Article Text

Randomised controlled trial
Elective delivery of twins at 37 weeks gestation decreases infant complications
  1. Sarah Jane Stock1,
  2. Jane Elizabeth Norman2
  1. 1Department of Obstetrics and Gynaecology, King Edward Memorial Hospital, Perth, Western Australia, Australia
  2. 2MRC Centre for Reproductive Health, University of Edinburgh, Edinburgh, UK
  1. Correspondence to : Dr Sarah Jane Stock
    Department of Obstetrics and Gynaecology, King Edward Memorial Hospital, Bagot Road, Subiaco, Perth, Western Australia 6160, Australia; sarah.stock{at}

Statistics from

Commentary on: OpenUrlCrossRefPubMed


In singleton pregnancies, offering elective delivery at or beyond term (40 weeks gestation) reduces perinatal mortality.1 ,2 Epidemiological data suggest that in twin pregnancies ‘term’ may be earlier than in singletons. Morbidity and mortality in twin pregnancies is the lowest in association with delivery at 36–38 weeks gestation, leading the National Institute for Health and Clinical Excellence (NICE) to endorse elective delivery around 37 weeks for dichorionic twins and 36 weeks for monochorionic twins.3 Data from randomised trials evaluating such a policy is lacking.


Dodd and colleagues report the results of a randomised controlled trial of elective delivery at 37 weeks gestation or standard care (delivery planned from 38 weeks gestation) in women with uncomplicated twin pregnancies. Women were recruited at 36+6 weeks gestation or greater from 13 hospitals in Australia, New Zealand and Italy between 2003 and 2010.

The primary outcome was a composite outcome of serious adverse infant outcomes. Although a sample size of 460 was planned (80% power to detect a reduction in primary outcome from 16.3% to 6.7%, with alpha 0.05), the study was stopped before this target was reached owing to lack of continued funding. Treatment allocation was masked from outcome assessors and analysis was performed on an intention to treat basis.


A total of 116 women were randomised to elective delivery at 37 weeks and 119 women to standard care (delivery planned after 38 weeks), with outcome data available on all infants. The mean gestational age at delivery in the elective delivery group was 37.3 (±0.4) weeks vs 37.9 (±0.5) weeks in the expectant management group.

Infants in the elective delivery group had lower serious adverse outcomes than infants in the standard care group (elective delivery group 11/232 (4.7%) versus standard care group 29/238 (12.2%); RR 0.39%, 95% CI 0.20% to 0.75%, p=0.005). Infants in the elective delivery group were at lower risk of morbidity (elective delivery group 10/232 (4.3%) versus standard care group 28/238 (11.8%); RR 0.37%, 95% CI 0.18% to 0.73%, p=0.004). This was due to reduction in the incidence of birthweight <third centile for gestational age and sex, which was included in the composite outcome of morbidity. As prespecified in the trial protocol, birthweight centiles were calculated using singleton growth standards. A post hoc secondary analysis using twin specific growth standards found that although there was a trend for a reduction in serious adverse infant outcome, serious infant morbidity and birthweight <third centile in the elective delivery group, the differences were no longer significant.


This study attempts to answer an important question regarding the management of twin pregnancies: whether planned delivery at 37 weeks improves infant outcome. A significant reduction in adverse infant outcome was seen in the elective delivery group, primarily driven by the inclusion of birthweight <third centile (against singleton standards) in the composite outcome of morbidity. Given the link between growth restriction and adverse perinatal outcome, the findings should, if confirmed, alter clinical practice in favour of planned delivery at 37 weeks. Although the study was terminated prematurely, the effect size was bigger than anticipated, thus a significant difference in the incidence of the prespecified primary outcome was observed. However, when twin specific growth standards were used to calculate birth centiles, no significant difference was found between the two groups. Additionally, the reason for the difference in the incidence of severe growth restriction is unclear as there was only a 4-day difference in gestational age between the two groups, and the authors themselves suggest that the observed findings could be a chance effect.

The study included both monochorionic and dichorionic twin pregnancies, but the sample size of each was not sufficient to perform subgroup analysis to explore the potential different effects of chorionicity on outcome. Importantly, the major confounder of mode of delivery was similar in both groups; thus, the findings of the recently completed Twin Birth Trial ( will not change the interpretation of this study.

Although Dodd and colleagues’ trial does not provide strong evidence of a clinically important benefit of elective delivery, it does suggest that some morbidity is reduced by elective delivery. The trial thus supports observational data, and current national recommendations that women with uncomplicated twin pregnancy should be offered delivery around 37 weeks to optimise infant outcome.3 The authors should be congratulated in providing the best current evidence on timing of twin delivery.


View Abstract


  • Competing interests None.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.