Preterm twins: what threshold of birth weight discordance heralds major adverse neonatal outcome?

This study was undertaken to establish the optimal threshold of birth weight discordance for prediction of adverse outcome in liveborn, nonmalformed preterm twins. We accessed a cohort of twin gestations for the period 1990 through 2000 delivered at less than 37.0 weeks' gestation. Adverse neonatal outcome was defined as stillbirth or occurrence of major morbidities. A total of 335 twin gestations (670 twins) were included, of which 104 (31%) experienced adverse neonatal outcome. The average birth weight discordance was 12.4% ± 10.6% versus 19.4% ± 14.8% in those with good versus adverse outcome (P <. 001). Birth weight discordance was a significant predictor of adverse neonatal outcome that was independent of gestational age at delivery, small for gestational age, and chorionicity. Receiver operating characteristic curve analysis suggested that optimal thresholds of birth weight discordance for clinical use should take into account gestational age. Birth weight discordance is an independent predictor of adverse neonatal outcome in liveborn, nonmalformed preterm twins. © 2004 Elsevier Inc. All rights reserved.