In 1997 we started a collaboration among three groups, combining our experience with Doppler examination of the human fetus, blood flow studies on fetal lamb, and mathematical modeling of human circulation. In preliminary investigations on fetal lambs, the same Doppler method designed for the human fetus was used to measure venous blood flow in the umbilical veins of seven fetal lambs. Doppler measurements and diffusion technique groups for umbilical venous flow were 210.8+/-18.8 and 205.7+/-38.5 ml/min/kg, respectively (p = 0.881). In human pregnancy the interobserver variabilities for the vein diameter, mean velocity, and absolute umbilical venous blood were 2.9%, 7.9%, and 12.7%, respectively. A cross-sectional study allowed us to establish normal reference values. Venous blood flow/kg of estimated fetal weight showed a nonsignificant linear reduction with gestational age, from 128.7 ml/min/kg at 20 weeks to 104.2 ml/min/kg at 38 weeks. In a series of 37 growth-restricted fetuses, the UV flow per kilogram was significantly lower in the more severe growth-restricted fetuses (abdominal circumference below the second percentile and abnormal umbilical arterial p.i.) than in normal comparable fetuses (p < 0.001). In a series of 140 normal fetuses, we calculated that the absolute blood flow rate in the ductus venosus (DV) increases significantly with advancing gestational age from 20 to 38 weeks of gestation (from 23.2+/-9.6 ml/min to 43.5+/-21.5 ml/min). This means that the percentage of umbilical blood flow shunted through the DV decreases significantly during gestation (from 50% at midgestation to 20% at 38 weeks). In a series of 45 growthrestricted fetuses, delivered because of nonreactive fetal heart rate (group 2) and for other reasons but still with a normal heart rate pattern (group 1), we measured the ductal inlet diameter. In these fetuses, the diameters at the ductal isthmus, normalized for the dimension of the abdominal circumference (inlet diameter/abdominal circumference), were significantly larger (group 1 = 6.8+/-2.3; group 29.4+/-2.8 ) than in the control group (6.1+/-0.3). This means that in this subset of fetuses the amount of blood shunted can be increased as a compensatory mechanism.