Inverse Solution Mapping of Epicardial Potentials

BACKGROUND: Catheter ablation of ventricular tachycardia (VT) is still one of the most challenging procedures in cardiac electrophysiology, limited, in part, by unmappable arrhythmias that are nonsustained or poorly tolerated. Calculation of the inverse solution from body surface potential mapping (sometimes known as ECG imaging) has shown tremendous promise and can rapidly map these arrhythmias, but we lack quantitative assessment of its accuracy in humans. We compared inverse solution mapping with computed tomography-registered electroanatomic epicardial contact catheter mapping to study the resolution of this technique, the influence of myocardial scar, and the ability to map VT. METHODS AND RESULTS: For 4 patients undergoing epicardial catheter mapping and ablation of VT, 120-lead body surface potential mappings were obtained during implantable defibrillator pacing, catheter pacing from 79 epicardial sites, and induced VT. Inverse epicardial electrograms computed using individualized torso/epicardial surface geometries extracted from computed tomography images were compared with registered electroanatomic contact maps. The distance between estimated and actual epicardial pacing sites was 13 ± 9 mm over normal myocardium with no stimulus-QRS delay but increased significantly over scar (P=0.013) or was close to scar (P=0.014). Contact maps during right ventricular pacing correlated closely to inverse solution isochrones. Maps of inverse epicardial potentials during 6 different induced VTs indicated areas of earliest activation, which correlated closely with clinically identified VT exit sites for 2 epicardial VTs. CONCLUSIONS: Inverse solution maps can identify sites of epicardial pacing with good accuracy, which diminishes over myocardial scar or over slowly conducting tissue. This approach can also identify epicardial VT exit sites and ventricular activation sequences.