Background:
Power levels influence time-dependent lesion formation during catheter ablation. Emerging technologies combine assessment of physical tip-tissue contact (via contact force, CF) and electrical coupling as well as energy transfer to tissue (via baseline local impedance, LI and the maximum local impedance drop, ∆LI) which assist lesion size estimation. However, whether application of different power and CF levels interact and influence LI dynamics remains unclear. 

 

Objective:
To investigate the impact of applied power and CF on LI dynamics in relation to underlying myocardial substrate. 

 

Methods:
Consecutive patients presenting for atrial fibrillation (AF) / atrial tachycardia (AT) catheter ablation after previous failed AF ablation were prospectively enrolled. Following basket catheter-guided high-density mapping, ablation was performed using an open-irrigated single-tip catheter (IntellaNav StablePoint, Boston Scientific, Marlborough, MA, USA) capable of continuously assessing CF and LI. Radiofrequency current (RFC) applications were compared between power levels of ≤25/30/35/40 W. A cut-off level of mean CF during ablation was determined to predict a ∆LI ≥20 Ω which was reported to result in transmural lesion formation in previous studies. The time course of RFC applications was analyzed regarding time until ∆LI. Baseline LI tertiles were determined to classify underlying myocardial substrate into low, intermediate and high baseline LI.  

 

A total of 48 patients (52 % female, mean age 68±11 years, 50 % with persistent AF, median of 1 (interquartile range 1-2) prior ablations, 43 % with heart failure) were enrolled. Conducted procedures included repeat AF (68 %), left AT (65 %) and right AT (14 %) ablation. A total of 1410 RFC applications were analyzed (power levels: ≤25 W, 18 %; 30 W, 58 %; 35 W, 15 %; ≥40 W, 9%). Receiver operating characteristics analysis determined a mean CF of ≥13 g as the best predictor of a ∆LI ≥20 Ω (AUC 0.65). 

Correlation of mean CF to ∆LI during ablation was enhanced with power levels ≥30 W, while RFC applications ≥40 W showed the highest correlation (Figure 1A). A mean CF level ≥13 g led to a significant increase in ∆LI with increasing power ≥30 W (p<0.001), whereas an increase in power did not increase ∆LI during a mean CF level <13 g regardless of applied power (Figure 1B). The LI drop rate until reaching ∆LI differed significantly between power levels when applying a mean CF <13 g (p<0.001), with the fastest drop being observed during 30 W (1.4±0.9 Ω/s) and slowest with 35 W (0.9±0.5 Ω/s). However, no difference was observed in LI drop rate (p=0.52) and time until reaching ∆LI (p=0.96) when applying ≥13 g between power levels.  

Increase of applied power did not alter ∆LI in regions of the low baseline LI tertile (baseline LI <138 Ω, p=0.052). In the intermediate baseline LI tertile group (138-153 Ω), a CF <13 g only influenced ∆LI when increasing power from ≤25 W to 40 W (∆LI: 14 vs 25 Ω, p<0.001), while in the high tertile group (≥154 Ω) a power of 30 W already resulted in increased ∆LI when applying a mean CF <13 g (p<0.001). In regions of intermediate and high baseline LI, a mean CF ≥13 g enhanced ∆LI with every power increase when comparing ≤25 W vs ≥30 W vs 40 W (∆LI 19 Ω vs 25 Ω vs 40 Ω, p<0.001). 

Chosen power levels impact the interaction of CF and LI depending on underlying baseline LI which should be considered during repeat atrial arrhythmia ablation.