Left ventricular global longitudinal strain calculated from manually traced endocardial border lengths utilizing the images for routine ejection fraction measurement by biplane method of disks

Purpose The purpose of this study was to test whether the fractional change in the endocardial border length between end-diastole and end-systole as manually traced in left ventricular ejection fraction (LVEF) measurement using the biplane method of disks (MOD) was consistent with the global longitudinal strain derived from speckle-tracking echocardiography. Methods For 105 patients who underwent echocardiography, two- and four-chamber images with manually traced endocardial lines for LVEF measurement by MOD were stored. LV endocardial lengths at end-diastole and at end-systole were measured on both images to calculate the fractional length changes, which were averaged (GLS(MOD)). Speckle-tracking analysis was performed to measure global longitudinal strains in the apical two- and four-chamber and long-axis images, and the three values were averaged (GLS(STE)) according to the ASE and EACVI guidelines. Results There was no significant difference between GLS(MOD) and GLS(STE). GLS(MOD) correlated well with GLS(STE) (r = 0.81, p < 0.001), and there was no fixed bias in the Bland-Altman analysis. The intraclass correlations for the intra- and inter-observer comparisons for GLS(STE) were excellent, and those for GLS(MOD) were adequate. Conclusion The fractional LV endocardial border length change, GLS(MOD), showed sufficient agreement with GLS(STE) to justify its use as a substitute for the STE-derived global longitudinal strain.