The effect of pulsed field versus radiofrequency pulmonary vein isolation on the cardiac autonomic nervous system in the BEAT PAROX-AF randomized trial
D Wichterle, J Kautzner, P Stojadinovic, J Petru, P Neuzil, S Boveda, I Deisenhofer, E Frison, S Knecht, P Maury, V Renaudeau, M Rousset, E Sauer, D Scherr, P JaisAbstract
Background
Significant alteration of the cardiac autonomic nervous system (ANS) was described as a collateral effect of pulmonary vein (PV) isolation (PVI). The impact of ganglionated plexi ablation is stronger with thermal than with non-thermal ablation energy. The prespecified ANS substudy investigated these differences in detail, both acutely and in the long term.
Methods
BEAT-PAROX-AF was a multicentre, open-label, randomized, controlled trial conducted across nine European sites [1]. Patients with symptomatic, drug-resistant paroxysmal AF were randomized (1:1) to either pentaspline catheter pulsed field ablation (PFA) with the FARAPULSE system or point-by-point radiofrequency ablation (RFA) with the Thermocool SmartTouch catheter and 3D navigation by CARTO 3, following the CLOSE protocol. Patients eligible for the ANS substudy, conducted at two study sites, were sub-randomized to the order of PVI, either ipsilateral left or right PVs ablated first. Extracardiac vagal nerve stimulation (ECVS) was performed at baseline, after unilateral PVI, and at the end of the procedure. Atropine test was performed a day before and after the ablation, and then at 6- and 12-month visits.
Results
Of 292 patients randomized, 110 patients (aged 61±10 years, 51% males, left atrial diameter of 41±6 mm, and left ventricular ejection fraction of 61±5%) were included in the ANS substudy. They were treated with PFA (n=54), with a total of 41±21 applications, or RFA (n=56), with an RF time of 24±6 min. The duration of ECVS-induced sinus arrest was reduced from 8.5±4.5 s to 6.1±3.9 s (P<0.00001) after PFA and from 8.3±4.2 s to 1.8±2.1 s (P<0.00001) after RFA; more after RFA [RFA-PFA difference]: -4.0 s (95% CI, -5.5 to -2.6 s). Similarly, the duration of ECVS-induced AV block was reduced from 6.2±3.3 s to 4.7±3.3 s (P<0.0001) after PFA and from 5.9±2.6 s to 1.1±3.3 s (P<0.00001) after RFA; more after RFA: -3.4 s (95% CI, -4.4 to -2.4 s). Figure 1 shows the dependence of ECVS-induced sinus arrests on the order of PVI, with right PVI having a generally stronger effect. The response of the AV node was comparable to that of the sinus node. The acceleration of SR after atropine on the first day after the procedure was reduced from +49±18% to +43±16% (P<0.05) after PFA and from +48±19% to +24±15% (P<0.00001) after RFA; more after RFA: -20% (95% CI, -28 to -11%). Figure 2 shows the profile of post-atropine SR acceleration during follow-up, with acute suppression and gradual recovery that was faster after PFA.
Conclusions
The alteration of cardiac ANS induced by PFA was significantly weaker and less durable compared to that after RFA. Ablation of the right PVs accounted for the majority of the overall effect. The maximum attenuation of vagal modulation of both SA and AV nodes was observed immediately after the procedure, with almost complete recovery in the PFA arm and partial recovery in the RFA arm during the 1-year follow-up period.Figure 1Figure 2