Composite biometry versus head circumference alone in second trimester dating of twin gestations: A retrospective cohort study

In the second trimester, U.S. societal guidelines recommend using a composite of the fetal biparietal diameter, head circumference, abdominal circumference, and femur length to establish the due date in a singleton gestation. In twin pregnancies, evidence supports using either the head circumference or composite biometry of the larger twin. Our objective was to compare the accuracy of these two approaches in determining the true gestational age. We conducted a retrospective cohort study of dichorionic, diamniotic gestations conceived with assisted reproductive technologies (ART) between 2006-2016, excluding fetuses with major anomalies that could impact accurate dating. The earliest second trimester ultrasound study for each subject was reviewed and the gestational age calculated using head circumference alone and composite biometry based on Hadlock's formulas. Paired sample statistics were used to calculate the mean difference in days from the observed versus true gestational age to assess accuracy of each strategy and represented graphically using Bland Altman plots. A total of 179 subjects with a second trimester ultrasound were identified over the 10-year period that met criteria. Ultrasounds were performed at a mean gestational age of 19.9 days (range 14.43-26.14). Use of head circumference alone resulted in a mean absolute difference of 3.11 days from the true gestational age (standard deviation (SD) 4.51 standard error mean (SEM) 0.34). Composite biometry resulted in a mean absolute difference of 1.74 days (SD 3.60 SEM 0.27). The difference of approximately 1-2 days was significant (p<0.001). In our twin population, use of composite biometry of the larger twin more accurately approximates gestational age compared to head circumference alone, though the difference between the two approaches is modest.