A comparison of first trimester measurements for prediction of delivery date

Objective. We evaluated the biometric values of first trimester measurements for prediction of delivery date by computing new dating formulas and evaluating them within the same settings in a large population. Methods. We studied unselected pregnancies undergoing first trimester ultrasound examination by a single experimented sonographer. We used 331 pregnancies conceived by assisted reproductive treatment to build a new dating formula and another 3667 normal pregnancies with spontaneous delivery to test the accuracy of these measurements and formulas. The reproducibility of the two measurements which performed best was done. Results. Crown-rump length (CRL) had the lowest random and systematic errors. The mean (SD) of errors in predicting day of delivery were 0.023 (7.873), 0.092 (7.928), 0.088 (8.208) and 0.269 (8.310) for CRL, biparietal diameter (BPD), head circumference (HC) and abdominal circumference (AC) respectively. CRL and BPD proved to be highly reproducible. The percentages of deliveries within +/- 10 and +/- 14 days of the predicted term were comparable for CRL and BPD but were significantly smaller when using HC, AC. Conclusions. Both CRL and BPD proved to be highly reproducible and able to predict with good precision the date of delivery. However, CRL has significantly lower random and systematic errors than all other biometric parameters.