Late preterm rupture of membranes: it pays to wait.

www.thelancet.com Published online November 9, 2015 http://dx.doi.org/10.1016/S0140-6736(15)00809-0 1 In contrast to previous assumptions, there is increasing evidence that being born in the late preterm period— between 34 and 36 weeks gestation—is associated with important long-term adverse eff ects. Several adverse outcomes have been reported, including cerebral palsy, more hospital admissions in early childhood, lower childhood height, asthma, limiting long-term illness, and po orer educational attainment. Findings from studies show a gradient of health outcomes with decreasing gestation. An estimated 4–5% of infants are born at 34–36 weeks, and 30% of preterm births follow pre-labour rupture of the membranes. Because of the potential risks of fetal and neonatal infection—although with limited evidence to support this assumption—present guidance favours planned early delivery in women presenting with ruptured membranes at 34–36 weeks. With the emerging evidence of diff erences in long-term outcomes between late preterm and term infants, robust assessment of the risks and benefi ts of this strategy is essential, because a small increase in gestation at birth is likely to be benefi cial to the infant. In The Lancet, Jonathan Morris and colleagues present the results of a pragmatic randomised controlled trial of planned immediate delivery versus expectant management in women presenting with pre-labour ruptured membranes at 34–36 weeks. Findings from this trial advance substantially the evidence on the optimum management strategy in these women. 1839 women in whom there was no indication for urgent delivery were randomly assigned to immediate delivery (n=924) or expectant management (n=915). There was no diff erence in the primary outcome of defi nite or probable neonatal sepsis between neonates in the immediate birth and expectant management groups (23 [2%] of 923 vs 29 [3%] of 912; relative risk [RR] 0·8, 95% CI 0·5–1·3). Additionally, there was no diff erence between groups in a secondary composite neonatal outcome of sepsis, ventilation for 24 h or more, or death (73 [8%] of 923 in the immediate delivery group vs 61 [7%] of 911 in the expectant management group; RR 1·2, 95% CI 0·9–1·6). Infants of women assigned to planned immediate delivery had a signifi cantly higher risk of respiratory distress syndrome (76 [8%] of 919 vs 47 [5%] of 910; RR 1·6, 95% CI 1·1–2·3) and mechanical ventilation (114 [12%] of 923 vs 83 [9%] of 912; 1·4, 1·0–1·8) compared with those whose mothers were assigned to expectant management. These infants also had a signifi cantly longer stay in a special care nursery or neonatal intensive care unit (median 4·0 days, IQR 0·0–10·0 vs 2·0 days, 0·0–7·0) and a longer total hospital stay (6·0 days, 3·0–10·0 vs 4·0 days, 3·0–8·0). However, in contrast to their infants, mothers in the expectant management group had a longer hospital stay than mothers who were assigned to planned immediate delivery (median 6·0 days, IQR 4·0–9·0 vs 5·0 days, 3·0–7·0), owing to the fact that most women in the expectant group were managed in hospital and were not discharged home to await the onset of labour. Almost 90% of women randomly assigned to expectant management received antibiotics before delivery, but this was not universal despite clear evidence of benefi t. Also, a planned subgroup analysis of women who had group B streptococcus cultured from a vaginal swab showed no diff erence in the primary outcome of neonatal sepsis between the groups (RR 0·9, 95% CI 0·2–4·5). The main strength of Morris and colleagues’ study is its size; previous meta-analyses included a total of only 1230 infants. However, a concern associated with the present study is the time taken to recruit suffi cient women—10 years. Patterns of obstetric care are unlikely to have changed suffi ciently over that time to have a major eff ect on the study’s results, but preferences for antibiotic Late preterm rupture of membranes: it pays to wait