PURPOSE: to evaluate the correlation between the estimated fetal weight (EFW) by ultrasonography and the neonatal weight (NW), as well as the EFW's capacity to predict changes in NW among pregnant women in João Pessoa, Paraíba, Brazil. METHODS: a diagnostic validation study, including 122 pregnant women who have had the EFW calculated by ultrasonography up to seven days before delivery and the NW established immediately after birth, with a specific newborn's scale. The correlation between EFW and NW measurements was assessed by Pearson's correlation coefficient and by the mean difference between them. EFW and NW were classified as: low for the gestational age (LGA), adequate for the gestational age (AGA) and high for the gestational age (HGA), according to the percentiles 10 and 90 of the respective reference curves. The diagnosis of EFW deviation has been validated using the values of the Alexander's NW reference curve as gold-standard, by estimating the sensitivity, specificity, and positive and negative predictive values. RESULTS: there has been a high linear correlation between the EFW and NW (R=0.96), and the difference between them has varied from -474 g to +480 g, with an average of +3 g. Most of the highest percent weight estimate variations were between 10 and 15%. EFW has had 85.7% of sensitivity and 100% of specificity for the detection of LGA, and 100 and 77.2%, respectively, for the detection of HGA. CONCLUSIONS: EFW is able to predict NW adequately, and the reference EFW tested has had a good performance in the screening of fetal growth deviation, in the population studied.
Birth weight; Ultrasonography; prenatal; Fetal development; Fetal growth retardation
