Optimal Technique for Measurement of Linear Left Ventricular Dimensions

Abstract

Echocardiographic assessment of the left ventricle (LV) begins with the measurement of linear dimensions that approximate its ellipsoid diameter. These linear dimensions have historically been measured at the basal level of the LV, which is not representative of its true diameter. The objective of this study was to determine the optimal level to measure LV cavity dimensions to more accurately estimate its size and mass. The derivation study included 75 patients who had a clinically indicated cardiac magnetic resonance (CMR) exam for ischemic heart disease (n=15), nonischemic cardiomyopathy (n=25), or normal studies (n=35). The three-chamber bright blood cine sequence was analyzed using a custom MATLAB program to measure the LV cavity diameter and wall thickness at 15 equidistant levels from base to apex. The linear measurements from each of these levels were compared against the CMR volumetric reference standard. The validation study included 100 patients who had a clinically indicated echocardiogram and CMR within 3days for ischemic heart disease (n=20), nonischemic cardiomyopathy (n=44), and normal or near-normal studies (n=36). The parasternal long-axis cine sequence was analyzed to measure the LV cavity diameter and wall thickness at the traditional basal level and the midventricular level, which were compared against the CMR volumetric reference standard. In both the derivation and validation studies, the midventricular linear dimensions, defined as those located at the true (maximal) diameter of the LV ellipsoid cavity, were found to be more closely correlated with the volumetric reference standard for LV mass, LV end-diastolic size, and LV ejection fraction. Measurement of linear dimensions at the midventricular level better reflects the ellipsoid geometry of the LV cavity and provides a more accurate estimate of LV mass, size, and systolic function as compared with the traditionally recommended basal level.