The use of Doppler studies of the uterine arteries in the prediction of pre-eclampsia and intrauterine growth retardation has had mixed success. The introduction of color Doppler imaging and the use of the "notch' to define an abnormal waveform have helped to improve the predictive value of uterine artery Doppler screening. The aim of this study was to evaluate the use of uterine artery Doppler in a group of women of mixed race and parity. This study was a prospective, cross-sectional analysis of 1326 unselected women who were screened with continuous wave uterine Doppler at 19-21 weeks, as part of a fetal anomaly/dating scan. A total of 214 women with abnormal uterine artery waveforms (notching) were referred for assessment at 24 weeks; 191 attended and had color Doppler imaging/pulsed Doppler studies of both uterine arteries. Data from 185 pregnancies were suitable for analysis. There were abnormal uterine Doppler findings (uni- or bilateral notching) in 110 patients at 24 weeks; 48 had bilateral notching. The sensitivity of notching for the prediction of proteinuric pregnancy-induced hypertension (PPIH) was similar in primiparas (76.9%), multiparas (77.7%), African-Caribbean women (82.6%) and Caucasian women (71.4%). The sensitivity of bilateral notching for the prediction of PPIH requiring delivery before 34 weeks was 81.2%, and 57.6% for babies small for gestational age (SGA), with positive predictive values of 27% (PPIH), 31.2% (SGA) and 37.5% (any complication). Patients with persistent bilateral notching are particularly at risk of developing PPIH or delivering an SGA baby before 34 weeks' gestation; they warrant increased surveillance, and may be a group that could benefit from prophylactic therapies.