OBJECTIVES Antenatal surveillance of intrauterine growth aims to detect growth-restricted fetuses (FGR), which face increased risk of stillbirth. Improving their detection could be an effective strategy for prevention of stillbirth. The French REPERE study was conducted to estimate the association between antenatal detection of FGR and risk of stillbirth. METHODS REPERE is a case-control study performed in three French districts with a combined total of approximately 30 000 births annually. Cases were singleton small-for-gestational-age (SGA) stillbirths ≥ 24 weeks' gestation and without severe congenital anomaly, between 2012 and 2014, identified using a population-based stillbirth registry; controls were live births fulfilling the same inclusion criteria over a 9-week period from 7 April to 8 June 2014. Data were extracted by trained investigators from medical records and ultrasound reports. SGA was defined as birth weight < 10th percentile of French customized standards. FGR was defined by the presence of at least one of seven predefined parameters (suspected FGR mentioned in medical records or in ultrasound report, suspected faltering growth mentioned in an ultrasound report, documented abdominal circumference or estimated fetal weight < 10th percentile, referral for additional ultrasound examination to monitor growth or abnormal umbilical artery Doppler). We used logistic regression to estimate crude and adjusted odds ratios (ORs) for the association between detection of FGR and risk of stillbirth. Included covariables were parity, maternal medical history, vascular complications during pregnancy and birth-weight percentile, which are known to be associated with risk of detection of FGR and of stillbirth. RESULTS During the study period, there were 92 182 births ≥ 22 weeks' gestation, including 669 stillbirths, of which 79 were singleton SGA stillbirths ≥ 24 weeks and without severe congenital anomaly. Of these cases, 44.3% (35/79) had FGR detected, compared with a detection rate of 36.2% in controls (154/426). The crude OR expressing the association between detection of FGR and risk of stillbirth was 1.4 (95% CI, 0.9-2.3) and the OR adjusted for parity, presence of risk factors for FGR, presence of vascular disorder and birth-weight percentile was 0.6 (95% CI, 0.3-1.0). Among deliveries ≥ 28 weeks, detection rates were 38.3% vs 36.0% for cases and controls, with an adjusted OR of 0.5 (95% CI, 0.2-1.0). CONCLUSION Antenatal detection of FGR was protective against stillbirth, but over 40% of stillbirths among SGA fetuses occurred despite detection of FGR, pointing to the need to improve management following detection. Copyright © 2019 ISUOG. Published by John Wiley & Sons Ltd.