Neonatal outcome in a Danish national cohort

BACKGROUND: In Denmark, one-third of twin pregnancies are the result of IVF/ICSI treatment. Limited data on neonatal outcome in IVF/ICSI twins are available in the literature. METHODS: A register study was conducted on neonatal morbidity and mortality in a complete national twin cohort including all 3438 (3393 live-born) IVF/ICSI and 10 362 (10 239 live-born) non-IVF/ICSI twins born between 1995 and 2000. Twins were identi®ed in the National Medical Birth Registry and dichotomized into IVF/ICSI and non-IVF/ICSI by cross-reference with the Danish IVF Registry. Data on neonatal morbidity and mortality were retrieved from the Danish Patient Registry and the Danish Registry of Causes of Deaths. In order to exclude monozygotic twins, sub-analyses on unlike-sex twins were conducted. RESULTS: A birth weight discordance of>20 % was observed in 20.6 % of IVF/ICSI versus 15.7 % of control twin pairs (P < 0.001). The risk of discordant birth weight>20 % was OR 1.29 (95 % CI 1.04±1.58) in unlike-sex IVF/ICSI twins versus control twins. The risk of delivery at <37 completed weeks and birth weight <2500 g was similar in the two cohorts; however, in unlike-sex IVF/ICSI versus control twins the risk of delivery at <37 weeks and birth weight <2500 g was OR 1.22 (95 % CI 1.09±1.38) and OR 1.25 (1.11±1.40) respectively. After strati®cation for maternal age and parity, these risks disappeared. IVF/ICSI twins carried a higher risk of admit-tance to a neonatal intensive care unit (NICU) than control twins (OR 1.18, 95 % CI 1.09±1.27), and this was even more pronounced in unlike-sex twins [OR 1.34 (95 % CI 1.19±1.51)]. No differences were observed in malformatio