Fetal weight estimation, evaluation of different methods

Introduction : The accuracy of ultrasound estimation of fetal weight is influenced by the imprecision of ultrasound methods. The aim of this study is to compare the accuracy of fetal weight estimation between conventional (2D) models and those including 2D and 3D subcutaneous tissue measurements. Methods : Prospective study of 199 low-risk pregnant women at the Hospital Universitario La Plana, Vila-Real, Spain. Data acquisition included the 2D and 3D parameters described by Lee, Hadlock and Scioscia. Results : All ultrasound formulas correlated positively with actual birth weight calculated by Sperman's method. Hadlock's model has the highest correlation with a different percentage < 5% of 42% and Sperman's correlation (r = 0.68).If we disaggregate the results by fetal weight categories, in newborns > 3000 g, Hadlock’s method was more accurate, 50% of the fetuses had less than 5% difference calculated as mean percentage difference, in newborns < 3000 g Lee’s method obtained more accurate results, 68.8% of the estimated weigths were within 5% of birth weight.Overall, systematic and random errors were lower for the Hadlock models than for the Lee and Scioscia models. Conclusions : The inclusion of fractional thigh volume in the ultrasound estimation of fetal weight improves the estimation of fetuses weighing less than 3000g. In our study Hadlock's method provided more accurate estimates of fetal weight in fetuses with normal range growth.