Feasibility of Assessing Cardiac Systolic Function Using Longitudinal Fractional Shortening Calculated by Two-Dimensional Speckle Tracking Echocardiography

Abstract

The left ventricular longitudinal fractional shortening (LVLFS) evaluated by the speckle tracking method reflects the longitudinal contraction of left ventricle (LV). We aimed to evaluate the accuracy of LVLFS as an index of LV systolic function. Two hundred thirty-three patients with or without heart failure (HF) were divided into four groups according to the progress of HF. Conventional echocardiography and two-dimensional speckle tracking echocardiography (2DSTE) were performed in all subjects. Besides, cardiac magnetic resonance (CMR) images were acquired in a subgroup of 30 patients with poor echo image quality. LVEF was measured using the biplane Simpson's method by two-dimensional echocardiography and CMR, respectively. LVLFS was calculated using QLAB 7.0, which performed speckle tracking analysis automatically. Generally, LVLFS declined accordingly with the progression of HF (P < 0.05). In a liner regression model, LVLFS correlated well with conventional LVEF (r = 0.856, P < 0.001). In the subgroup study performed in patients with poor acoustic windows, LVLFS correlated better with LVEF measured by CMR than conventional LVEF measured using biplane Simpson's method (r = 0.899 vs. 0.848). As for the reproducibility, better intra- and interobserver variability was observed in LVLFS than in conventional LVEF (6.2 ± 3.5% vs. 8.5 ± 4.7% and 7.8 ± 5.6% vs. 11.3 ± 6.2%, P < 0.05), especially for patients with poor image quality. Besides, measurement of LVLFS is much less time-consuming than conventional LVEF. The results of this study suggest a linear relationship between LVLFS and LVEF. LVLFS may provide sensitive and accurate assessment of LV systolic function, and its measurement reproducibility is better than that of LVEF.