Systematic correlation of continuous-wave Doppler and hemodynamic measurements in patients with aortic stenosis

Abstract

The purpose of this study was to compare estimates of pressure gradients obtained from continuous-wave (CW) Doppler recordings with direct pressure measurements derived from cardiac catheterization in patients with aortic stenosis. Forty patients who underwent cardiac catheterization for evaluation of aortic stenosis were prospectively studied with CW Doppler spectral recordings of the aortic valve prior to catheterization. Thirty-three patients underwent a second Doppler examination simultaneously with pressure recordings in the catheterization laboratory. Nineteen of the patients had catheterization pressures measured using high-fidelity, micromanometer-tip catheters. Doppler and pressure tracings were digitized using a microprocessor-based computer with a software program which allowed for calculation of maximal instantaneous, mean, and peak-to-peak gradients, plus ejection and acceleration times. Maximal instantaneous gradient by CW Doppler showed an excellent correlation with maximal instantaneous catheterization gradient ( r = 0.93, SEE = 9 mm Hg). The correlation of maximal instantaneous Doppler gradient with peak-to-peak catheterization gradient was also linear ( r = 0.85, SEE = 12 mm Hg), but there was a consistent overestimation of peak-to-peak gradient in 38 of 40 cases (mean = 17 mm Hg). Mean gradient as calculated by the two techniques correlated best of all measurements performed ( r = 0.95, SEE = 6 mm Hg). When patients were grouped into subsets of mild (0 to 25 mm Hg), moderate (25 to 50 mm Hg), and severe (>50 mm Hg) levels of stenosis, the correlation of maximal instantaneous Doppler and peak-to-peak catheterization gradients were r = 0.22, 0.44, and 0.77, respectively. Doppler and catheterization maximal instantantous gradients correlated better in the 19 patients who had micromanometer recordings ( r = 0.96) than in the 21 patients in whom fluid-filled systems were used ( r = 0.89). Surprisingly, the correlation was also better for Doppler studies performed prior to, rather than simultaneous with cardiac catheterization ( r = 0.93 and 0.85, respectively). The data indicate that Doppler spectral signals accurately reflect instantaneous catheterization pressure gradients, and that mean systolic pressure can be calculated using planimetry methods. However, the peak-to-peak catheterization gradient is consistently overestimated by Doppler maximal instantaneous gradient, especially in mild and moderate degrees of aortic stenosis. In addition, certain technical factors such as types of catheter pressure recording systems and the timing of Doppler studies in relation to catheterization, may be important in the accuracy of CW Doppler predictions of hemodynamic parameters.