Frontal Plane Studies of Homogeneous Torso Models

Abstract

The accuracy of currently used vector methods of analysis of frontal-plane electrocardiograms and the behavior of the Wilson central-terminal voltage are investigated experimentally by means of an accurate unipolar voltage technic, using human torso models filled with a homogeneous conducting fluid in which is immersed a heart dipole. The errors entailed in the determination of the manifest heart vector from torso surface voltages by means of Einthoven's vector projection methods are found to be very sizable and strongly dependent upon the position of the heart dipole which is placed at various points within the heart volume in this investigation. Einthoven's angle α is found to depart from the actual heart-vector angle by –7 degrees to +41 degrees, at the instant when the manifest heart vector is a maximum, and the amplitude and shape of frontal-plane vector loops and limb leads are found to be considerably distorted, in many cases, as compared with the actual heart-dipole behavior. The Wilson central-terminal voltage is found to depart from its ideal value of zero by approximately 15 to 40 per cent of the maximum value of the manifest heart vector as seen in the frontal plane, in typical cases, and by more than 200 per cent in extreme cases. It is shown that the component of the dipole perpendicular to the frontal plane is not negligible in influencing the limb-electrode potentials. The results are found to be consistent within 10 per cent or better for a male and female torso differing widely in proportions and contour, except for an increase in the female torso voltages arising from its smaller physical dimensions. The results are presented principally in terms of the QRS complex, but the geometric triangle representation is given as are numerical coefficients which describe torso surface voltages.