Novel two-dimensional method for assessment of right ventricular function: validation study and prognostic value
D Stahie, C Grigore, B Spataru, A Paduraru, D Vasile, C Simionica, L Balan, A Nicula, A Velcea, R Dragoi Galrinho, H Septar, D Mihalcea, S Mihaila Baldea, D VinereanuAbstract
Background
Assessment of right ventricular (RV) function using 2D-echocardiography (2DE) remains challenging because of the complex RV geometry. Most conventional 2D parameters do not reflect global RV function (1,2). Conversely, 3D-echocardiography (3DE) (3) and cardiac magnetic resonance (CMR) (4) provide more accurate assessment of RV function, by measuring true RV volumes and ejection fraction (RVEF). These parameters were previously validated and showed prognostic value in cardiac diseases, however, their costs and limited availability impact implementation into routine clinical use.
Purpose
To validate a novel and fast 2D-derived method for estimating global RV function, and to assess its prognostic value in patients with heart failure with reduced or mildly reduced ejection fraction (HFrEF / HFmrEF).
Methods
86 patients hospitalized for HFrEF or HFmrEF underwent comprehensive echocardiography, including 2D RV-focused imaging and 3D volume acquisitions; 20 patients also underwent CMR. RV volumes and RVEF were calculated by both 3DE and CMR. By 2DE, a RV fractional volume index (RVFVI) was calculated from end-diastolic and end-systolic geometric volumes (RVEDV and RVESV), obtained from the RV-focused apical view, as follows: RVFVI = (RVEDV − RVESV) / RVEDV × 100. 2D RV volumes were derived using a simplified semi-cone subtraction model, whereby two geometric volumes (X and Y) were obtained by Simpson’s method of disks, and final RV volume was calculated as their difference (X − Y) (Figure 1). Agreement between 2DE RVFVI and 3DE RVEF was assessed using intraclass correlation coefficients (ICC) and Bland–Altman analysis; agreement between 2DE RVFVI and CMR RVEF was evaluated in the subset of 20 patients. Intra- and interobserver reproducibility were assessed in 20 patients. Primary endpoint was all-cause mortality, during a follow-up of 28 months.
Results
2DE RVFVI showed excellent agreement with 3DE RVEF (ICC=0.96, Figure 2), while Bland–Altman showed also good limits of agreement (±8.5%), with no evidence of systematic bias. Agreement was also good for 3DE RVESV (−30 to +25 mL), whereas limits of agreement were wider for 3DE RVEDV. In the CMR subgroup, 2DE RVFVI showed good agreement with CMR-derived RVEF (ICC=0.85). Intra- and inter- observer reproducibility was excellent (ICC 0.99 and 0.92, respectively). By univariate Cox regression analysis, 2DE RVFVI, as well as 3DE RVEF, were associated with mortality (both p<0.05). 2DE RVFVI remained independently associated with mortality after adjustment for tricuspid annular plane systolic excursion (TAPSE).
Conclusions
2DE RVFVI is a fast and reproducible 2D-derived parameter for assessment of global RV function, showing good agreement with both 3DE and CMR. RVFVI provides clinically relevant prognostic information for all-cause mortality in patients with HFrEF or HFmrEF, remaining independently associated with outcomes after adjustment for TAPSE.Figure 1.RVFVI derivation methodFigure 2.2DE RVFVI vs 3DE RVEF