The aim of our study was to obtain measurements of the umbilical artery pulsatility index (PI) in pregnancies before invasive procedures for prenatal diagnosis, in order to investigate its potential prognostic value in predicting trisomy 18. We performed a prospective study including 1785 consecutive women from 10 to 18 weeks with singleton pregnancies undergoing chorionic villus sampling (n = 559) or genetic amniocentesis (n = 1226) in our unit. Doppler measurements were performed transvaginally (tenth to 13th week of gestation) or transabdominally (14th to 18th week of gestation) immediately before the invasive procedure. In 7 out of 10 fetuses subsequently diagnosed as trisomy 18, the PI was above the 95th centile, providing a detection rate of 70 per cent, a specificity of 95.1 per cent, a positive predictive value of 7.7 per cent, and a negative predictive value of 99.8 per cent. When the 90th percentile was assayed as a cut-off, the efficacy of PI as a marker of trisomy 18 yielded a sensitivity of 90 per cent and a specificity of 90.4 per cent, with a positive predictive value of 5.2 per cent and a negative predictive value of 99.9 per cent. We suggest that although the use of a single PI measurement for screening purposes needs to be confirmed by further investigation, trisomy 18 fetuses show an abnormal increase in umbilical PI in the first half of pregnancy, and its relation to the early onset of fetal growth retardation needs to be further explored.