The authors advocate a semiquantitative evaluation of mitral regurgitation (MR), based on the extent of the left intra-atrial systolic jet. Their study was performed in a series of 63 consecutive patients explored by: (1) catheterization with left ventriculography on two planes for either ischaemic heart disease (24 cases) or cardiac valve disease (39 cases including 2 with mitral valve prosthesis), and (2) pulsed Doppler ultrasound combined with two-dimensional echocardiography for evaluation of MR carried out 24 or 48 hours prior to catheterization. Five grades of MR (0 to IV) were determined by ventriculography according to the time and degree of left atrial opacification. Similarly, five grades of MR (0 to IV) were determined by Doppler ultrasound, considering only the maximum distance from the mitral annulus plane. This distance was evaluated on at least 3 projections: apical for the 4 and 2 cavities, parasternal for the greater axis. The Doppler method correlated fairly well with the angiography, whether or not grade 0 was included (r = 0.87 and r = 0.72 respectively). However, some major MR may be underestimated; thus, out of 13 RM of angiographic grade III or IV, three (23 p. 100) were graded II by the Doppler technique. Conversely, some minute or moderate MR may be overestimated: out of 33 MR of angiographic grade I or II, four (12 p. 100) were graded III or IV by the Doppler technique.(ABSTRACT TRUNCATED AT 250 WORDS)