A novel assessment of local impedance during catheter ablation: initial experience in humans comparing local and generator measurements.

A novel measure of local impedance (LI) has been found to predict lesion formation during radiofrequency current (RFC) catheter ablation. The aim of this study was to investigate the utility of this novel approach, while comparing LI to the well-established generator impedance (GI). In 25 consecutive patients with a history of atrial fibrillation, catheter ablation was guided by a 3D-mapping system measuring LI in addition to GI via an ablation catheter tip with three incorporated mini-electrodes. Local impedance and GI before and during RFC applications were studied. In total, 381 RFC applications were analysed. The baseline LI was higher in high-voltage areas (>0.5 mV; LI: 110.5 ± 13.7 Ω) when compared with intermediate-voltage sites (0.1-0.5 mV; 90.9 ± 10.1 Ω, P < 0.001), low-voltage areas (<0.1 mV; 91.9 ± 16.4 Ω, P < 0.001), and blood pool LI (91.9 ± 9.9 Ω, P < 0.001). During ablation, mean LI drop (△LI; 13.1 ± 9.1 Ω) was 2.15 times higher as mean GI drop (△GI) (6.1 ± 4.2 Ω, P < 0.001). Baseline LI correlated with △LI: a mean LI of 99.9 Ω predicted a △LI of 12.9 Ω [95% confidence interval (12.1-13.6), R2 0.41; P < 0.001]. This relationship was weak for baseline GI predicting △GI (R2 0.06, P < 0.001). Catheter movements were represented by rapid LI changes. The duration of an RFC application was not predictive for catheter-tissue coupling with no further change of △LI (P = 0.247) nor △GI (P = 0.376) during prolonged ablation. Local impedance can be monitored during ablation. Compared with the sole use of GI, baseline LI is a better predictor of impedance drops during ablation and may provide useful insights regarding lesion formation. However, further studies are needed to investigate if this novel approach is useful to guide catheter ablation.

Informations de copyright

Published on behalf of the European Society of Cardiology. All rights reserved. © The Author(s) 2019. For permissions, please email: [e-mail].