Is size discordancy an indication for delivery of preterm twins?

Our goal was to determine the clinical significance of size discordancy in preterm twins.

A retrospective study was performed to review outcomes of twins delivered between Jan. 1, 1988, and June 30, 1995. Maternal and neonatal records were assessed for demographic data, maternal medical history, and neonatal mortality and morbidity outcomes. Discordancy was defined as > or = 20% difference in birth weight. The chi 2 analysis was performed.

There were 42 sets of discordant twins and 77 sets of concordant twins in the final analysis. The distribution of gestational ages in both groups was similar. We found no difference in maternal morbidity between the groups. Discordant sets had a significantly longer hospital stay (p = 0.003) and more cases of hyperbilirubinemia (p = 0.01), but there were no other differences in morbid outcomes. There were no differences in outcome variables between the two twins within discordant sets with respect to gender, size, birth order, growth restriction, or route of delivery. There were no stillbirths among any of the 238 infants. Of the 15 neonatal deaths, none occurred in infants delivered after 32 weeks' gestation or in infants weighing > 2000 gm at birth. Infants who were small for gestational age had a higher incidence of sepsis (p = 0.043) and longer hospital stays (p = 0.005) compared with infants who were appropriate for gestational age.

Size discordancy alone does not appear to be an indication for preterm delivery of twins. When results of antenatal testing are normal and growth restriction is absent, attempts should be made to achieve a gestational age > 32 weeks and weight > 2000 gm before delivery is considered.