Validation of cardiac magnetic resonance tissue tracking in the rapid assessment of RV function: a comparative study to echocardiography

Abstract

To investigate the accuracy of cardiac magnetic resonance (CMR) tissue tracking (CMR-TT) and speckle tracking echocardiography (STE) against CMR determined right ventricular (RV) ejection fraction (RVEF) and to identify an optimal cut-off value for STE and CMR-TT to determine RVEF <45% and compare this to other conventional methods for estimating RVEF in dilated cardiomyopathy (DCM) patients. Twenty-nine DCM patients were recruited prospectively. CMR and echocardiography were performed within 48 hours and four-chamber views were used for strain analysis. Contoured CMR short axis images provided RVEF. Intraclass correlation coefficient (ICC), bias, levels of agreement, and receiver operating characteristic (ROC) curve analyses were performed. CMR-TT RV free-wall longitudinal strain (FLS) and STE RV global longitudinal strain (GLS) showed the best correlation with RVEF (r=-0.68, r=-0.82, p<0.001 respectively). There was moderate correlation between echocardiography RV GLS and CMR RV FLS (r=0.64, p<0.001). CMR-TT FLS showed excellent intra-observer and interobserver reliability (ICC=0.980; ICC=0.968 respectively). STE GLS correlated better with RVEF than with peak systolic annular velocity (S'; r=0.45), tricuspid annular plane systolic excursion (TAPSE; r=0.56), and fractional area change (FAC; r=0.78). CMR-TT RV FLS had better correlation with RVEF than CMR TAPSE (r=0.69 versus 0.40). ROC analysis demonstrated the optimal cut-off value for CMR-TT RV FLS and STE GLS in detection of RVEF <45% was ≥-24.4% (area under the curve=0.87, 100% sensitivity, 66.7% specificity) and ≥-20.9% (area under the curve=0.88, 100% sensitivity, 60% specificity) respectively. CMR-TT FLS and STE GLS showed potential to provide rapid assessment of RV function and had superior correlation with RVEF compared to conventional parameters.