Introduction: Atrial fibrillation is one of the most prevalent cardiac arrhythmias worldwide. Pulsed-Field Ablation (PFA) is a non-thermal catheter-based ablation strategy using high-voltage electric pulses to induce irreversible electroporation and cell death, to restore sinus rhythm. Although some pre-clinical studies used electrocardiogram-timed ablation protocols, the role of the cardiac cycle on PFA efficiency is unclear. This study assessed whether the mechanical state of the heart may affect the geometry of lesions induced by PFA.
Methods: Langendorff-perfused wild-type mouse hearts were either arrested in their relaxed state using a mechanical uncoupler (2,3-butanedione 2-monoxyme, 30 mmol/L), or put into contracture using lithium and caffein (10 mmol/L). Then, 250 V biphasic electric voltages (8 bursts of 60.42 ms each) were delivered using a PFA generator system (AQUILA, Stockert GmbH) and a 1.25 mm long and 0.25 mm wide monopolar linear-tip catheter (Stockert GmbH). The ablation protocol was designed to obtain lesions that are on purpose not fully transmural, to not be limited in the assessment of lesion depth. Every heart received two epicardial ablations per ventricle, followed by a 30 min recovery period perfused with physiological solution to allow the heart to recover its contractile activity and the electroporated cells to die or recover. The heart was stained by perfusion of 2% TTC (2,3,5-triphenyltetrazolium chloride) during 15 minutes and tissue blocs including the lesions were cut. Then, tissue was cryo-embedded and sectioned at 30 µm thickness, perpendicular to the epicardial surface and parallel to the lesion’s long axis, before imaging one section every 90 µm. The short and long axes, depth and lesion area were assessed for each slice.
Results: PFA resulted in formation of ellipsoidal lesions that were clearly identifiable after TTC staining on mouse ventricles. In both ventricles, the maximal lesion area of transmural tissue sections was significantly larger when PFA was delivered during contracture, compared to the relaxed state (1.3 ± 0.4 mm2 vs 0.6 ± 0.1 mm2 for the right ventricle; n = 2 animals / 4 lesions for each condition) and 1.4 ± 0.2 mm2 vs 0.9 ± 0.1 mm2 for the left ventricle (n = 3/6 for contracture and n = 4/8 for relaxed state; p < 0.001 for both). In left ventricle, this was linked to significantly larger long and short axes, whereas no significant differences were measured in maximal lesion depth. In both ventricles, the lesion volume was significantly greater when PFA was delivered to hearts in contracture compared to relaxed hearts, by 116.6 ± 39.8 % in the right ventricle (n = 2/4 for both condition; p < 0.01) and by 114.9 ± 18.5 % in the left ventricle (n = 3/6 for contracture and n = 4/8 for relaxed state; p < 0.001).
Conclusion: PFA generates larger lesions when delivered during contracture, compared to relaxed hearts. This observation is of potential interest for optimising clinical protocols, and further studies will assess whether it can be confirmed in beating hearts. For this, we are developing a protocol enabling heartbeat-synchronised ablation, targeting either systole or diastole in Langendorff-perfused pig hearts. In this setup, the intra-ventricular pressure signal is used to track the mechanical cycle and to trigger ablations.
