Aim To assess interobserver reproducibility of the offline measurement of fetal head biometry of the 3D volume datasets and to compare them with measurement taken during the real time 2D scan.

Methods 79 3D volume datasets of fetal head were obtained at the time of the scan. 27 were normal, 52 had abnormalities. Offline analysis of them were carried out by three independent fetal medicine experts (E1, E2 and E2) using 4D view software. Measurements of the biparietal diameter (BPD), head circumference (HC), posterior horn of the lateral ventricle (Vp) and transcerebellar diameter (TCD) were compared between examiners and to 2D measurements recorded in patient record. All examiners were blinded to history and patient details.

Results The intraclass co-efficient for data analysis demonstrated a high level of measurement reproducibility between 2D and 3D E1, 3D E2 and 3D E3 with values greater than 0.9 throughout. Bland–Altman plots showed good levels of agreement between 2D and all three 3D observers. The mean percentage differences for the prespecified parameters are 0.3%, 0.7%, 1% and 1.1% for BPD, HC, Vp and TCD respectively.

Conclusion Offline measurement of fetal head biometry using 3D volume dataset is a reliable and reproducible method and correlates well with real time 2D ultrasound. Further work is required to assess the potential use of offline 3D volume dataset analysis as a teaching aid in the measurement of fetal biometry and to examine brain abnormalities.