Right ventricular (RV) systolic function is prognostically important, but its assessment by echocardiography remains challenging, in part because of the multitude of available measurement methods. The purpose of this prospective study was to rank these methods against the reference of RV ejection fraction (EF) as obtained in a broad clinical population by magnetic resonance imaging (MRI).
Two hundred and twenty-three individuals were included in the study. The following seven Doppler echocardiographic parameters were tested using receiver operating characteristic (ROC) analysis for their accuracy to distinguish between normal and moderately impaired RVEF by MRI (RVEF cut-off 50%), respectively, between moderately and severely reduced RVEF (cut-off 30%): RV fractional area and fractional long-axis change (FLC), RV myocardial performance index (MPI), tricuspid annular peak systolic excursion, Doppler tissue imaging-derived isovolumic acceleration and peak systolic velocity (S') at the lateral tricuspid annulus, and strain at the lateral free wall as obtained by speckle-tracking echocardiography. Survival analysis was performed. All seven Doppler echocardiographic parameters correlated significantly with RVEF by MRI (range between 5 and 85%). RVEF <50% was best detected by S' < 11 cm/s: area under the ROC curve 0.779 (95% confidence interval 0.716-0.843), sensitivity 0.740, and specificity 0.753. RVEF ≤30% was best detected by MPI > 0.50: area under the ROC curve 0.948 (95% confidence interval 0.906-0.991), sensitivity 0.947, and specificity 0.852. The Kaplan-Meier analysis revealed reduced cumulative survival among patients with RVEF ≤30% (P = 0.0003).
A systolic long-axis peak velocity of <11 cm/s at the lateral tricuspid annulus most accurately detects moderately impaired RVEF as obtained by MRI; severely reduced RVEF ≤30% is best detected by RV MPI at a value of >0.50.