Myocardial and Neural Damage due to Pulmonary Vein Isolation: Comparison of Pulsed Electric Field and Thermal Ablation

Background Pulmonary vein isolation (PVI) using thermal energy-radiofrequency (RF) or cryoablation (CRYO)-is associated with direct myocardial injury and collateral neural damage. In contrast, pulsed electric field (PEF) energy has emerged as a neural-sparing alternative. There is a lack of quantitative data on proportional neural damage resulting from individual ablation energies.Objective To assess the impact of PVI by RF, CRYO and PEF on the release of plasma biomarkers of myocardial and neural damage.Methods This study included 85 patients with paroxysmal atrial fibrillation undergoing PVI (age: 61 +/- 12 years, 74% males). Ablation was performed with PEF in 32 patients, RF in 23, and CRYO in 30. Peripheral venous blood samples were collected before, immediately after PVI, and the day after the procedure to measure plasma levels of S100 binding protein (S100B), high-sensitivity troponin I (hsTnI), and high-sensitivity troponin T (hsTnT). Groups were compared based on maximum biomarker levels and the (hsTnI/hsTnT)/S100B ratio as an index of cardioselectivity.Results Following PVI, all energy modalities caused significant myocardial injury, with PEF showing the most pronounced effects (maximum hsTnT: 1476, 958 and 967 ng/L; maximum hsTnI: 12659, 1699 and 8109 ng/L; for PEF, RF and CRYO, respectively). Maximum S100B levels indicating the neural impact were highest in the CRYO group (80, 80 and 190 ng/L for PEF, RF and CRYO, respectively). PEF exhibited the greatest cardioselectivity, as indicated by the highest (hsTnI/hsTnT)/S100B ratio (0.11 +/- 0.04, 0.02 +/- 0.01 and 0.05 +/- 0.05 for PEF, RF and CRYO, respectively). Interestingly, no correlation was observed between S100B release and heart rate acceleration (R = 0.07, p = 0.51).Conclusion PEF is the most cardioselective modality of ablation energy currently available for the treatment of atrial fibrillation. The neuromodulatory effect of PVI on the sinus node appears to be independent of the magnitude of neural damage.