The Definition of a Single Measure of Regional Wall Motion Abnormality in Scintigraphic Ventriculography

Abstract

We undertook this study to investigate if a single measure of focal or regional wall motion abnormalities could be devised. Since it can be shown that the global ejection fraction does not necessarily decrease until a substantial fraction of the myocardium has been damaged, and since global parameters are easily influenced by non-myocardial variables such as after-load, pre-load and heart rate, it seemed that a more precise measure of focal damage was needed.

Our approach is based on the hypothesis that in the absence of focal damage, each myocardial segment has a constant relative contribution to the global ejection fraction. While global, non-cardiac factors which influence the global ejection fraction will also influence all segments, they will not influence the relative (amongst segments) contraction of the segments. Hence, all data could be normalized to a “normal” global ejection fraction.

The first step of the method is analogous to the circumferential profile methods in Thallium myocardial perfusion scintigraphies; after the (automatic) delineation of the left ventricular borders and the isolation of left ventricular activities by interpolation from those borders, 255 radii evenly spaced over 360 degrees are defined, originating in the center of mass of the left ventricle computed in the end-diastolic frame. Along those radii counts are integrated over the outer half, between center and border, in the end-diastolic image, and over the identical pixels in the end-systolic image. From those data 256 radial ejection fractions are computed.

From 12 middle-aged volunteers a normal resting and a normal stress profile is then computed, with a standard deviation value for each radial value. The index of regional or focal wall motion abnormality is computed from the cumulative difference between corresponding points in the test cases versus the “normal” profile, expressed either in absolute units or in fractions of the local standard deviations. The difference, however, can be either the sum of the absolute differences, the sum of the square, or the cubes of the differences. Those definitions favor, respectively, wide or deep abnormalities.

To test the feasibility (and not yet the clinical efficacy), we studied 8 patients with infarction who for two years had no further cardiac problems, and 8 patients chosen from the same cohort but who did have a medical cardiac event during the same time. Both groups differ significantly from the “normal” and to some extent from each other.