Prediction of outcome in twin pregnancy with first and early second trimester ultrasound.

OBJECTIVE To establish if first or second trimester biometry is a useful adjunct in the prediction of adverse perinatal outcome in twin pregnancy. METHODS A consecutive cohort of 1028 twin pregnancies was enrolled for the Evaluation of Sonographic Predictors of Restricted growth in Twins (ESPRiT) study, a prospective study conducted at eight academic centers. Outcome data was recorded for 1001 twin pairs that completed the study. Ultrasound biometry was available for 960 pregnancies. Biometric data obtained between 11 and 22 weeks were evaluated as predictors of a composite of adverse perinatal outcome (mortality, hypoxic ischemic encephalopathy, periventricular leukomalacia, necrotizing enterocolitis, respiratory distress, or sepsis), preterm delivery (PTD) and birthweight discordance greater than 18% (18% BW). Outcomes were adjusted for chorionicity and gestational age using Cox Proportional Hazards regression. RESULTS Differences in crown-rump length (CRL) were not predictive of adverse perinatal outcome. Between 14 and 22 weeks, a difference in abdominal circumference (AC) of more than 10% was the most useful predictor of adverse outcome, PTD and 18% or more BW discordance in all twins. Overall the strongest correlation was observed for intertwin differences in biometry between 18 and 22 weeks. CONCLUSION Biometry in the early second trimester can successfully identify twin pregnancies at increased risk. Intertwin AC difference of greater than 10% between 14 and 22 weeks gestation was the best individual predictor of perinatal risk in all twins. Sonographic biometry in the early second trimester should therefore be utilized to establish perinatal risk, thus allowing prenatal care to be improved.