To determine which ultrasound measurement for predicted fetal macrosomia most accurately predicts adverse delivery and neonatal outcomes. Four biomedical databases searched for studies published after 1966. Randomised trials or observational studies of women with singleton pregnancies, resulting in a term birth who have undergone an index test of interest measured and recorded as predicted fetal macrosomia ≥28 weeks. Adverse outcomes of interest included shoulder dystocia, brachial plexus injury (BPI) and Caesarean section. Twenty-five observational studies (13,285 participants) were included. For BPI, the only significant positive association was found for Abdominal Circumference (AC) to Head Circumference (HC) difference > 50 mm (OR 7.2, 95 % CI 1.8–29). Shoulder dystocia was significantly associated with abdominal diameter (AD) minus biparietal diameter (BPD) ≥ 2.6 cm (OR 4.2, 95 % CI 2.3–7.5, PPV 11 %) and AC > 90th centile (OR 2.3, 95 % CI 1.3–4.0, PPV 8.6 %) and an estimated fetal weight (EFW) > 4000 g (OR 2.1 95 %CI 1.0–4.1, PPV 7.2 %). Estimated fetal weight is the most widely used ultrasound marker to predict fetal macrosomia in the UK. This study suggests other markers have a higher positive predictive value for adverse outcomes associated with fetal macrosomia.