Abstract 17967: Association of 3D Echocardiographic Principal and Planar Strains with Left Ventricular Remodeling Parameters and Abnormal Ejection Fraction by Cardiac Magnetic Resonance Imaging

Abstract

Background: 3D principal (P), circumferential (C), longitudinal (L) and radial (R) strains by speckle tracking echocardiography (ECHO) have been reported to be sensitive markers of left ventricular (LV) mechanical dysfunction. We explored the association of these parameters with markers of adverse LV remodeling as well as abnormal ejection fraction (EF) by cardiac magnetic resonance imaging (CMR). Methods: Subjects were prospectively recruited and underwent CMR and 3DECHO within 2 hours. 3D principal strain (P) was the 3D vector from which C, L, and R strain components were derived. CMR LV volumes (EDV, ESV), LV mass indexed to BSA (LVMi), LVM/EDV, LVM/ESV ratios and EF were also obtained. A Trend test was used to compare the variables across subgroups and ROC analysis was used to compare the sensitivity and specificity of each strain variable for association with CMREF<50%. . Results: 120 subjects (62% male, 68 normal (Gp1), 15 with heart failure and preserved LVEF (Gp2) and 37 with heart failure and LV dysfunction (Gp3)) were evaluated. There were significant differences in LVMi (49, 64, and 63 g/m2, p<0.001), in LVM/EDV (0.4, 0.9, 0.6 g/ml, p<0.001) in LVM/ESV (1.6, 2.3, 1.1 g/ml, p<0.001) and in LVEF (59%, 60%, 39%, p<0.001). Larger LVMi and larger LVM/EDV and LVM/ESV ratios were associated with lower 3D strain indices (all p values <0.001). Of the strain indices (table), the highest area under the curve for CMREF<50% was obtained for 3D L strain. However, 3D P strain had the best sensitivity (86.5%) and specificity (79.5%) at a cutoff value of -31.2%. Conclusions: 1) Higher indexed values of LV mass and abnormal LV mass/volume ratios, parameters of adverse LV remodeling are associated with lower 3D strain indices. 2) Cutoff values of -15.7% and -31.2% for 3D L strain and 3D P strain, respectively are sensitive and specific markers of reduced EF as measured by CMR in a mixed population of normals, heart failure patients with preserved LVEF and those with systolic dysfunction.