Variability of right ventricular global and segmental longitudinal strain measurements.

ABSTRACT

OBJECTIVES: We evaluated intra-observer, test-retest, and inter-observer reproducibility of right ventricular (RV) longitudinal strain (LS) measurements derived from two-dimensional speckle tracking echocardiography. BACKGROUND: The assessment of RV deformation has demonstrated to hold valuable prognostic and diagnostic data in clinical practice. Data about the reproducibility of the measurements, however, are missing. METHODS: In 200 subjects, apical modified four-chamber view focused on the right ventricle (RV) was obtained. In 150 subjects (75%), a second set of images was acquired for test-retest analysis. Global RV LS (RVGLS-4C), free wall strain (RVFW-4C), and segmental maximum peak strain (base, mid, and apex) were calculated (EchoPac vBT13, GE Vingmed Ultrasound, Horten, Norway). Additionally, the reliability and feasibility of RV free wall strain derived from single wall tracking (RVFW-SWT) were tested. The absolute difference between repeated measurements and inter-class correlation coefficients (ICC) for consistency was calculated. RESULTS: RVGLS-4C and RVFW-4C measurement could be obtained in all subjects. RVGLS-4C demonstrated lower intra-observer, test-retest, and inter-observer absolute difference compared to RVFW-4C (1.4 ± 1.2%, 1.8 ± 1.5%, and 1.7 ± 1.9%, respectively, vs 2.4 ± 2.4, 2.2 ± 1.9, and 2.6 ± 3.0, respectively, P < 0.01). ICC ranged between 0.83 and 0.90 for RVGLS-4C measurements and between 0.76 and 0.79 for RVFW-4C measurements. RVFW-SWT was feasible in 82.5% of the subjects and showed noninferior reproducibility compared to RVFW-4C. Basal and apical segments demonstrating the highest variability. CONCLUSIONS: The measurement of RVGLS-4C appears to be reliable in clinical practice and demonstrates better reproducibility compared to RV free wall LS measurements. Segmental LS measurements presented high variability and such parameters should be interpreted with caution.