A direct experimental study of three systems of spatial vectorcardiography.

Three commonly used systems of vectorcardiography are compared on an absolute and direct quantitative basis in terms of equations, geometric interpretations, scalar leads and vectorcardio-grams, using experimental results obtained on an accurate, three-dimensional homogeneous torso model of the human subject with a dipole fixed in position in the center of the heart. It is concluded that the accuracy of the three systems, as presently used, is unsatisfactory and that further consideration of the systems of Duchosal and Grishman is questionable because of excessive intrinsic errors. Certain chance properties of the Wilson tetrahedron enable standardization modifications which lead to fairly accurate results for some dipole positions, and this system also possesses other features which appear to be desirable. The applicability of the results and conclusions to the human subject depends upon the degree to which the human heart may be represented by a fixed-position dipole and the degree to which human body electrical inhomogeneities influence body surface potentials. RECO GNITION that the electrical forces in the heart are three-dimensional in nature has precipitated an international, active quest for accurate means of recording their spatial aspects. A variety of electrode arrangements on the body surface of the human subject are presently being utilized by many electrocardiographers.1-6 Most of these systems of vectorcardiography are founded upon the assumptions that the human torso is a substantially homogeneous volume conductor and that the heart forces may be represented by a single fixed-position, time-varying current dipole. The degree to which these assumptions are applicable to the human subject has not been firmly established. Additional assumptions are usually made by the advocates of various specific systems concerning the geometry of the body and the position of the dipole. However, it has been shown in this laboratory by exhaustive studies on hu-101 man torso models that the shape of the human torso and the position of the equivalent heart dipole are important factors which introduce large errors in currently employed methods of vectorcardiography, even granting that the torso medium is homogeneous and that the heart forces may be consolidated into a single fixed-position dipole. It is the purpose of this paper to extract from the tremendous amount of data obtained with male and female homogeneous torso models, moulded accurately to the human subject, representative findings which pertain to a few of the commonly used systems of spatial vectorcardiography. It is hoped that the shortcomings of all of these …