3D-echocardiography for evaluation of the right ventricle in patients with repaired Tetralogy of Fallot: a comparison study with cardiac magnetic resonance

Abstract

Abstract Background Patients with repaired Tetralogy of Fallot (ToF) need periodic instrumental assessments, due to complications they can meet during lifetime. Right ventricle (RV) dilation and dysfunction are some examples that explain why those patients regularly undergo Cardiac Magnetic Resonance (CMR), currently the gold standard method to detect these complications. Purpose 3D-echocardiography is an emerging tool to study RV, but there is little data supporting its use in congenital heart diseases. This study aimed at evaluating its accuracy in assessing RV dimensions and systolic function in ToF patients, in comparison to CMR. Methods 34 patients were prospectively enrolled after CMR performance. They all underwent standard 2D-echocardiography and Multi-Beat-ECG-triggered 3D Full-Volume acquisition. RV End Diastolic Volume (RV-EDV), End Systolic Volume and Ejection Fraction (RV-EF) were defined both in 3D-echocardiography and CMR. Results Post-hoc analysis of 3D images was performed through a vendor-independent software in 30 patients, resulting in a feasibility of 88%. A Bland-Altman analysis was performed, showing that 3D-echocardiography underestimates the RV-EDV/BSA of −27.39 mL/m2 on average (CI 95%: −29.46; −25.31, SD: 5.68 mL/m2), compared to CMR, as expected. However, when compared to normality reference, 3D-echocardiography identified RV dilatation with a significant correlation to CMR (V=0.66, p<0.001). Moreover, ROC curves showed that 3D-echo-RV-EDV/BSA is the echocardiographic measure that best correlates to CMR in detecting RV dilatation (AUC 1, CI 95%: 1.00; 1.00), followed by RVOT diameter in PLAX (AUC 0.89, CI 95%: 0.69; 0.99). Youden analysis suggested 77 mL/m2 (Se 95%, Sp 100%; J 0,95) as optimal limit to define RV dilatation in 3D-echocardiography. Regarding the RV-EF, the Bland-Altman analysis showed that 3D-echocardiography overestimates the RV-EF by 1,97% on average (CI 95%: 0.01; 3.95, SD: 5.43), compared to CMR. Secondly, accuracy of different echocardiographic parameters (both 2D and 3D) in detecting a RV systolic impairment was analyzed: ROC curves showed that RV-EF in 3D echocardiography is the measurement that best correlates to CMR in detecting RV dysfunction (AUC 0.92, CI 95%: 0.79–1.00), followed by Fractional Area Change (FAC) (AUC 0.78, CI 95%: 0.60; 0.96). Tricuspid Annular Plane Systolic Excursion (TAPSE) and tricuspid valve annular motion velocities in systole (S'-TDI), instead, showed very poor correlation. Conclusions 3D-echocardiography showed good agreement with CMR in defining RV dimensions and systolic function in ToF patients, even more than the bidimensional measurements cardiologists are used to. This could lead to a wider use of this tool in daily clinical practice, involving a reduction in the execution of the numerous CMR patients usually undergo. Moreover, it could be a valid instrument in patients with contraindication to CMR. Funding Acknowledgement Type of funding sources: None.