From novice to expert: the reproducibility of 3D echocardiographic right ventricular assessment in heart failure patients

Background Right ventricular (RV) function is a key prognostic factor in patients with heart failure with mildly reduced (HFmrEF) or reduced ejection fraction (HFrEF). While two-dimensional echocardiography (2DE) is used due to its availability, three-dimensional echocardiography (3DE) provides more reproducible measurements, though its use is limited by training requirements. Objective To assess whether cardiologists experienced in 2DE with limited 3DE exposure can obtain feasible and reproducible 3DE measurements of RV size and function after a short training in patients with HFmrEF/HFrEF. Methods 161 patients hospitalized for decompensated HFmrEF/HFrEF (mean age 58 ± 17 years, 71% males, 3D LVEF 35 ± 10%) were analyzed in the study using 2DE and 3DE assessments. Measurements were performed by an Expert in 2DE and 3DE, and by a Beginner with experience in 2DE but only three months of practical training in 3DE. Measurements were taken at baseline (T0) and after three months of practical training in 3DE (T1) to assess intra- and inter-observer reproducibility. Results The study demonstrated high intra-observer reproducibility for 2DE parameters by the Beginner with 95% ICCs of: 0.98 (0.98–0.99) for RV diameter, 0.97 (0.94–0.98) for TAPSE, 0.92 (0.90–0.99) for RVFAC, 0.96 (0.95–0.98) for S’, and 0.98 (0.97–0.99) for RVFWS. Conversely, there was a slightly lower inter-observer reproducibility compared to the Expert for the same 2D parameters, with ICCs of: 0.81 (0.71–0.87) for RV diameter, 0.91 (0.88–0.94) for TAPSE, 0.86 (0.81–0.90) for RVFAC, 0.90 (0.88–0.93) for S’, and 0.93 (0.85–0.96) for RVFWS, respectively. The Beginner's intra-observer reproducibility for 3DE parameters was good at baseline, after short theoretical training in 3DE, with ICCs of: 0.87 (0.83–0.91) for RVEDV, 0.85 (0.79–0.89) for RVESV, and 0.90 (0.87–0.93) for RVEF, respectively, and improved significantly after 3 months of practice in 3DE, with ICCs of: 0.96 (0.92–0.97) for RVEDV, 0.95 (0.94–0.98) for RVESV, and 0.95 (0.91–0.97) for RVEF. Bland-Altman analysis showed no systematic bias between the Expert and Beginner for both 2DE and 3DE measurements, confirming the robustness of 3DE across different experience levels. Conclusions After brief training, 2DE-proficient cardiologists can perform accurate and reproducible 3DE measurements of RV function, supporting broader clinical use of 3DE in heart failure assessment.