Electromechanical wave imaging vs electrocardiographic imaging: a direct comparison of non-invasive ventricular...

Precise non-invasive identification of the site of origin (SoO) of ventricular arrhythmias (VA) could inform ablation strategies.

To compare spatial accuracy of ultrasound-based electromechanical wave imaging (EWI) and ECG imaging (ECGI) to estimate the anatomical and axial (endo- vs epicardial) SoO of focal VA or pace maps employing contact mapping as gold standard.

Patients awaiting a catheter ablation procedure underwent preprocedural EWI and ECGI to non-invasively map the SoO of VE/VT or RV and LV pacing sites. A commercial CT-ECGI system was used to reconstruct epicardial activation maps. Unipolar EGM morphology and slew rate were employed to estimate axial SoO. EWI was performed using high frame rate (2000fps) transthoracic echocardiography with simultaneous ECG. Contact mapping and pacing sites were used as gold standard to define SoOs.

Thirty-three patients with 36 maps in total were included, 24 patients for VE/VT activation and 9 for pace-mapping. ECGI correctly identified the segmental VA-SoO/pacing-site in 28/36 maps (77.8%) compared to 28/35 by EWI (80%, p = ns). Erroneous annotations in ECGI related to septal and papillary muscle foci, whereas EWI mostly misannotated in outflow tract and RV SoOs. One patient had insufficient VEs to allow EWI VA mapping. Reconstructed unipolar EGM features of ECGI maps did not reliably differentiate endo- from epicardial SoO. Direct transmural mapping with EWI correctly identified the "transmural" SoO in 27/35 (77.1%).

Both EWI and ECGI localized the anatomical SoO in the majority of cases with site-specific advantages but also shortcomings for both modalities. EWI determines the transmural SoO which cannot be reliably localized using ECGI. ECGI has the advantage of providing mapping of multifocal and/or infrequent VA by offering panoramic single beat mapping.