Prediction of growth discordance in twins with the use of abdominal circumference ratios.

Abstract

The purpose of this study was to assess the accuracy of sonographic prediction of twin birth weight discordance using the abdominal circumference ratio. This was a prospective cohort of diamniotic twin gestations that underwent serial ultrasound examinations every 2 to 4 weeks from 11 to 38 weeks of gestation. Birth weight discordance was defined as > or =25% difference in birth weight, relative to the larger twin. The sensitivity, specificity, and predictive values for the abdominal circumference ratio were assessed for the prediction of growth discordance. Of 503 diamniotic twin pregnancies, 64 pregnancies (12.7%) had discordant fetal growth. The abdominal circumference ratio could be measured consistently throughout gestation in 100% of twin pairs. Receiver operating curve analysis showed that the abdominal circumference ratio was a good predictor of birth weight discordance (area under the curve = 0.80). An abdominal circumference ratio cutoff of 0.93 yielded a sensitivity and specificity of 61% and 84%, respectively. Twin birth weight discordance may be predicted at any gestational age with an abdominal circumference ratio <0.93.