Deviation of sonographic estimated fetal weight from actual birth weight in two consecutive pregnancies of the same parturients

Objective: To determine whether maternal features affect the accuracy of sonographic estimated fetal weight (SEFW) by evaluating the consistency of the systematic error of SEFW with regards to the birth weight (BW) in two consecutive pregnancies of the same gravida. Methods: The cohort included women with SEFW within 1 week of delivery (32-42 weeks' gestation) in two consecutive singleton pregnancies from 2007 to 2012. The systematic error was calculated as (SEFW - BW)/BW × 100 and expressed as a percentage of the BW. Results: A total of 636 pregnancies (318 gravidas) were eligible for analysis. The BW and SEFW were correlated in both first examined (r = 0.873, p < 0.001) and consecutive (r = 0.843, p < 0.001) pregnancies. There was a significant difference in mean systematic error between first examined and consecutive pregnancies (3.13 ± 8.95 vs. 0.34 ± 8.75%, p < 0.001), with a very weak correlation between the two (r = 0.135, p = 0.16). Nulliparity or multiparity at the first examined pregnancy was not found to be a significant factor, and in both groups the error was higher in the first examined pregnancy. There were no significant differences between parturients with a minor (10% and below) or major (>10%) difference in the systematic error between the two pregnancies. Conclusion: The systematic error between the SEFW and BW in two consecutive pregnancies is inconsistent, suggesting that it is unaffected by maternal biometric features.