Combined influence of ventricular loading and relaxation on the transmitral flow velocity profile in dogs measured by Doppler echocardiography

The relation of the Doppler transmitral flow velocity profile to left ventricular loading conditions and diastolic properties remains poorly described. We studied seven adult mongrel dogs with an open-chest right heart bypass model in which left atrial pressure, representing preload, was varied by controlling blood flow into the pulmonary artery and left ventricular systolic pressure, representing afterload, was controlled independently by pumping blood into or from the femoral arteries. Heart rate was kept constant by crushing the sinus node and pacing the right atrium. Mitral inflow velocity profiles were measured by pulsed-wave Doppler echocardiography at multiple left atrial and left ventricular systolic pressures. In individual dogs, the peak E-wave velocity increased linearly with increasing left atrial V-wave pressure at constant left ventricular systolic pressure and decreased with increasing left ventricular systolic pressure at constant left atrial pressure. Stepwise multiple linear regression analysis of data pooled from all experimental stages in all dogs identified left atrial V-wave pressure, the time constant of relaxation (T(L)), and left ventricular systolic pressure, in order of decreasing significance, as predictors of the peak E-wave velocity (n = 82, multiple r = 0.87, p < 0.0001). Multivariate analysis with the same three factors in individual dogs yielded higher r values (mean r = 0.89; range, 0.85-0.97), suggesting the presence of important interdog differences that were not accounted for by these three factors alone. When the values of codeterminant hemodynamic factors were kept within narrower limits, correlations between peak E-wave velocity and left atrial V-wave pressure (n = 35, multiple r = 0.83, p < 0.0001), T(L) (n = 76, multiple r = -0.54, p < 0.0001) and left ventricular systolic pressure (n = 20, multiple r = -0.59, p < 0.005) improved substantially. In the pooled data, the relation of the peak E-wave velocity to left atrial V-wave pressure was shifted downward by an increase in T(L) (reduced relaxation rate), and the relation of the peak E-wave velocity to T(L) was shifted upward by an increase in left atrial V-wave pressure. Multivariate analysis also selected left atrial V-wave pressure and T(L) as the two most significant correlates of the velocity-time integral and deceleration rate of the E wave. Those Doppler parameters that represented proportions rather than absolute values (the peak E-wave velocity:peak A-wave velocity ratio, the A-wave velocity-time integral:total velocity-time integral ratio, and the first third velocity-time integral:total velocity-time integral ratio) correlated with T(L) but were relatively independent of left atrial V-wave pressure. Thus, the mitral inflow profile is determined in a complex manner by multiple factors, which include left atrial pressure, relaxation rate, and left ventricular systolic pressure. Simultaneous variations in each of these factors often obscure simple relations between individual Doppler parameters of interest and a single hemodynamic factor. These results suggest that Doppler filling parameters must be interpreted cautiously when used as indexes of diastolic function.