Perinatal outcomes of singleton live births with and without vanishing twin following transfer of multiple embryos: analysis of 113 784 singleton live births

STUDY QUESTION: Does transfer of multiple embryos affect perinatal outcomes of resulting singleton live births following ART? SUMMARY ANSWER: There is a higher risk of preterm birth (PTB) and low birthweight (LBW) in singleton live births associated with spontaneous reduction of an initial multiple to singleton gestation following transfer of multiple embryos. WHAT IS KNOWN ALREADY: Singleton pregnancies following ART are at a higher risk of adverse perinatal outcomes compared to spontaneous conceptions. Earlier studies have found an increased risk of PTB and LBW in singletons following transfer of multiple embryos versus single embryo transfer (SET). However, these studies did not address the specific role of vanishing twin, i.e. spontaneous reduction of an initial multiple to singleton gestation. STUDY DESIGN, SIZE, DURATION: Anonymised data on all ART cycles performed in the UK were obtained from the Human Fertilization and Embryology Authority. Data from 1991 to 2011 involving 508 410 fresh and 131 157 frozen autologous ART cycles resulting in 95 779 and 18 005 singleton live births, respectively, were analyzed. PARTICIPANTS/MATERIALS, SETTING, METHODS: Fresh and frozen ART cycles were analyzed separately to compare perinatal outcomes of PTB and LBW of singleton live births resulting from transfer of multiple (>= 2) embryos versus SET. Logistic regression analysis was performed adjusting for confounders. Subgroup analyses were carried out for perinatal outcomes of singleton live births with initial multiple or initial single gestational sacs following transfer of multiple embryos versus singleton live births following SET. MAIN RESULTS AND THE ROLE OF CHANCE: In fresh cycles, there was a significantly higher risk of PTB (adjusted odds ratio (aOR) 2.70, CI 2.37-3.05) and LBW (aOR 2.76, CI 2.44-3.13) in singleton live births with initial multiple gestational sacs but there was no significant difference in the risk of PTB (aOR 1.08, CI 1.00-1.16) or LBW (aOR 1.08, CI 1.00-1.16) in singleton live births with an initial single gestational sac following transfer of >= 2 embryos compared to those following SET. In frozen cycles, there was a significantly higher risk of PTB (aOR 2.13, CI: 1.55-2.93) and LBW (aOR 2.61, CI: 1.87-3.64) in singleton live births with initial multiple gestational sacs but there was no significant difference in the risk of PTB (aOR 1.02, CI: 0.88-1.18) or LBW (aOR 0.91, CI: 0.77-1.07) in the singleton live births with an initial single gestational sac following transfer of >= 2 embryos compared to those following SET. LIMITATIONS, REASONS FOR CAUTION: While the analysis was adjusted for a number of known confounders, the dataset had no information for confounders such as smoking, BMI, previous obstetric history and comorbid medical conditions during pregnancy. The lack of information about the timing of occurrence of the vanishing phenomenon is another limitation because poorer perinatal outcomes of a surviving twin have been reported following second trimester fetal demise compared to the first trimester. WIDER IMPLICATIONS OF THE FINDINGS: The study results suggest that the vanishing twin phenomenon is associated with increased risk of PTB and LBW in the resulting singleton live births and there was no increased risk when there was a single gestational sac from the outset following transfer of multiple embryos. STUDY FUNDING/COMPETING INTERESTS: Nil.