Abstract Funding Acknowledgements Type of funding sources: Public grant(s) – National budget only. Main funding source(s): This work was supported by the New National Excellence Programme (ÚNKP-19-3-I) of the Ministry for Innovation and Technology in Hungary, and the Artificial Intelligence Research Field Excellence Programme of the National Research, Development and Innovation Office of the Ministry of Innovation and Technology in Hungary. Background Right ventricular (RV) ejection fraction (EF) has established prognostic significance, which is independent of left ventricular (LV) EF in various cardiac diseases. However, RV EF is a cumulative result of the complex interplay between distinct mechanical components (i.e., shortening along the longitudinal, radial, and anteroposterior directions), and the prognostic value of RV motion decomposition remains to be quantified. Objective Our aim was to explore whether the assessment of longitudinal, radial, and anteroposterior motion components of the RV with 3D transthoracic echocardiography offers prognostic value in patients with left-sided heart disease. Methods Two hundred and ninety-two consecutive patients (age 59 ± 17 years, 70% male) with left-sided heart disease underwent standard clinical investigations and 3D echocardiographic examination. They were followed-up for 6.7 ± 2.2 years, and cardiac death served as the primary endpoint. LV and RV volumes and ejection fractions were quantified by the offline analysis of 3D datasets. The ReVISION method was applied to the 3D models of the RV to decompose the motion along the three orthogonal axes and to calculate longitudinal, radial, and anteroposterior EF (LEF, REF, AEF, respectively). Conventional parameters of RV systolic function (tricuspid annular plane systolic excursion [TAPSE], fractional area change [FAC]) were also assessed. Results Cardiac death occurred in 60 (21%) patients. Patients who died had lower LV EF (39 ± 16 vs. 52 ± 12%, p < 0.001), RV EF (40 ± 11 vs. 48 ± 8%, p < 0.001), and each mechanical component showed significantly lower values compared to patients alive (LEF: 13 ± 6 vs. 19 ± 6%; REF: 22 ± 7 vs. 25 ± 7%; AEF: 14 ± 6 vs. 18 ± 5%, all p < 0.001). LEF was decreased to a greater degree compared to RV EF (relative %: -30 vs. -18). In univariate Cox regression models, RV EF (Hazard Ratio [HR]: 0.928, 95% Confidence Interval [CI] 0.909 – 0.948, p < 0.001), LEF (0.855 [0.816 – 0.896], p < 0.001), REF (0.932 [0.898 – 0.967], p < 0.001), AEF (0.879 [0.841 – 0.919], p < 0.001), TAPSE (0.881 [0.841-0.923], p < 0.001), and FAC (0.955 [0.933-0.977], p < 0.001) were all found to be significant predictors of cardiac death. From all parameters that were predictive, the optimal combination of variables was identified with an automated stepwise selection algorithm. The final multivariate model included serum creatinine (1.015 [1.010 – 1.020], p < 0.001), haemoglobin concentration (0.965 [0.948 – 0.982], p < 0.001), LV EF (0.977 [0.955 – 0.999], p < 0.05), and LEF (0.899 [0.843 – 0.959], p < 0.01) as independent predictors of cardiac death. Notably, the algorithm rather selected LEF and not RV EF. Conclusions 3D echocardiography-derived measurements of RV systolic function are able to predict outcomes in patients with left-sided heart disease independently of LV function. The separate quantification of RV mechanical components can hold additional prognostic value compared to conventional echocardiographic parameters.