Precision and accuracy of Doppler flow measurements.In vitroandin vivostudy of the applicability of the method in human fetuses

The precision of the Doppler method for quantitative blood flow measurement in the fetal descending thoracic aorta and in the umbilical vein and for estimation of the Pulsatility Index from the velocity curve from fetal aorta was tested in vivo by examination of six pregnant women eight times. Two investigators examined each patient twice in random order upon two successive days. The diameter of the vessel was measured using planimetry on a magnified time-motion image of the diameter variations during the heart cycle, while the angle between the ultrasound Doppler beam and the vessel of interest was measured on the hard copy image. No systematic variation was found between observers, days, repeated observations or repeated readings of curves and images. The mean coefficient of variations was 5.6% for the quantitative flow per kilogram estimated fetal weight measured in the fetal descending aorta, 6.8% for the quantitative flow per kilogram estimated fetal weight measured in the umbilical vein and 9.8% for the Pulsatility Index. When the diameter of aorta was calculated as the mean of the maximal and the minimal diameter measured on the hard copy image, the mean coefficient of variation for the flow increased to 9.4%. In vitro tests of the Doppler instrument and the real-time scanner revealed a systematic overestimation of Doppler measured flow of only 4.4% compared with the true flow, and a real-time scanner underestimation of vessel diameter of only 1.1%.