DIAGNOSTIC METHODS VENTRICULAR PERFORMANCE Advantages and applications of the centerline method for characterizing regional ventricular function

We sought to identify theoretical advantages and applications of the centerline method for quantitative assessment of regional ventricular function. Motion was measured along 100 chords constructed perpendicular to a centerline drawn midway between the end-diastolic and end-systolic contours, and normalized for heart size. Abnormality was expressed in units of standard deviations from the mean motion in a normal reference population to indicate both the severity and significance of the wall motion abnormality. The mean abnormality averaged over 100 chords correlated highly with the area ejection fraction (r = .97). The centerline method uses a "sliding window" to measure motion where it is abnormal, because assessment of wall motion in predefined regions of the ventricular contour underestimates abnormality. From the 100 data points, the extent (% of contour) of regional abnormalities can also be determined. The severity of hypokinesis at the site of acute myocardial infarction correlated better with infarct size estimated from creatine kinase release (r = .78) than did the ejection fraction or the circumferential extent of hypokinesis. Because the centerline method measures motion along locally determined vectors, and requires no apex, origin, coordinate system, or geometric reference figure, it can be applied to contours as dissimilar as the 60 degree left anterior oblique projection of the left ventricle and the 75 degree left anterior oblique projection of the right ventricle. Circulation 74, No. 2, 293-305, 1986. IN CURRENT CLINICAL PRACTICE, wall motion is assessed subjectively from contrast ventriculograms. The poor reproducibility of such qualitative evaluations' has led to the development of quantitative methods that measure the extent of wall motion at 4 to 100 points around the endocardial contour.'More recent studies have questioned the assumptions on which many of these methods are based. The validity of radial methods has been challenged by reports showing that motion proceeds toward many points rather than a single point in the ventricular chamber, and that motion measurement along radii from a single origin is erroneous."5 16 Assessments of intraobserver and interobserver variability indicate that the apex is one of the least reliably visualized parts of the contour; the dependence of rectangular and radial methods on the apex as a landmark in defining a long axis may introduce error in wall motion measureFrom the Cardiovascular Research and Training Center, University of Washington, Seattle, and University Hospital Eppendorf, Hamburg, West Germany. Supported in part by NHLBI grants HL-27819 and HL-19451. Address for correspondence: Florence H. Sheehan, M.D., University of Washington RG-22, Seattle, WA 98195. Received March 14, 1986; accepted April 17, 1986. ment.'7' 18 Finally, due to technologic advances in imaging and expanding interests, there is a need for a method that can measure regional function in both right and left anterior oblique (RAO; LAO) projections, in longand short-axis views, and in the right and left ventricles. The centerline method of regional wall motion analysis was developed to address these concerns. Its sensitivity and specificity in detecting abnormal regional function compare favorably with those of other methods'9 and its reliability has been previously documented.'7 The centerline method has been applied extensively to analyze the efficacy of thrombolytic therapy in salvaging left ventricular function.2023 However, the clinical significance of reduction in the severity of hypokinesis at the site of acute myocardial infarction has not been established, and the clinical relevance of measurement of the severity of hypokinesis at the infarct site as opposed to measurement of the circumferential extent of hypokinesis has not been determined. Therefore, we performed a study to (1) analyze the relationship between regional and global ventricular function, (2) compare the correlation of the extent and the severity of regional hypokinesis and the ejection Vol. 74, No. 2, August 1986 293 by gest on M ay 4, 2017 http://ciajournals.org/ D ow nladed from