This study was undertaken to compare the accuracies of the two-dimensional echocardiographic (2DE) and Doppler pressure half-time methods for the noninvasive estimation of cardiac catheterization measurements of mitral valve area in patients with pure mitral stenosis both with and without a previous commissurotomy. Data were retrospectively obtained from 74 consecutive patients who underwent cardiac catheterization within a 30 month period for evaluation of mitral stenosis, and who had two-dimensional echocardiograms performed before catheterization. Six patients (8.1%) had technically inadequate 2DE images and their data were excluded from analysis. Two of these patients had undergone commissurotomy, while the remaining four had not. Continuous-wave Doppler echocardiographic examinations were attempted in 45 patients and adequate measurements of pressure halftimes were obtained in all patients studied. Mitral valve area by two-dimensional echocardiography was measured as the planimetered area along the inner border of the smallest mitral orifice visualized during short-axis scanning, while pessure half-time was calculated as the interval between the peak transmitral velocity and velocity/V2 as measured from the envelope of the Doppler spectral signal. Calculations from catheterization represented the minimal valve area at rest as derived from the Gorlin formula with the use of pressure gradients and thermodilution measurements of cardiac output. Thirtyseven of the patients had had a previous mitral commissurotomy a mean of 11.2 ± 5.4 years before, while the remaining 37 patients were previously unoperated. Mean valve area as determined at catheterization for the total group of patients ranged from 0.37 to 2.30 cm2 (mean = 1.08 ± 0.42 cm2). Linear regression analysis of data from the group of 33 previously unoperated patients revealed a good correlation between 2DE and catheterization measurements of mitral valve area (r = .83, y = 0.79x + 0.29, SEE = 0.26 cm2). Similarly, the correlation between Doppler measurements of mitral valve area were also good (r = .85, y = 0.84x + 0.17, SEE = 0.22 cm2). However, in the group of 35 patients who had undergone commissurotomy, the Doppler pressure half-time correlated much better with catheterization measurements (r = .90, y = 0.63x + 0.39, SEE = 0.14 cm2) than with 2DE estimates (r = .5 8, y = 0.47x + 0. 6 1, SEE = 0. 28 cm2) Reproducibility was similar for the two noninvasive methods, with a mean error of 0.14 cm2 for 2DE planimetry, and of 0.15 cm2 for Doppler pressure half-time. Thus, our data show that both 2DE and Doppler pressure half-time methods provide accurate noninvasive estimates of mitral orifice area in patients who have not undergone surgery. However, the Doppler pressure half-time is superior to two-dimensional echocardiography in estimating mitral valve area in patients who have undergone commissurotomy.