Validation of noninvasive indices of right ventricular diastolic function. Simultaneous echocardiography and pressure-volume catheterization studies.

Abstract

Background: The reliability of the recommended echocardiographic methods for assessing RV diastolic function has been questioned. We aimed to validate noninvasive indices of RV diastolic function, derived from tricuspid Doppler and myocardial deformation metrics, against intrinsic diastolic chamber properties and filling pressures.

Methods: We obtained simultaneous high-fidelity pressure-volume loops and echocardiographic data in separate animal and clinical settings: (1) a porcine model of acute hemodynamic interventions (n = 13), and (2) patients with Fallot tetralogy and pulmonary hypertension (n = 9). These designs allow for within- and between-subject validation. From the PV loops data, we obtained the reference values of RV stiffness (S₊), elastic recoil (S_-) and relaxation (τ) constants, as well as the contribution of passive properties to instantaneous diastolic pressures.

Results: In the animal setting, only the tricuspid E/A ratio and e' velocity weakly correlated with S₊ (R_rm:0.36 and 0.28 respectively, p < 0.01 for both). In the clinical group, no correlation was found between the echocardiographic indices and the intrinsic diastolic properties. Isovolumic relaxation time and early diastolic global strain-rate (GSR) correlated with mean right atrial pressure (RAP) (Spearman r: -0.73 and 0.85, respectively, p < 0.05 for both). E/e' and E/GSR ratio were not associated with RAP. Tricuspid e' and GSR negatively correlated with passive pressure component (only due to) at valve opening (R_rm -0.27 and - 0.33, respectively, p < 0.01 for both).

Conclusions: Recommended echocardiographic indices of RV diastolic function do not reflect intrinsic RV diastolic properties. Therefore, the application of these indices for inferring RV diastolic function and filling pressures is limited.