Reliability of ultrasound estimation of fetal weight in term singleton pregnancies

To assess the reliability of ultrasound estimation of fetal weight undertaken antenatally at Wellington Hospital (Wellington City, New Zealand) in women with a singleton pregnancy = or >37 weeks gestation.

Data were collected retrospectively for pregnant women who had undergone ultrasound estimation of fetal weight <7 days prior to a term delivery (= or >37 weeks gestation) over the period of July 1998-June 2005. Stillbirths and multiple pregnancies were excluded. Ultrasound fetal weight estimations, calculated using a locally modified Woo formula, were compared with the infant's actual birth weight.

A total of 1177 infants were studied. The mean absolute error and mean signed error (+/-SD) of ultrasound fetal weight estimations were 7.0+/-5.7% and -0.2+/-9.0%, respectively (n=1177). Three-quarters of estimations were within 10% of birth weight. Ultrasonic estimation of fetal weight tended to overestimate the weight of small infants (<2500 g; mean signed error = +3.5%+/-9.1%, n=98) and underestimate the weight of large infants (= or >4000 g; mean signed error = -3.3+/-8.7%, n=170). Both large and normal weight infants of women with diabetes tended to have their weight underestimated (mean signed error = -5.1+/-9.2%, n=48). Sensitivity, specificity, positive predictive value, and negative predictive value for ultrasonic detection of fetal weight = or >4000 g in non-diabetic women were 61%, 96%, 69%, and 94%, respectively. For detection of fetal weight = or >4500, the figures were 50%, 98%, 47%, and 98%, respectively.

The accuracy of ultrasound estimations of fetal weight performed at Wellington Hospital within 7 days of delivery in term singleton pregnancies was at least similar and sometimes better than that reported in other studies. For one in four women, however, the fetal weight estimation was more than 10% different from the actual birth weight of their infant. Ultrasound measurements had a tendency to overestimate the weight of small infants while underestimating the weight of both large infants and the infants of diabetic mothers. As the reliability of ultrasound estimation of fetal weight to detect larger babies was poor, the use of such an objective measurement in the management of suspected macrosomia in term singleton pregnancies should be avoided.