Measuring left ventricular global longitudinal strain in different contrast-enhanced echocardiography modes: a feasibility study

Left ventricular global longitudinal strain (LVGLS) is an important parameter to evaluate left ventricular (LV) systolic function, which is more sensitive and reproducible than left ventricular ejection fraction (LVEF). However, whether LVGLS can be accurately measured with ultrasound enhancing agent (UEA) remains controversial. The aim of this study was to verify the feasibility and reproducibility of LVGLS in different contrast-enhanced echocardiography (CE) modes. Two-dimensional speckle tracking echocardiography (2D-STE) was used to measure LVGLS on 75 patients included in this study. Patients were divided into optimal group ( n = 33) and suboptimal group ( n = 42) according to image quality. LVGLS measurements were performed in 2D mode, left ventricular opacification (LVO) mode, and myocardial contrast echocardiography (MCE) mode. 2D mode LVGLS measurements of the optimal group were used as a reference standard. The comparisons among different modes of each group included Bland-Altman analysis, linear regression and intra-class correlation (ICC). 20 patients were randomly selected from each group to analyze the inter- and intra-observer variability. The LVO and MCE modes of optimal group had similar LVGLS measurements with 2D mode (-14.30 ± 4.73% 2D vs. -15.32 ± 4.78% LVO vs. -15.36 ± 5.41% MCE , ICC: 0.814 2D−LVO vs. 0.781 2D−MCE , P < 0.001). There was good agreement between LVO and MCE mode in two groups (ICC: 0.855 optimal vs. 0.935 suboptimal , P < 0.001). In suboptimal group, both LVO and MCE mode had poor agreement with 2D mode in terms of LVGLS measurements (-13.86 ± 5.46% 2D vs. -15.39 ± 7.74% LVO vs. -15.66 ± 7.78% MCE , ICC: 0.739 2D−LVO vs. 0.687 2D−MCE, P < 0.001). CE improved inter-observer variability in suboptimal group (ICC: 0.630 2D vs. 0.864 LVO vs. 0.830 MCE , P = 0.022, P < 0.001, P = 0.001, respectively). It is feasible to measure LVGLS in different CE modes, even in patients with suboptimal acoustic windows, which can reduce the impact of image quality on LVGLS measurements.