Presented at the Neonatal Society 2007 Spring Meeting.

Ray B, Ward Platt MP (on behalf of the Multiple Pregnancy Register Steering Group)

Royal Victoria Infirmary, Newcastle Upon Tyne, UK

Objective: To determine the mortality rates of live born twins compared to singletons of less than 30 weeks’ gestation in relation to gestational age, chorionicity and year of birth in a geographically defined population.

Study Design: We identified all live born babies of 23-29 completed weeks of gestation born in the North of England between January 1998 and Dec 2004, which allowed us to compare 397 twin babies with 1322 singletons. We used the regional Multiple Pregnancy Register, Perinatal Mortality Survey, and the regional neonatal intensive care unit (NICU) database, and hand searched for live-born babies not captured in the NICU database. We excluded all triplets and other higher order multiple births and also babies with significant congenital anomalies. We ascertained all deaths up to 1 year. We did not require any ethical approval, because none of the patient identifying information was used.

Results: Both the groups were demographically comparable, with median birth weight of 990 in singletons & 995 in twins (IQR 775-1230 & 765-1200 respectively). Gender ratio showed slight male preponderance on both sides; 54% in singletons and 56% in twins. Babies born by LUCS were 37% in singletons and 41% in twins. We observed an overall higher death rates amongst twins compared to singletons with an odds ratio 1.36 (CI 1.06-1.36). These deaths were particularly noticeable in gestation of 23-25 weeks, in the early neonatal period [OR (CI) 2.01 (1.31-3.07)], and the late neonatal period [OR (CI) 2.16 (1.09-4.25)], but not after the first month. Among babies of 26-29 weeks, twins and singletons had similar early neonatal, late neonatal and post-neonatal infant mortality rates. Between 1998 and 2004 mortality improved significantly in twins of 25-26 weeks and in singletons of 24-25 weeks’ gestation. We found no effect of chorionicity on the mortality outcome of twins.

Conclusion: The excess mortality among twins of less than 30 weeks’ gestation is confined to neonatal deaths among babies of 25 weeks or less. However the outlook for both twins and singletons has improved substantially in recent years.