First, the risks of stillbirth and neonatal mortality in twin pregnancies with varying levels of development discordance and connection to small for gestational age (SGA) were assessed, and ideal gestational ages for delivery were determined. Second, these optimal gestational ages were compared to previously established optimal delivery timing for twin pregnancies not complicated by fetal growth restriction, which was calculated at 37 0/7 weeks of gestation for dichorionic pregnancies and 36 0/7 weeks for monochorionic pregnancies in a previous individual patient meta-analysis.

Between 2015 and 2018, a search of cohort studies reporting risks of stillbirth and infant mortality in twin pregnancies from 32 to 41 weeks of gestation was conducted in MEDLINE, EMBASE,, and Ovid. In addition, a prior meta-analysis that used a similar search approach (from inception to 2015) was integrated. Women carrying monoamniotic twins were not eligible. In all, 20 cohort studies submitted original data on 7,474 dichorionic and 2,281 monochorionic twin pairs out of 57 eligible investigations. Researchers calculated the risk of perinatal mortality (the risk difference between prospective stillbirth and neonatal death) per gestational week using individual participant data meta-analysis. The analyses were divided into three groups based on chorionicity, degrees of growth discordance, and SGA occurrence in one or both twins. 

The absolute risks of stillbirth and infant mortality were higher in dichorionic and monochorionic twins when one or both twins were SGA and increased with higher degrees of growth discordance. In dichorionic and monochorionic twin pregnancies, the perinatal risk was balanced between 36 0/7–6/7 and 37 0/7–6/7 weeks of gestation, with a likely larger chance of stillbirth than neonatal mortality from 37 0/7–6/7 weeks forward. SGA, or growth discordance, was related to increased absolute chances of stillbirth and infant mortality. However, after considering these two hazards, they found no indication that the existence of development abnormalities alone changes the best date of delivery.