Comparison of two-dimensional speckle and tissue velocity based strain and validation with harmonic phase magnetic resonance imaging

Abstract

Two-dimensional (2-D) strain (epsilon2-D) on the basis of speckle tracking is a new technique for strain measurement. This study sought to validate epsilon2-D and tissue velocity imaging (TVI)-based strain (epsilonTVI) with tagged harmonic-phase (HARP) magnetic resonance imaging (MRI). Thirty patients (mean age 62 +/- 11 years) with known or suspected ischemic heart disease were evaluated. Wall motion (wall motion score index 1.55 +/- 0.46) was assessed by an expert observer. Three apical images were obtained for longitudinal strain (16 segments) and 3 short-axis images for radial and circumferential strain (18 segments). Radial epsilonTVI was obtained in the posterior wall. HARP MRI was used to measure principal strain, expressed as maximal length change in each direction. Values for epsilon2-D, epsilonTVI, and HARP MRI were comparable for all 3 strain directions and were reduced in dysfunctional segments. The mean difference and correlation between longitudinal epsilon2-D and HARP MRI (2.1 +/- 5.5%, r = 0.51, p <0.001) were similar to those between longitudinal epsilonTVI and HARP MRI (1.1 +/- 6.7%, r = 0.40, p <0.001). The mean difference and correlation were more favorable between radial epsilon2-D and HARP MRI (0.4 +/- 10.2%, r = 0.60, p <0.001) than between radial epsilonTVI and HARP MRI (3.4 +/- 10.5%, r = 0.47, p <0.001). For circumferential strain, the mean difference and correlation between epsilon2-D and HARP MRI were 0.7 +/- 5.4% and r = 0.51 (p <0.001), respectively. In conclusion, the modest correlations of echocardiographic and HARP MRI strain reflect the technical challenges of the 2 techniques. Nonetheless, epsilon2-D provides a reliable tool to quantify regional function, with radial measurements being more accurate and feasible than with TVI. Unlike epsilonTVI, epsilon2-D provides circumferential measurements.