Right ventricular function parameters in pulmonary hypertension: echocardiography vs. cardiac magnetic resonance.

ABSTRACT

BACKGROUND: Right ventricular (RV) function is a major determinant of outcome in patients with pulmonary hypertension. Cardiac magnetic resonance (CMR) is gold standard to assess RV ejection fraction (RVEFCMR), however this is a crude measure. New CMR measures of RV function beyond RVEFCMR have emerged, such as RV lateral atrio-ventricular plane displacement (AVPDlat), maximum emptying velocity (S'CMR), RV fractional area change (FACCMR) and feature tracking of the RV free wall (FWSCMR). However, it is not fully elucidated if these CMR measures are in parity with the equivalent echocardiography-derived measurements tricuspid annular plane systolic excursion (TAPSE), S'-wave velocity (S'echo), RV fractional area change (FACecho) and RV free wall strain (FWSecho). The aim of this study was to compare regional RV function parameters derived from CMR to their echocardiographic equivalents in patients with pulmonary hypertension and to RVEFCMR. METHODS: Fifty-five patients (37 women, 62 ± 15 years) evaluated for pulmonary hypertension underwent CMR and echocardiography. AVPDlat, S'CMR, FACCMR and FWSCMR from cine 4-chamber views were compared to corresponding echocardiographic measures and to RVEFCMR delineated in cine short-axis stack. RESULTS: A strong correlation was demonstrated for FAC whereas the remaining measurements showed moderate correlation. The absolute bias for S' was 2.4 ± 3.0 cm/s (relative bias 24.1 ± 28.3%), TAPSE/AVPDlat 5.5 ± 4.6 mm (33.2 ± 25.2%), FWS 4.4 ± 5.8% (20.2 ± 37.5%) and for FAC 5.1 ± 8.4% (18.5 ± 32.5%). In correlation to RVEFCMR, FACCMR and FWSecho correlated strongly, FACecho, AVPDlat, FWSCMR and TAPSE moderately, whereas S' had only a weak correlation. CONCLUSION: This study has demonstrated a moderate to strong correlation of regional CMR measurements to corresponding echocardiographic measures. However, biases and to some extent wide limits of agreement, exist between the modalities. Consequently, the equivalent measures are not interchangeable at least in patients with pulmonary hypertension. The echocardiographic parameter that showed best correlation with RVEFCMR was FWSecho. At present, FACecho and FWSecho as well as RVEFCMR are the preferred methods to assess and follow up RV function in patients with pulmonary hypertension. Future investigations of the CMR right ventricular measures, beyond RVEF, are warranted.