To assess the performance of four previously reported Doppler abnormalities of uterine artery velocity waveforms (presence of a protodiastolic notch, peak systolic over protodiastolic velocities (A:C ratio) > 2.5, peak systolic over end diastolic velocities (A:B ratio) > 90th centile, resistance index (RI) ([A-B]/A) > or = 0.58) in predicting pre-eclampsia, low birthweight and prematurity. Prospective cohort study. Tertiary care university hospital in Quebec City. 1311 nulliparous women. Evaluation of pulsed Doppler abnormalities of uterine artery velocity waveforms was carried out in 1000 and 1194 of women at 18.3 (SD 0.9) or 26.7 (SD 0.9) weeks of pregnancy. Pre-eclampsia, birthweight below the 10th centile for gestational age and spontaneous preterm birth (< 37 completed gestation weeks). Pre-eclampsia, low birthweight for gestation and prematurity occurred in 4%, 11% and 7% of the pregnancies, respectively. At 26 weeks all the abnormalities of the studied Doppler indices were significantly associated with pre-eclampsia and low birthweight for gestation as reflected by the 95% confidence intervals of the positive likelihood ratios which did not include the value 1. However, sensitivities (26% to 34%) and positive predictive values (7% to 28%) were low. No Doppler indices performed significantly better than the others. The abnormalities of the Doppler indices were not associated with spontaneous prematurity. The performance of the Doppler measurements performed at 18 weeks was poor. Uterine artery Doppler velocimetry waveform analysis does not qualify as a reliable screening test for pre-eclampsia or low birthweight for gestation in low risk pregnancies but may be useful in selected high risk populations.