Inverse calculation of QRS-T epicardial potentials from body surface potential distributions for normal and ectopic beats in the intact dog.

Abstract

Inverse epicardial potential distributions was calculated from potential distributions measured from the body surface in 12 intact dogs, divided into 6 pairs. The calculation procedure made use of measurements from the initial dog of each pair, giving the geometric location of each epicardial and body surface electrode and the approximate variance of the epicardial potentials and of the body surface noise. The same calculation procedure subsequently was applied to the other dog of the pair. For all dogs, inverse solutions were calculated throughout QRS-T for several sequences of excitation and repolarization which were produced by stimulating singly and in pairs any of 8 ventricular sites. All calculated inverse results were checked by detailed quantitative comparison to the corresponding measured epicardial potential distributions, which were obtained from chronically implanted epicardial electrodes. The root square (RMS) numerical differences between the inverse computed and the measured epicardical distributions were a substantial fraction of the RMS measured epicardial voltages, often 0.7 or more, and the correlation coefficients between measured and computed epicardial distributions were in the range 0.6-0.8. The major epicardial events of excitation and repolarization easily were seen in all of the inverse maps in the initial dogs of each pair, and with only slightly less clarity in the subsequent dogs. Events readily observed included the initial minimum around the stimulus site as excitation began, the movement of the zero contour line across the epicardium as excitation progressed, and the characteristic pattern of repolarization with a maximum near the stimulus site. Specific physical features of epicardial potential distributions can be determined from body surface measurements and can be verified by comparison with direct epicardial measurements.