Layer-Specific Strain Is Preload Dependent: Comparison between Speckle-Tracking Echocardiography and Cardiac Magnetic Resonance Feature-Tracking

Abstract

Speckle-tracking echocardiographic (STE) imaging and cardiac magnetic resonance feature-tracking (CMR-FT) are novel imaging techniques enabling layer-specific quantification of myocardial deformation. Conventional echocardiographic parameters are load dependent, but few studies have investigated the effects of loading conditions on STE and CMR-FT layer-specific strain and the interchangeability of the two modalities. The aim of this study was to evaluate the effects of acute preload augmentation by saline infusion on STE and CMR-FT longitudinal and circumferential layer-specific strain parameters and their intermodal agreement. A total of 80 subjects, including 41 control subjects (mean age, 40±12years; 49% men) and 39 patients with cardiac disease (mean age, 47±15years; 92% men) were examined using STE and CMR-FT layer-specific strain analysis before and after saline infusion (median, 2.0L) with quantification of transmural global longitudinal strain (GLS), epicardial GLS, endocardial GLS, transmural global circumferential strain (GCS), epicardial GCS, and endocardial GCS in addition to epicardial-endocardial gradients. Bland-Altman plots and Pearson correlation coefficients were used to evaluate agreement between the two modalities across all strain parameters. Acute saline infusion increased all STE and CMR-FT layer-specific strain parameters in both groups. STE and CMR-FT GLS increased by 1.4±1.5% and 1.5±2.0% (P<.001) in control subjects and by 0.9±1.8% and 0.9±1.9% (P<.001) in patients with cardiac disease. STE and CMR-FT GCS increased by 2.0±2.2% and 1.8±2.3% (P<.001) in control subjects and by 1.8±2.3% and 1.7±3.6% in patients with cardiac disease (P<.001 and P=.03). STE longitudinal strain correlated strongly with corresponding CMR-FT longitudinal strain (GLS, epicardial GLS, and endocardial GLS: r=0.81, r=0.82, and r=0.81, respectively) despite poor intermodal agreement (bias±limits of agreement, -2.84±4.06%, 0.16±3.68%, and 2.33±3.52%, respectively) whereas GCS, epicardial GCS, and endocardial GCS correlated weakly between the two modalities (r=0.28, r=0.19, and r=0.34, respectively) and displayed poor intermodal agreement (bias±limits of agreement, -1.33±6.86%, 4.43±6.49%, and -9.92±8.55%, respectively). STE and CMR-FT longitudinal and circumferential layer-specific strain parameters are preload dependent in both control subjects and patients with cardiac disease. STE and CMR-FT longitudinal layer-specific strain parameters are strongly correlated, whereas circumferential layer-specific strain parameters are weakly correlated. STE and CMR-FT longitudinal and circumferential strain should not be used interchangeably, because of poor intermodal agreement.