Early birth compared with waiting for birth in babies who are thought to be coping poorly towards the end of pregnancy

What is the issue?

What care should be given to babies who are thought to be coping poorly towards the end of pregnancy (beyond 37 weeks)? A baby may be in this situation because the placenta is no longer functioning well and this means the baby may be short of nutrition or oxygen. We asked in this Cochrane review if it is better to induce labour or do a caesarean section (both ways of ensuring the baby is born earlier) rather than letting the pregnancy continue until labour starts by itself.

Why is this important?

Sometimes, when a healthy pregnant woman gets towards the end of pregnancy, there may be signs that her baby may be having difficulty coping. Some of these babies are born sick, very occasionally they do not survive, or they have problems in their later development. Several factors may indicate problems. A baby may not be growing normally and so is smaller than expected (this is termed intrauterine growth restriction - IUGR). The baby may show decreased movements, which may indicate the placenta is no longer functioning well. Fetal heart monitoring (known as cardiotocography or CTG) may show up a possible problem. Ultrasound can also measure amniotic fluid and blood flow in order to assess the baby’s well-being.

Induction of labour or caesarean section might help these babies by taking them out of the uterus. But intervening early in this way may mean that these babies’ lungs are not mature enough to deal well with the outside world, and they might be better to continue inside the uterus. It is not clear which option is best for mothers and babies.

What evidence did we find?

We found three trials involving 546 pregnant women and their babies at term (search – 31 May 2015). All three trials looked at using induction of labour for an early birth. Two trials looked at babies thought to have growth restriction and one trial looked at babies thought to have a small volume of amniotic fluid (oligohydramnios). All three trials were of reasonable quality and most of the evidence comes from the largest trial which compared babies who were growth restricted. There is no information about funding sources for these trials.

Overall, we found no major differences between these two strategies in terms of the babies’ survival, the numbers of very sick babies nor in the numbers of babies with problems in development.

We looked at many other outcomes, too, including how many caesarean sections there were, and how many operative vaginal births (with forceps or ventouse). However, we cannot be sure there are no real major differences because the numbers of women and babies included were too small. There were more inductions of labour in the early birth group.

What does this mean?

There is not enough evidence from trials to guide clinical practice regarding earlier birth versus waiting for birth in healthy women towards the end of pregnancy who are carrying babies who may be coping poorly. Further large trials are needed. We also need research into better tests to identify babies who are not coping well towards the end of pregnancy. Women should discuss their specific circumstances with their caregivers when coming to a decision.

Authors' conclusions: 

A policy for planned early delivery versus expectant management for a suspected compromised fetus at term does not demonstrate any differences in major outcomes of perinatal mortality, significant neonatal or maternal morbidity or neurodevelopmental disability. In women randomised to planned early delivery, the gestational age at birth was on average 10 days earlier, women were less likely to have a baby beyond 40 weeks' gestation, they were more likely to be induced and infants were more likely to be admitted to intermediate care nursery. There was also a significant difference in the proportion of babies with a birthweight centile < 2.3rd, however this did not translate into a reduction in morbidity. The review is informed by only one large trial and two smaller trials assessing fetuses with IUGR or oligohydramnios and therefore cannot be generalised to all term pregnancies with suspected fetal compromise. There are other indications for suspecting compromise in a fetus at or near term such as maternal perception of DFM, and ultrasound and/or CTG abnormalities. Future randomised trials need to assess effectiveness of timing of delivery for these indications.

Read the full abstract...

Fetal compromise in the term pregnancy is suspected when the following clinical indicators are present: intrauterine growth restriction (IUGR), decreased fetal movement (DFM), or when investigations such as cardiotocography (CTG) and ultrasound reveal results inconsistent with standard measurements. Pathological results would necessitate the need for immediate delivery, but the management for ‘suspicious’ results remains unclear and varies widely across clinical centres. There is clinical uncertainty as to how to best manage women presenting with a suspected term compromised baby in an otherwise healthy pregnancy.


To assess, using the best available evidence, the effects of immediate delivery versus expectant management of the term suspected compromised baby on neonatal, maternal and long-term outcomes.

Search strategy: 

We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (31 May 2015) and reference lists of retrieved studies.

Selection criteria: 

Randomised or quasi-randomised controlled trials comparing expectant management versus planned early delivery for women with a suspected compromised fetus from 37 weeks' gestation or more.

Data collection and analysis: 

Two review authors independently assessed trials for inclusion and assessed trial quality. Two review authors independently extracted data. Data were checked for accuracy. We assessed the quality of the evidence using the GRADE approach.

Main results: 

Of the 20 reports identified by the search strategy, we included three trials (546 participants: 269 to early delivery and 277 to expectant management), which met our inclusion criteria. Two of the trials compared outcomes in 492 pregnancies with IUGR of the fetus, and one in 54 pregnancies with oligohydramnios. All three trials were of reasonable quality and at low risk of bias. The level of evidence was graded moderate, low or very low, downgrading mostly for imprecision and for some indirectness. Overall, there was no difference in the primary neonatal outcomes of perinatal mortality (no deaths in either group, one trial, 459 women, evidence graded moderate), major neonatal morbidity (risk ratio (RR) 0.15, 95% confidence interval (CI) 0.01 to 2.81, one trial, 459 women, evidence graded low), or neurodevelopmental disability/impairment at two years of age (RR 2.04, 95% CI 0.62 to 6.69,one trial, 459 women, evidence graded low). There was no difference in the risk of necrotising enterocolitis (one trial, 333 infants) or meconium aspiration (one trial, 459 infants), There was also no difference in the reported primary maternal outcomes: maternal mortality (RR 3.07, 95% CI 0.13 to 74.87, one trial, 459 women, evidence graded low), and significant maternal morbidity (RR 0.92, 95% CI 0.38 to 2.22, one trial, 459 women, evidence graded low).

The gestational age at birth was on average 10 days earlier in women randomised to early delivery (mean difference (MD) -9.50, 95% CI -10.82 to -8.18, one trial, 459 women) and women in the early delivery group were significantly less likely to have a baby beyond 40 weeks' gestation (RR 0.10, 95% CI 0.01 to 0.67, one trial, 33 women). Significantly more infants in the planned early delivery group were admitted to intermediate care nursery (RR 1.28, 95% CI 1.02 to 1.61, two trials, 491 infants). There was no difference in the risk of respiratory distress syndrome, (one trial, 333 infants), Apgar score less than seven at five minutes (three trials, 546 infants), resuscitation required (one trial, 459 infants), mechanical ventilation (one trial, 337 infants), admission to neonatal intensive care unit (NICU) (RR 0.88, 95% CI 0.35 to 2.23, three trials, 545 infants, evidence graded very low), length of stay in NICU/SCN (one trial, 459 infants), and sepsis (two trials, 366 infants).

Babies in the expectant management group were more likely to be < 2.3rd centile for birthweight (RR 0.51, 95% CI 0.36 to 0.73, two trials, 491 infants), however there was no difference in the proportion of babies with birthweight < 10th centile (RR 0.98, 95% CI 0.88 to 1.10). There was no difference in any of the reported maternal secondary outcomes including: caesarean section rates (RR 1.02, 95% CI 0.65 to 1.59, three trials, 546 women, evidence graded low), placental abruption (one trial, 459 women), pre-eclampsia (one trial, 459 women), vaginal birth (three trials 546 women), assisted vaginal birth (three trials 546 women), breastfeeding rates (one trial, 218 women), and number of weeks of breastfeeding after delivery one trial, 124 women). There was an expected increase in induction in the early delivery group (RR 2.05, 95% CI 1.78 to 2.37, one trial, 459 women).

No data were reported for the pre-specified secondary neonatal outcomes of the number of days of mechanical ventilation, moderate-severe hypoxic ischaemic encephalopathy or need for therapeutic hypothermia. Likewise, no data were reported for secondary maternal outcomes of postnatal infection, maternal satisfaction or views of care.