ECG vs ECGi to predict the area of interest in VT ablation in structural heart disease patients

Pre-procedural planning for ventricular tachycardia (VT) ablation in structural heart disease is critical for procedural success. Electrocardiogram (ECG) is widely used but has significant limitations in accurately localizing the VT exit site. Electrocardiographic imaging (ECGi) may be a promising alternative to predict VT breakout site.

To compare the accuracy of VT exit site/isthmus prediction, using ECG analysis by general cardiologists or electrophysiologists, and ECGi.

Single-center prospective study included patients (pts) with structural heart disease referred for left ventricular VT ablation. Clinical VT was induced in all patients using noninvasiveprogrammedventricularstimulationusingthe ICD.BoththeECGoftheclinical VT and VT mapping using ECGi were analysed to predict the exit site /isthmus of the VT. ECGs were analysed by three general cardiologists (GCs) and three electrophysiologists (EPs), in a blinded manner, using the 17-segment prediction algorithm validated by Berruezo. Predictions were considered accurate if they matched the VT exit site/isthmus with one-segment margin of error. ECGi was performed using a 252-electrode noninvasive 3D mapping system (CardioInsightTM) under mild sedation, after suspending anti-arrhythmic drugs for 72 hours. The ECGi area of interest was defined as the segment with the earliest activation region (initial 20 ± 5 ms from the first dV/dT). VT maps were segmented with ADAS 3D software and compared to LAVA location or invasive VT activation maps documented in invasive electrophysiology (EP) study. Paired T sampled test was performed for statistical analysis

We included 17 pts, 88% male, with a median age of 71 ± 13 years; 53% had ischemic cardiomyopathy. Physicians correctly predicted the VT exit site/isthmus in 48% of cases, with significantly higher accuracy among EPs compared to GCs (p < 0.001) - Table 1. ECGi achieved an accuracy of 88%, significantly outperforming both EPs and GCs (p=0.003) - Table 2. Furthermore, ECGi identified the exact VT exit site segment without the margin of error in 82.4% of patients, compared to 5.8% for physicians (EPs: 3.9%, GCs: 7.8%; p<0.001). Interoperator concordance was higher among EPs (76.5%) compared to GCs (66.7%), with an overall concordance of 71.6%. In cases where ECG analysis was inaccurate but adjacent to the corrected segment, the clinician prediction was more apical in 16% of cases and more basal in 2% than the corrected location observed in the EP study.

ECGi presents a superior accuracy than ECG in predicting the VT exit site/isthmus, presenting a valuable tool in pre-procedural planning of VT ablation in structural heart disease patients.