Impact of the Position of Ectopic Ventricular Activation Origin on Its Noninvasive Assessment: A Simulation Study.

The accuracy of a method for noninvasive localization of the origin of premature ventricular complex (PVC) was tested on simulated data. The method is based on inverse solution in terms of a single dipole that uses integrals of body surface potentials (BSPs) and a model of the torso volume conductor. ECGSIM software was used to simulate ectopic activation in a realistic ventricular model and BSPs on an inhomogeneous torso. Ectopic activation origins were simulated in 120 positions located in anterior, lateral, posterior, and septal segments of the right and left ventricle (RV, LV). In both septal segments 15 endocardial positions were specified, in all other segments 5 epicardial, 5 endocardial, and 5 in the valvular region were defined. Corresponding BSPs for initial 5 ms interval of the ectopic activity were simulated in 300 torso nodes as well as in 192 points representing BSP mapping leads. Simulated BSPs and the torso model with lungs and heart cavities were then used for the inverse localization of each PVC origin. The localization error (LE) was defined as the distance between the simulated and inversely estimated PVC origin position. All simulated PVC origins were localized in the correct ventricular segment. There was no significant difference between the LEs in the RV and LV, except the septal segments where the LE was significantly smaller in the RV (p<0.05). There were no significant differences between the LEs obtained using BSP maps from 300 torso nodes or 192 mapping leads (mean values LE300=5.2±4.4 mm and LE192=5.3±4.5 mm). [ABSTRACT FROM AUTHOR]