Estimation of birth weight by use of ultrasonographic formulas targeted to large-, appropriate-, and small-for-gestational-age fetuses.

This report sought to determine whether estimates of fetal weight are enhanced by using ultrasonographic formulas targeted to the large-, appropriate-, and small-for-gestational-age fetus in both preterm and term pregnancies. Ultrasonographic fetal measurements from 575 singleton pregnancies were obtained within 7 days of delivery. The first 194 fetuses were classified into three groups on the basis of the growth percentile rank of the abdominal circumference (greater than or equal to 90%, greater than 5% and less than 90%, and less than or equal to 5%). Regression analyses were performed to generate three formulas for estimating fetal weight on the basis of the following: Gestational age (weeks) + Head circumference (cm) + 2 x Abdominal circumference (cm) + Femur length (cm). This mathematic model allowed formulation of accurate and concise tables relating the sum of these measurements to estimated birth weight. The accuracy of these formulas was then prospectively compared, first, with the formula published by Hadlock et al. (in 32 large-, 279 appropriate-, and 70 small-for-gestational-age fetuses) and second, with the formula of Weiner et al. (in 82 preterm fetuses). The difference between actual and estimated birth weights generated by the three study formulas had no systemic error (t test, p greater than 0.05). Cumulatively, there was a statistically significant reduction of random error in the birth weight estimates by use of the three study formulas versus the best single formulas of Hadlock et al. (2 SD reduced from 21.6% to 19.8% and absolute 2 SD reduced from 15.6% to 12.2%). Additionally, the three study formulas resulted in a statistically significant reduction in the absolute 2 SD error compared with the best formula by Weiner et al. in fetuses less than or equal to 34 weeks' gestation (2 SD reduced from 19.1% to 13.6%).