Objective. To compare perinatal outcomes of late small-for-gestational age (SGA, birthweight <10th percentile) infants, according to antenatal recognition. Design. Retrospective study. Setting. Tertiary referral center. Population. All singleton pregnancies with SGA (birthweight <10th percentile) infants born ≥36 weeks’ gestation from January 2007 to April 2009. Methods: Chart review of 771 pregnancies. SGA infants identified prior to delivery (group A) were compared with those not identified (group B). Main outcome measures. Mode of delivery, perinatal complications, admission to neonatal intensive care unit (NICU). In group A, receiver operating characteristic (ROC) analysis and area under the curve (AUC) of antenatal tests to predict NICU admission were calculated. Results. In 17% of infants, SGA was recognized before birth (group A), whereas in 83% it was recognized only at birth (group B). Infants with the most severe degree of SGA (birthweight <3rd percentile) were more frequently diagnosed antenatally (30%). Admission to NICU (14 vs. 3%, p < 0.001) and cesarean delivery (42 vs. 26%; p < 0.001) were more likely in group A. Adverse outcomes related to hypoxia were all observed in group B, but the difference was not significant. In antenatally detected fetuses a combination of fetal abdominal circumference, umbilical artery pulsatility index and middle cerebral artery pulsatility index z-scores was useful for prediction of NICU admission (AUC = 0.94). Conclusion. Antenatal recognition of late SGA may improve perinatal outcomes. However, admission to neonatal intensive care and cesarean delivery were more frequent among SGA infants recognized antenatally.