OBJECTIVE To investigate the value of second-trimester uterine artery Doppler in the prediction of complications resulting from uteroplacental insufficiency in low- and high-risk multiparous women. METHODS Color flow pulsed Doppler imaging of both uterine arteries at 20 weeks' gestation was performed on 628 multiparous women; 458 of them had no known risk factors and 170 had clinically identifiable high-risk factors at booking. An abnormal result was defined as bilateral notches and a mean resistance index (RI) >/= 0.55 (50th centile) or unilateral notches and a mean RI >/= 0.65 (80th centile). The main outcome measure was adverse pregnancy outcome defined as any case of pre-eclampsia, small-for-gestational age birth weight (< 5th centile), placental abruption, stillbirth or early neonatal death. RESULTS There was an adverse outcome in 30 women (6.6%) in the low-risk group and 48 (28.2%) women in the high-risk group. In the high-risk group the sensitivity to predict adverse pregnancy outcome in screen-positive women was 81.4% for a specificity of 89.0%, a positive predictive value of 71.4% and a negative predictive value of 93.4%. Normal Doppler studies in the high-risk group conferred a risk of adverse perinatal outcome of 6.6%, similar to the risk of adverse outcome in the low-risk population (6.6%). In the low-risk group the sensitivity for an adverse outcome in screen-positive women was 33.3% for a specificity of 92.8% and a positive predictive value of 24.4%. CONCLUSIONS In high-risk multiparous women, persistent bilateral notches with mean RI >/= 0.55 and unilateral notches with mean RI >/= 0.65 at 20 weeks' gestation identifies the vast majority of women who will subsequently develop complications secondary to uteroplacental insufficiency. Normal uterine artery Doppler studies in these women confers a risk of adverse outcome similar to that of women with an uncomplicated obstetric history. In low-risk women, the screening efficacy of uterine artery Doppler for adverse perinatal outcome is poor and does not justify routine screening.