In previous reports that evaluated pulsed Doppler transmitral filling, the sampling site has varied; we examined the effect of the sample volume location on Doppler measurements. Studied were 97 patients: 58 with normal echocardiograms, 20 with mitral regurgitation, and 19 with miscellaneous cardiac diseases. Transmitral filling was recorded at the mitral anulus and at the left atrial and left ventricular (LV) sides of the mitral tips. As the sample volume was moved from the mitral anulus to the LV side, the peak velocity and time-velocity integral of early diastole increased (40.6 +/- 13.8 versus 59.0 +/- 19.0 cm/sec, 5.26 +/- 1.65 versus 8.35 +/- 2.37 cm; p less than 0.001) as did those of late diastole (48.7 +/- 11.5 versus 57.5 +/- 17.0 cm/sec, 3.48 +/- 0.97 versus 4.59 +/- 1.39 cm; p less than 0.001). The late-to-early diastolic peak velocity and time-velocity integral ratios and the late-to-total diastolic time-velocity integral ratio decreased (1.33 +/- 0.51 versus 1.06 +/- 0.41, 0.71 +/- 0.24 versus 0.58 +/- 0.19, 0.40 +/- 0.09 versus 0.36 +/- 0.08; p less than 0.001). The dependency of these indices on the sampling site was in the same degree in all three groups. However, the peak filling rate normalized to mitral stroke volume (4.71 +/- 1.43 versus 4.63 +/- 1.32 l/sec; p = NS) was not influenced by the sample volume location; thus this parameter may be more reliable for assessing LV filling.