Background
In response to the challenges of assessing fetal growth in obese women, guidelines recommend routine third trimester ultrasound scans.

Aim
The aim of this study was to assess the diagnostic performance of this routine scan in obese women (body mass index (BMI) ≥ 35 kg/m2).

Methods
A retrospective cohort study of 1008 pregnancies with maternal BMI ≥ 35 kg/m2 born after 37 weeks gestation at a Victorian hospital from 2015 to 2017. Multiple pregnancies and those affected by diabetes were excluded. Growth ultrasounds were performed between 34 + 0 and 36 + 6 weeks gestation. Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) for the detection of large for gestational age (LGA > 90%) and small for gestational age (SGA < 10%) were calculated using ultrasound estimated fetal weight (EFW) or abdominal circumference (AC) and compared with gestational age and gender‐based birthweight percentiles. Results Using EFW, sensitivity for detecting SGA at birth was 8.1% (six of 74) with a PPV of 100%. Sensitivity for detecting LGA at birth was 61.0% (119 of 195), PPV 54.8%. Sensitivity, specificity, PPV and NPV percentages were all lower using AC. Only 40% of actual birthweight percentiles (405/1008) were within ±10 percentiles of their growth ultrasound EFW percentile. Conclusion The performance of a routine third trimester ultrasound in women with BMI ≥ 35 kg/m2 suggests limited utility in helping identify aberrant fetal growth. This has important implications for the management of obese pregnant women.