuously recorded X-, Y-, and Z-axis acceleration. 6 The pre-specified primary Z-axis reflected movement perpendicular to the abdominal surface. Z-axis acceleration values during nsPFA delivery were compared with baseline recordings without energy delivery. Accelerometer signals were recorded at 30 Hz using an IM948 six-axis inertial measurement unit with digital low-pass filtering and transmitted via Bluetooth for analysis. A median of 4320 pulses per patient (range, 4320–5670) was delivered. Accelerometer data were analysed by investigators blinded to patient identity and procedural information. Each 2.5-s application comprised five cycles of 18 biphasic pulses (approximately 300 ns/phase) delivered at up to 4 kV. Local anaesthesia was achieved by local infiltration of 2% lidocaine hydrochloride (20 mg/mL; 0.4 g/20 mL) at the vascular access sites. Procedural pain and discomfort were assessed using a 10-point visual analogue scale immediately after the procedure. A total of 154 patients were enrolled (mean age 59.3 ± 10.3 years; 86 men [55.8%]). Sixty patients underwent procedures under local anaesthesia and were included in the skeletal muscle activation analysis. Compared with the general-anaesthesia group, the local-anaesthesia group was older and had a higher CHA 2 DS 2 -VASc score and a higher prevalence of diabetes mellitus. The local-anaesthesia group also had fewer ablation applications and shorter procedure and fluoroscopy times. Acute pulmonary vein isolation was achieved in all 618 targeted pulmonary veins (100%). The mean left atrium diameter was 36.2 ± 4.1 mm ( *Figure 1C* ). During 6 months of follow-up, sinus rhythm was documented in 145 patients (94%) based on scheduled electrocardiography and 24-h Holter monitoring at 1, 3, and 6 months. Non-pulmonary vein ablation included superior vena cava (SVC) isolation in 22 patients and roof ablation in one. At one redo, all pulmonary veins remained isolated; SVC isolation and Vein of Marshall ethanol infusion were performed. No device- or procedure-related death, atrio-oesophageal fistula, clinically apparent pulmonary vein stenosis, stroke/transient ischaemic attack, persistent phrenic nerve injury, coronary spasm, or haemolysis-related renal failure was observed. In the local-anaesthesia subgroup, nsPFA delivery was not associated with a measurable increase in abdominal-wall acceleration compared with baseline recordings without energy delivery (overall *P* = 0.70), with consistent findings observed for the left superior ( *P* = 0.45), left inferior ( *P* = 0.58), right superior ( *P* = 0.71), and right inferior pulmonary veins ( *P* = 0.69) ( *Figure 1B* ). Among 60 patients undergoing procedures without general anaesthesia, visual analogue scale (VAS) pain scores were available in 59 patients (98.3%). The median VAS pain score was 3.0 (interquartile range, 2.5–6.0; range, 0–10). This multicentre pilot study achieved acute pulmonary vein isolation with minimal skeletal muscle activation under local anaesthesia. The greater use of general anaesthesia than in SCENA-AF may reflect differences in device and catheter design. 7 Despite centre-specific preference for general anaesthesia during early adoption, 60 patients (39%) underwent nsPFA under local anaesthesia without sedation. The single-arm design precludes comparative conclusions regarding anaesthesia requirements, efficacy, or tolerability. Recent data indicate that nanosecond pulse duration does not eliminate PFA-related haemolysis. 8 The minimal muscle activation observed was specific to this nsPFA system and should not be generalized to other waveforms or platforms. Compared with thermal ablation, PFA may better preserve autonomic function through mid- to long-term follow-up; whether this benefit extends to nsPFA warrants investigation. 9 Our findings support feasibility and early safety but require confirmation in larger controlled studies. Several limitations should be acknowledged. Temperature was not measured, local-anaesthesia use was centre-specific, and intermittent monitoring with 6-month follow-up may have missed recurrences or late complications. Abdominal accelerometry may have underestimated diaphragmatic or lower-limb movements. Dedicated haemolysis markers were not measured; therefore, subclinical haemolysis cannot be excluded. Without systematic remapping, lesion durability and dose adequacy remain uncertain, and low-dose delivery may have contributed to the limited muscle activation observed. Confirmation with remapping-based durability data is therefore needed. This multicentre pilot study supports the feasibility and early safety of nsPFA, with minimal skeletal muscle activation under local anaesthesia without sedation. Larger controlled studies with longer follow-up are needed.