Quantification of left ventricular mechanics using vector-velocity imaging, a novel feature tracking algorithm, applied to echocardiography and cardiac magnetic resonance imaging

Abstract

Background: Tagged magnetic resonance imaging (MRI) is the non-invasive golden standard to measure myocardial deformity. Tissue Doppler Imaging can be used to assess myocardial deformity, however, it has the limitation of angle-dependence. Our study aimed to compare left ventricular torsion and strains measured by velocity-vector imaging (VVI) using echocardiography (echo-VVI) and MRI (MRI-VVI), and to validate them against harmonic phase tagged MRI (HARP MRI).

Methods: A total number of 34 subjects (14 normal and 20 patients) were evaluated. Apical and basal image of left ventricular short axis view were acquired for measurements of apical and basal rotation, circumferential and radial strain using both echo-VVI and MRI-VVI. An apical four-chamber view was obtained for measuring the distance between the apical and basal levels.

Results: The correlations of segmental rotations, circumferential and radial strains were high between echo-VVI and HARP MRI, while the agreement of apical rotation was poor. Left ventricular torsion showed much better correlation and agreement between echo-VVI and HARP MRI than apical rotation: the coefficient was 0.97, P < 0.001. The correlation between MRI-VVI and HARP MRI in quantifying rotational parameters and strains was similar with echo-VVI and HARP MRI. Echo-VVI could discriminate normal and dysfunctional ventricles on either hypertensive or dilated cardiomyopathy.

Conclusion: The data from this study show that (1) it is feasible to quantify left ventricular torsion and myocardial strain using echo-VVI and MRI-VVI in normal subjects, patients with left ventricular global systolic dysfunction and segment systolic dysfunction; (2) the agreement among all mechanical parameters derived from echo-VVI, MRI-VVI, and HARP MRI remained with clinically acceptable ranges.