Performance and Limitations of Multipolar Electrogram Mapping in Scar-Related Ventricular Tachycardia

Background Multipolar electrogram is designed to improve near-field (NF) signal clarity by subtracting far-field (FF) components using adjacent electrodes. Objective To assess the performance of multipolar electrogram mapping in identifying critical activation pathways and substrates in ventricular tachycardia (VT). Methods Scar-related VT maps created using the CARTO™3 system and Optrell™ catheter were analyzed. Bipolar VT activation maps were created using wavefront algorithm (annotating unipolar maximum -dV/dt), while substrate maps employed late annotation mapping (LAM). Bipolar maps were converted into multipolar maps offline using the CARTOREPLAY TM module. Electrograms with both NF and FF components—including local abnormal ventricular activities (LAVA), late potentials (LP), and diastolic signals—were compared between bipolar and multipolar maps. Auto-annotation accuracy was evaluated. Results A total of 3,636 electrograms from 53 maps (45 patients, age 64±9 years, 87% male, 57% ischemic) were analyzed. Multipolar signals showed higher NF/FF amplitude ratio than bipolar signals (1.50 [0.61–3.08] vs. 1.19 [0.52–2.60], p<0.0001). Multipolar activation mapping more accurately identified diastolic potentials than bipolar wavefront mapping (76% vs. 46%, p<0.0001). However, multipolar substrate mapping was less sensitive in detecting LAVA/LP than bipolar LAM (80% vs. 92%, p<0.0001). Of 30 maps with LPs, 87% had LPs missed by the multipolar algorithm but captured by LAM. Smaller NF / larger FF amplitudes and perimeter electrodes predicted multipolar annotation errors. Conclusions Multipolar activation mapping enhances diastolic potential detection but reduces sensitivity for VT substrate identification compared to bipolar LAM. These findings inform optimal electrogram annotation approaches for VT activation and substrate mapping.