ERRORS INVOLVED IN RADIOLOGICAL HEART VOLUME DETERMINATION BY THE ELLIPSOID-APPROXIMATION TECHNIQUE

Open Access
Publisher Website
Google Scholar
Abstract
Post-mortem evidence indicates that the basic assumption involved in the ellipsoid-approximation technique is applicable to the great majority of human hearts. The biggest potential error involved in heart volume estimation in life by this technique is observer variation in making the axis measurements from the X-ray films. When this error is combined with technical error (including that imposed by cardiac cycle changes) it is found that measured differences of 160 ml or more are significant (p< 0.05) of real volume change in an individual heart, and that differences of 210 ml or more similarly signify real differences in the volumes of 2 different hearts. Axis "b" is best simply measured, from the junction of pulmonary conus and left heart border, as nearly perpendicular to "1" (the long axis of the cardiac ellipse) as possible; to facilitate this measurement some extrapolation of the right lower cardiac contour is permissible. "d" is best measured as the horizontal diameter of the cardiac ellipse as seen in the lateral projection. With erect subjects, 2 m postero-anterior and 1.5 m left lateral focus film distances, and skin film distances of approximately 0.5 cm and 3.5 cm accurate estimates of over-all heart volume (V) are provided by the formula V=0.41.1.b.d The method is unsuitable for use where the heart is of bizarre configuration, where the relevant parts of its contour cannot be clearly seen in the X-ray films, and where the subject is too dyspneic to maintain the full inspiratory position for a few seconds.

This publication has 15 references indexed in Scilit: