Right ventricular diastolic function in aging: a head-to-head comparison between phase-contrast MRI and Doppler echocardiography

• Published in: International Journal of Cardiovascular Imaging Vol. 37; no. 2; pp. 663 - 674
• Main Authors: Kachenoura, Nadjia, Bollache, Emilie, Soulat, Gilles, Clément-Guinaudeau, Stéphanie, Ashrafpoor, Golmehr, Perdrix, Ludivine, Diebold, Benoit, Ladouceur, Magalie, Mousseaux, Elie
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Netherlands 01.02.2021; Springer Nature B.V; Springer Verlag
• Subjects: Adult; Age; Age Factors; Aging; Bioengineering; Cardiac Imaging; Cardiology; Cardiology and cardiovascular system; Coefficient of variation; Contrast Media; Correlation coefficients; Doppler effect; Echocardiography; Echocardiography, Doppler - methods; Female; Flow velocity; Heart; Heart Ventricles - diagnostic imaging; Human health and pathology; Humans; Image Enhancement - methods; Imaging; Life Sciences; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; Male; Medicine; Medicine & Public Health; Middle Aged; Original Paper; Parameter estimation; Radiology; Regression analysis; Reproducibility; Reproducibility of Results; Velocity; Ventricle; Ventricular Function, Right - physiology; Diastolic function; Aging; Right ventricle; Phase-contrast MRI
• ISSN: 1569-5794; 1875-8312; 1573-0743; 1875-8312
• DOI: 10.1007/s10554-020-02040-y

To evaluate right ventricle (RV) diastolic function from phase-contrast MRI (PC-MRI) in aging. 89 healthy individuals (50 men, 43 ± 15 years) underwent cardiac MRI including 2D PC-MRI (1.5T) and reference Doppler echocardiography of both ventricles on the same day. Conventional echocardiographic parameters were estimated: early (E, cm/s) and atrial (A) peak velocities as well as myocardial early peak longitudinal velocity (E′). PC-MRI images were analyzed using custom software, providing: E′, E and A waves along with respective peak flow rates (Ef, Af, mL/s) and filling volume (mL), for both ventricles. Intra- and inter-observer reproducibility was studied in 30 subjects and coefficients of variation (CoV) as well as intra-class correlation coefficients (ICC) were provided. RV diastolic function indices derived from PC-MRI data were reproducible (CoV ≤ 21%, ICC ≥ 0.75) and reliable as reflected by significant associations with left ventricular diastolic function indices assessed using both echocardiography (linear regression Pearson correlation coefficient r ≤ 0.59) and PC-MRI (r ≤ 71). Despite the fair associations between RV echocardiography and PC-MRI (r ≤ 0.25), the highest correlation with age was obtained for MRI Ef/Af ratio (r = − 0.64, p < 0.0001 vs. r = − 0.40, p = 0.0001 for echocardiographic E/A). Among PC-MRI E/A ratios, highest correlations with age were observed for flow rate and mean velocity ratios (r = − 0.61, p < 0.0001) as compared to maximal velocity ratios (r = − 0.56, p < 0.0001). Associations with age for E′ were equivalent between PC-MRI (mean velocity: r = − 0.40, p < 0.0001; maximal velocity: r = − 0.36, p = 0.0005) and echocardiography (r = − 0.36, p = 0.0006). Finally, the significant and age-independent associations between RV mass/end-diastolic volume and E′ were stronger for PC-MRI (mean velocity: r = − 0.36, p = 0.0006; maximal velocity: r = − 0.28, p = 0.007) than echocardiography (r = − 0.09, p = 0.38). PC-MRI tricuspid inflow and annulus myocardial velocity parameters were reproducible and able to characterize age-related variations in RV diastolic function.