ABSTRACT
OBJECTIVES: Accurate assessment of right ventricular (RV) systolic function is important, as it is an established predictor of mortality in cardiac and respiratory diseases. We aimed to compare speckle tracking-derived longitudinal deformation measurements with traditional two-dimensional (2D) echocardiographic parameters, as well as real time three-dimensional echocardiography (RT3DE) and cardiac magnetic resonance imaging (CMR)-derived RV volumes and ejection fraction (EF). METHOD: Subjects referred for CMR also underwent echocardiography. On both RT3DE and CMR, we measured RV volumes and EF. On 2D echocardiography, we analyzed RV fractional area change, RV internal diastolic diameter, tricuspid annular plane systolic excursion, tricuspid annular tissue Doppler-derived velocity, myocardial performance index, and RV global longitudinal strain (RV GLS). RESULTS: Sixty subjects were recruited (mean age = 45 ± 10 years; 60% male). RV GLS (R = -0.69, P < 0.001) and RT3DE RVEF (R = 0.56, P < 0.001) correlated well with CMR RVEF. RT3DE RV end-diastolic (RVEDV) and end-systolic (RVESV) volumes also correlated with CMR RV volumes: RVEDV, R = 0.74, P < 0.001 and RVESV, R = 0.84, P < 0.001. In addition, RV GLS best predicted the presence of RV dysfunction, defined as RVEF <48% on CMR (hazard ratio = 7.0 [1.5-31.7], P < 0.01). On receiver operator characteristic analysis, a RV GLS of -20% was the most sensitive and specific predictor of RV dysfunction (AUC 0.8 [0.57-1.0], P < 0.02). CONCLUSION: RVEF and volumes estimated on RT3DE were closely correlated with CMR measurements. When compared to more traditional markers of RV systolic function and RT3DE, RVGLS produced the highest correlation with CMR RVEF and was a good predictor of RV dysfunction. RV GLS should be considered a complementary modality to RT3DE and CMR in the assessment of RV systolic function.
