ABSTRACTObjectiveTo compare the accuracy of four published reference standards for the umbilical artery pulsatility index (UA‐PI) in predicting small‐for‐gestational age (SGA), adverse neonatal outcomes and obstetric complications in pregnancies at risk for fetal growth restriction.MethodsThis was a secondary analysis of a prospective study of singleton pregnancies that underwent fetal growth assessment by ultrasound between 26 and 36 weeks' gestation. Pregnancies with estimated fetal weight or abdominal circumference < 20th percentile with UA‐PI measurements available were included. We excluded fetuses with chromosomal anomaly or congenital malformation and those without delivery information. The predictive ability of UA‐PI > 95th percentile according to the reference standards of Acharya et al., the INTERGROWTH‐21st Project, the Fetal Medicine Foundation and Parra‐Cordero et al. for SGA, a composite of adverse neonatal outcomes and a composite of obstetric complications was compared using the area under the receiver‐operating‐characteristics curve (AUC). Sensitivity, specificity and positive and negative predictive values were calculated.ResultsOf the 1054 pregnancies that underwent fetal growth evaluation by ultrasound, 207 were included in our analysis. SGA, adverse neonatal outcomes and obstetric complications were diagnosed in 94 (45.4%), 50 (24.2%) and 69 (33.3%) cases, respectively. All reference standards had similar and statistically significant but poor predictive accuracy for SGA (AUC of 0.55 to 0.56), adverse neonatal outcomes (AUC of 0.57 to 0.60) and obstetric complications (AUC of 0.55 for all).ConclusionsThe reference standards for UA‐PI evaluated herein have poor predictive ability for SGA, adverse neonatal outcomes and obstetric complications. At present, no particular UA‐PI reference standard can be recommended over others. Larger trials are needed to answer this research question. © 2024 International Society of Ultrasound in Obstetrics and Gynecology.