Objective To determine the accuracy of fetal weight estimation in twin pregnancy using the Hadlock formula. A version of this formula which utilises only two measurements from each fetus will be compared to that which utilises four measurements to obtain an estimation of fetal weight.

Methods Statistics from twin pregnancies that had delivered successfully within 7 days of the last scan during 2004–2009 were analysed which included 242 twins (121 pregnancies). There were 137 fetuses who had all four measurements recorded; namely abdominal circumference (AC), Femur Length (FL), bi-parietal diameter (BPD) and head circumference (HC). There were 105 fetuses that had only two measurements recorded; namely AC and FL. The estimation of fetal weight was then compared to the birth weight (BW) of the twin immediately after birth.

Conclusion In twin pregnancy, compared to singleton pregnancy, it may be more difficult to obtain a full set of measurements to enable accurate fetal weight estimation (Secher et al)1. The use of Hadlock AC-FL-BPD-HC was more accurate compared to Hadlock AC-FL by approximately 1.2%. In Twin 1 overall the use of Hadlock AC-FL was as accurate as Hadlock BPD-HC-AC-FL and only marginally different in T2. Utilising HC and BPD will give a more accurate representation of the BW but the results will not be grossly inaccurate if only AC and FL are used. There was a tendency to overestimate the BW in both twins whether two or four measurements were used.