Association between Troponin release, ablation extent and recurrence of atrial fibrillation or atrial flutter after pulmonary vein isolation using pulsed field ablation

Introduction

The extent of myocardial injury during pulmonary vein isolation (PVI) with pulsed field ablation (PFA) may influence long-term rhythm outcomes. However, the relationship between biomarker release, ablation geometry, and atrial fibrillation (AF) or flutter (AFlu) recurrence remains unclear.

Purpose

In a previous single-center study, we observed an inverse association between post-procedural troponin levels and AF/AFlu recurrence. This extended analysis aimed to validate these findings in a larger cohort with longer follow-up and to investigate whether high-sensitivity troponin T (hs-cTnT) kinetics relate to ablation map characteristics.

Methods

We retrospectively analyzed 138 patients undergoing first-time PVI for paroxysmal or persistent AF between February 2024 and March 2025, predominantly (134/138) with PFA. In 19 patients, left atrial (LA) high-density 3D maps were acquired before and after ablation to quantify ablated atrial area or volume. Hs-cTnT was measured before ablation and on post-procedural days 1 and 2; the difference between baseline and peak values was used for analysis. Clinical follow-up assessed arrhythmia recurrence.

Results

Mean age was 69 ± 11 years; 52 patients (38%) were female and 64% had paroxysmal AF. Mean CHA2DS2-VA score was 2.9 ± 1.6. Skin-to-skin time was 59 ± 32 min without and 88 ± 43 min with ultra-high-density mapping. 75 patients were treated with Medtronic™ PulseSelect (PFA-P), 48 with Boston™ Farapulse (PFA-F), 11 with Sphere9™ in PFA mode (PFA-A), and 4 with radiofrequency ablation. All remapped patients received PFA-P. Mapping catheters included Orion™, HD Grid™, and Octaray™. The ablated area was expressed as a percentage of total LA surface (EnSite X™) or volume (RHYTHMIA™, CARTO™).

During a median follow-up of 167 ± 84 days, 32 patients (23%) had AF/AFlu recurrence. Patients without recurrence showed significantly higher hs-cTnT levels than those with recurrence (median 1343 vs 760 pg/ml, p = 0.0003). The calculated cutoff predicting recurrence was 1591 pg/ml. Median hs-cTnT release differed among systems (PFA-P 1423, PFA-F 1128, PFA-A 1865 pg/ml; p = 0.0027), but recurrence rates were comparable between modalities (p = 0.12). No significant correlation was observed between hs-cTnT levels and ablated area using pearson´s rank correlation (r = 0.10, p = 0.69) or between ablated area and AF/Aflu recurrence (p = 0.25).

Conclusion

PFA lesions during PVI are not homogeneously distributed, and 3D-mapping reflects lesion geometry rather than depth or transmurality. Higher hs-cTnT release after PFA-PVI indicates myocardial injury and is associated with improved rhythm outcomes, independent of mapped ablation area. Troponin may thus serve as a practical biomarker of lesion quality in PFA, whereas geometric lesion extent alone does not predict efficacy. Post-procedural troponin assessment could provide a simple measure of procedural effectiveness without additional mapping.