Invasive Validation of Right Ventricular Stroke Volume and Tricuspid Regurgitant Volume Obtained from Three-Dimensional Echocardiography

Background: Quantifying tricuspid regurgitation (TR) severity by conventional Doppler echocardiography is challenging. Objectives: We sought to validate right ventricular (RV) stroke volumes (SV3DE) and tricuspid regurgitant volumes (RegVol3DE) measured by three-dimensional echocardiography (3DE). Then we compared the quantitative metrics used to assess TR severity obtained by 3DE and by both the conventional and the corrected PISA methods. Methods: Three-dimensional echocardiography RV and left ventricular (LV) SV were collected simultaneously with direct Fick measurements at right heart catheterization (RV SVRHC) in 45 patients (66 ± 14 years, 76% women) with no or trivial TR and in 57 consecutive patients (73 ± 13 years, 61% women) with mild to torrential (14% mild, 37% moderate, 26% severe, 16% massive, and 7% torrential) TR and mild or no mitral or aortic regurgitation. In the latter group, RHC and 3DE were combined to derive TR regurgitant volume (RegVol3DE-RHC = RV SV3DE – RV SVRHC) and compared to the 3DE volumetric RegVol (RegVol3DE = RV SV3DE – LV SV3DE). Effective regurgitant orifice area (EROA) and regurgitant fraction (RegFr) were calculated from RegVol3DE, as well as through conventional and corrected PISA methods. Results: The feasibility of RV SV3DE was 74%. Among patients with no or trivial TR, RV SV3DE and RV SVRHC showed a strong correlation (R2 = 0.916, P < .0001), with a minimal bias (3.7 mL) and reasonable precision (limits of agreement, −10 mL; 18 mL). In those with mild to torrential TR, RegVol3DE correlated with RegVol3DE-RHC (R2 = 0.918, P < .0001) and was accurate (bias = −1.3 mL, limits of agreement, –17.9 mL; 15.2 mL). The EROA, RegVol, and RegFr derived from conventional PISA (0.48 ± 0.4 cm2, 38 ± 20 mL, and 36% ± 18%, respectively) were significantly smaller (P < .05) than those obtained from corrected PISA (0.61 ± 0.5 cm2, 48 ± 24 mL, and 46% ± 24 %, respectively) and volumetric 3DE (0.62 ± 0.5 cm2, 46 ± 25 mL, and 42% ± 18%, respectively). Conclusions: Right ventricular SV measured by 3DE is accurate (minimal bias, acceptable imprecision) when compared to direct Fick RV SVRHC.