Noninvasive assessment of left atrial maximum dP/dt by a combination of transmitral and pulmonary venous flow

OBJECTIVES The study assessed whether hemodynamic parameters of left atrial (LA) systolic function could be estimated noninvasively using Doppler echocardiography. BACKGROUND Left atrial systolic function is an important aspect of cardiac function. Doppler echocardiography can measure changes in LA volume, but has not been shown to relate to hemodynamic parameters such as the maximal value of the first derivative of the pressure (LA dP/dt(max)). METHODS Eighteen patients in sinus rhythm were studied immediately before and after open heart surgery using simultaneous LA pressure measurements and intraoperative transesophageal echocardiography. Left atrial pressure was measured with a micromanometer catheter, and LA dP/dt(max) during atrial contraction was obtained. Transmitral and pulmonary venous flow were recorded by pulsed Doppler echocardiography. Peak velocity, and mean acceleration and deceleration, and the time-velocity integral of each flow during atrial contraction was measured. The initial eight patients served as the study group to derive a multilinear regression equation to estimate LA dP/dt(max) from Doppler parameters, and the latter 10 patients served as the test group to validate the equation. A previously validated numeric model was used to confirm these results. RESULTS In the study group, LA dP/dt(max) showed a linear relation with LA pressure before atrial contraction (r = 0.80, p < 0.005), confirming the presence of the Frank-Starling mechanism in the LA. Among transmitral flow parameters, mean acceleration showed the strongest correlation with LA dP/dt(max) (r = 0.78, p < 0.001). Among pulmonary Venous flow parameters, no single parameter was sufficient to estimate LA dP/dt(max) with an r(2) > 0.30. By stepwise and multiple linear regression analysis, LA dP/dt(max) was best described as follows: LA dP/dt(max) = 0.1 M-AC + 1.8 P-V - 4.1; r = 0.88, p < 0.0001, where M-AC is the mean acceleration of transmitral flow and P-V is the peak velocity of pulmonary venous flow during atrial contraction. This equation was tested in the latter 10 patients of the test group. Predicted and measured LA dP/dt(max) correlated well (r = 0.90, p < 0.0001). Numerical simulation verified that this relationship held across a wide range of atrial elastance, Ventricular relaxation and systolic function, with LA dP/dt(max) predicted by the above equation with r = 0.94. CONCLUSIONS A combination of transmitral and pulmonary venous flow parameters can provide a hemodynamic assessment of LA systolic function. (J Am Coil Cardiol 1999;34:795-801) (C) 1999 by the American College of Cardiology.