Short period of training in 3D echocardiography provides good feasibility and reproducibility of right ventricular assessment in heart failure

Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): PROHEART 3D. Background Right ventricular (RV) dysfunction is an independent risk factor for increased mortality and morbidity. Standard approach for assessing is 2-dimensional echocardiography (2DE). However, 2DE is prone to errors due to increased variability of the anatomical cut-planes and geometrical assumptions due to the complex triangular and "wrapped-around" RV shape. Meanwhile, 2DE reflects mainly the RV longitudinal function, at the expense of the radial and global RV functions. 3D echocardiography (3DE) enabled itself as a more accurate method for assessment of cardiac volumes and calculation of ejection fraction (EF), but with a cumbersome learning time-curve. Objectives To establish if trainees in 3DE provide feasible and reproducible 3D measurements of RV size and function, by comparison to 2DE assessment, after a short period of training by an expert in echocardiography. Methods 161 consecutive patients (58±17 years, 71% males, 3D left ventricular EF 35±10%), hospitalized for heart failure (HF), underwent standard precordial 2DE acquisitions and 3DE multi-beat full-volumes of the RV. One expert in both 2DE and 3DE, and one fellow trained for 2 years in 2DE and for 3 months in 3DE, performed measurements of both 2DE and 3DE parameters for RV size and function, in a blinded fashion. 2DE parameters were measured first, in the following order: RV diameter, RV end-diastolic (RVED) and end-systolic (RVES) areas, RV fractional area change (RV FAC), tricuspid annular plane systolic excursion (TAPSE), tricuspid annular systolic velocity (RV S’), and RV free wall longitudinal strain (RV LS). Then, 3D RVED and RVES volumes were measured, and RV ejection fraction (RVEF) was calculated, using a semi-automated software package. Endocardial borders were corrected, as needed. Results Feasibility of 2DE and 3DE was 97% and 92%, respectively, patients being excluded due to poor quality images, similarly for expert vs. trainee. Mean RVED volume was 94.0±30.6 ml/m2, and mean RVEF was 37.3±9.6%. Reproducibility of 2DE parameters obtained by the expert vs. trainee were good for the assessment of RV longitudinal function (TAPSE, RV S', and RV LS), but suboptimal for the assessment of RV size and global function (RV diameter, RV areas, and RV FAC) (Table). Reproducibility of 3DE parameters obtained by the expert vs. trainee were also good, eve