### Abstract

#### Objectives

To determine the accuracy of established ultrasound dating formulae when used at 12–14 weeks of gestation.

#### Methods

One-hundred and sixty-seven singleton pregnancies conceived after *in-vitro* fertilization (IVF) underwent a dating scan at 12–14 weeks of gestation. Gestational age at the dating scan was calculated by adding 14 days to the number of days between the date of oocyte retrieval and the date of the ultrasound scan. Gestational age according to oocyte retrieval was regarded as the true gestational age. True gestational age was compared to gestational age calculated on the basis of 21 dating formulae based on fetal crown–rump length (CRL) measurements and to three dating formulae based on fetal biparietal diameter (BPD) measurements. In a previous study the three BPD formulae tested here had been shown to be superior to four other BPD formulae when used at 12–14 weeks of gestation. The mean of the differences between estimated and true gestational age and their standard deviation (SD) were calculated for each formula. The SD of the differences was assumed to reflect random measurement error. Systematic measurement error was assumed to exist if zero lay outside the mean difference ± 2SE (SE: standard error of the mean).

#### Results

The three best CRL formulae were associated with mean (non-systematic) measurement errors of −0.0, −0.1 and −0.3 days, and the SD of the measurement errors of these formulae varied from 2.37 to 2.45. All but two of the remaining CRL formulae were associated with systematic over- or under-estimation of gestational age, and the SDs of their measurement error varied between 2.25 and 4.86 days. Dating formulae using BPD systematically underestimated gestational age by −0.4 to −0.7 days, and the SDs of their measurement errors varied from 1.86 to 2.09.

#### Conclusions

We have identified three BPD formulae that are suitable for dating at 12–14 weeks of gestation. They are superior to all 21 CRL formulae tested here, because their random measurement errors were much smaller than those of the three best CRL formulae. The small systematic negative measurement errors associated with the BPD formulae are likely to be clinically unimportant. Copyright © 2005 ISUOG. Published by John Wiley & Sons, Ltd.