Assessment of Right Ventricular Longitudinal Strain in Patients with Ischemic Cardiomyopathy: Head-to-Head Comparison between Two-Dimensional Speckle-Based Strain and Velocity Vector Imaging Using Volumetric Assessment by Cardiac Magnetic Resonance as a "Gold Standard".

Abstract

Longitudinal strain of right ventricle (RV) can be used to determine RV systolic function. This study compared RV longitudinal strain values of two different speckle tracking software technologies, velocity vector imaging (VVI) and two-dimensional speckle tracking echocardiography (2DSTE), and longitudinal strain by cardiac magnetic resonance (CMR). We studied 36 patients (28 men, 63±11years) with ischemic cardiomyopathy (ICM) who underwent echocardiography with GE machines and CMR. Longitudinal strain of RV analyzed with 2DSTE and VVI in same DICOM files. Longitudinal RV strain analyzed with 2DSTE and VVI in same raw data. These values were compared with RVEF and longitudinal strain by CMR. VVI strain showed significant correlations with RVEF by CMR (global RV: r=-0.56, P<0.01, free wall: r=-0.52, P<0.01, and septum: r=-0.49, P<0.01). 2DSTE strain also revealed significant correlations (global RV: r=-0.40, P=0.02, and septum: r=-0.35, P=0.04). 2DSTE strain had significant bias with wide limits of agreement in global RV and septum compared with CMR strain. 2DSTE strain had significantly lower intra-observer variability than VVI (P=0.03) or CMR strain (P=0.04) in RV-free wall. RV longitudinal strains by VVI and 2SDTE demonstrated relatively good correlations with RVEF and longitudinal strain by CMR. However, when compared to CMR-derived strain, 2DSTE-derived strain underestimates longitudinal strain of RV septum and of global right ventricle. 2DSTE strain had significantly lower intra-observer variability compared with VVI or CMR strain analysis.