Prospective multicenter feasibility study of omnipolar near-field peak-frequency mapping for slow-pathway localization in typical atrioventricular nodal re-entrant tachycardia

Abstract Background Slow-pathway (SP) ablation is the established therapy for atrioventricular nodal re-entrant tachycardia (AVNRT). While conventional anatomical and electrogram-guided approaches are effective, SP localization may be challenging in a subset of procedures. Omnipolar near-field (OTNF) peak frequency (PF) mapping has been proposed as a signal-based approach to support SP localization; however, prospective multicenter data remain limited. Methods Fifty consecutive adults with inducible typical AVNRT underwent catheter ablation in three Italian hospitals. High-density right atrial mapping was performed during sinus rhythm using a 16-pole grid catheter (Advisor™ HD Grid) with OTNF on the EnSite X system. PF maps were generated (high-frequency threshold > 250 Hz), and RF applications were delivered to the region of highest PF. Voltage maps were also created to delineate low-voltage bridge (LVB) regions (typically 0.04–2 mV) for retrospective overlap/localization analyses; LVB was not used to select ablation targets. Results Acute procedural success was achieved in all patients, with no procedural complications. The effective ablation site (junctional beats with subsequent SP modification and/or non-inducibility) corresponded to the highest PF region in 45/50 patients (90%). In a retrospective overlap/localization analysis, the effective site fell within the LVB region in 36/50 patients (72%); concordance between PF-defined and LVB-defined regions at the effective site was observed in 35/50 patients (70%). Mean PF in the SP target region was 424 ± 86 Hz versus 272 ± 57 Hz in surrounding tissue ( P < 0.05 on paired testing). Exploratory ROC analysis yielded an AUC of 0.937 with an optimal cutoff of 357 Hz (sensitivity 0.87; specificity 0.94); these estimates should be interpreted cautiously because PF guided RF target selection (incorporation/target-selection bias). Conclusion In this prospective multicenter cohort, OTNF-derived PF mapping was feasible and reproducibly identified a discrete region used to guide SP ablation in typical AVNRT. ROC-derived thresholds and performance estimates are exploratory and require independent validation in cohorts where ablation targets are not selected using PF criteria.