The aim of this study was to evaluate the usefulness of umbilical artery and middle cerebral artery PI and the ratio of these values as predictors of fetal growth-retardation and adverse perinatal outcome. In 100 normal fetuses and 105 intrauterine growth-retarded fetuses Doppler velocity waveforms were recorded from the umbilical artery and middle cerebral artery, and we calculated the incidence of low birth weight, gestational age at delivery, fetal distress and umbilical artery blood gas. The results were as follows: 1. There were moderate correlations between umbilical PI and birth weight (r = -0.43, p < 0.0001) and between middle cerebral artery PI and birth weight (r = 0.37, p < 0.0001) and a close correlation between cerebral-umbilical PI ratio and birth weight (r = 0.71, p < 0.0001). 2. There was a significant association between the cerebral-umbilical PI ratio and the HC/AC ratio (r = -0.46, p < 0.0001). 3. The cerebral-umbilical PI ratio was the best predictor of birth weight, fetal distress and neonatal mortality compared with umbilical artery PI and middle cerebral artery PI. In growth retarded fetuses, high umbilical artery PI and low middle cerebral artery PI, the so called "brain sparing effect" was demonstrated. The cerebral-umbilical PI ratio provided a better predictor of IUGR and adverse perinatal outcome than umbilical artery PI and middle cerebral artery PI alone.