Does delaying cord clamping or using cord milking at birth improve the health of babies born too early?

What is the issue?

In this Cochrane Review, we set out to determine if delayed cord clamping or umbilical cord milking improves the health outcomes for babies born before 37 weeks' gestation. These interventions were compared with early cord clamping.

Why is it important?

Babies born before 37 weeks, or preterm, have poorer health outcomes than babies born at term, particularly if they are born before 32 weeks. Babies born preterm can experience problems with the functioning of many of their major organs including their lungs, gut and hearts. They have a greater risk of dying or having long-term problems such as cerebral palsy. After birth, the babies may need blood transfusions and drugs to strengthen their heart contractions (inotropes) and to raise their blood pressure. It is important to try to find ways of improving the health of these tiny babies.

Early clamping of the umbilical cord has been standard practice over many years. It allows the baby to be transferred quickly to care from a specialised team of doctors either at the side of the room or in another room. Yet, delayed clamping for half to three or more minutes allows continuing blood flow between the mother and her baby, and this may help the baby to adjust to breathing air. Squeezing blood along the umbilical cord towards the baby (milking the cord), can boost the baby's blood volume, and this may improve the baby's health. We wanted to see if there are any benefits or harms from either waiting to clamp or milking the cord.

What evidence did we find?

We collected and analysed all relevant studies to answer this question (date of search: November 2017). Our updated review included 40 studies which provided data on 4884 babies and their mothers. Studies were undertaken across the world, but mostly in high-income countries. Births were in hospitals which practiced early clamping. For many outcomes there were insufficient data to be really confident of our findings.

1) For delayed cord clamping (with immediate care of the baby after cord clamping) compared with early cord clamping, we found it likely that fewer babies died before discharge (20 studies, 2680 babies). Also, fewer babies may have had any bleeding in the brain (15 studies, 2333 babies), but there was probably no difference in the numbers of babies with very serious brain bleeds (10 studies, 2058 babies).

2) Only one study of 276 babies and their mothers provided data on delayed cord clamping with immediate care of the baby beside the mother with cord intact compared with early cord clamping. This study was small and did not identify any marked differences in health outcomes.

3) For delayed cord clamping (with immediate care of the baby after cord clamping) versus cord milking, there were insufficient data (three studies, 322 babies) to make comparisons between outcomes.

4) For cord milking versus early cord clamping, we found 11 studies providing data with 1183 babies and their mothers. Again, there were insufficient data to make clear comparisons on outcomes.

What does this mean?

Delayed cord clamping probably reduced the risk of death for babies born preterm. Early cord clamping probably causes harm. No studies showed what length of delay was best, and only a few studies followed babies for health outcomes in early childhood. There is insufficient evidence for reliable conclusions on providing immediate care for the baby beside the mother with the cord intact. Similarly, there is insufficient evidence for reliable conclusions on cord milking. Further studies are in progress.

Authors' conclusions: 

Delayed, rather than early, cord clamping may reduce the risk of death before discharge for babies born preterm. There is insufficient evidence to show what duration of delay is best, one or several minutes, and therefore the optimum time to clamp the umbilical cord remains unclear. Whilst the current evidence supports not clamping the cord before 30 seconds at preterm births, future trials could compare different lengths of delay. Immediate neonatal care with the cord intact requires further study, and there are insufficient data on UCM.

The nine new reports awaiting further classification may alter the conclusions of the review once assessed.

Read the full abstract...
Background: 

Infants born preterm (before 37 weeks' gestation) have poorer outcomes than infants at term, particularly if born before 32 weeks. Early cord clamping has been standard practice over many years, and enables quick transfer of the infant to neonatal care. Delayed clamping allows blood flow between the placenta, umbilical cord and baby to continue, and may aid transition. Keeping baby at the mother's side enables neonatal care with the cord intact and this, along with delayed clamping, may improve outcomes. Umbilical cord milking (UCM) is proposed for increasing placental transfusion when immediate care for the preterm baby is needed. This Cochrane Review is a further update of a review first published in 2004 and updated in 2012.

Objectives: 

To assess the effects on infants born at less than 37 weeks' gestation, and their mothers of: 1) delayed cord clamping (DCC) compared with early cord clamping (ECC) both with immediate neonatal care after cord clamping; 2) DCC with immediate neonatal care with cord intact compared with ECC with immediate neonatal care after cord clamping; 3) DCC with immediate neonatal care after cord clamping compared with UCM; 4) UCM compared with ECC with immediate neonatal care after cord clamping.

Search strategy: 

We searched the Cochrane Pregnancy and Childbirth Group Trials Register, ClinicalTrials.gov, the WHO International Clinical Trials Registry Platform (ICTRP) (10 November 2017), and reference lists of retrieved studies. We updated the search in November 2018 and added nine new trial reports to the awaiting classification section to be assessed at the next update.

Selection criteria: 

Randomised controlled trials (RCTs) comparing delayed with early clamping of the umbilical cord (with immediate neonatal care after cord clamping or with cord intact) and UCM for births before 37 weeks' gestation. Quasi-RCTs were excluded.

Data collection and analysis: 

Two review authors independently assessed trials for inclusion and risk of bias, extracted data and checked them for accuracy. Random-effects are used in all meta-analyses. Review authors assessed the certainty of the evidence using the GRADE approach.

Main results: 

This update includes forty-eight studies, involving 5721 babies and their mothers, with data available from 40 studies involving 4884 babies and their mothers. Babies were between 24 and 36+6 weeks' gestation at birth and multiple births were included. The data are mostly from high-income countries. Delayed clamping ranged between 30 to 180 seconds, with most studies delaying for 30 to 60 seconds. Early clamping was less than 30 seconds and often immediate. UCM was mostly before cord clamping but some were milked after cord clamping. We undertook subgroup analysis by gestation and type of intervention, and sensitivity analyses by low risk of selection and attrition bias.

All studies were high risk for performance bias and many were unclear for other aspects of risk of bias. Certainty of the evidence using GRADE was mostly low, mainly due to imprecision and unclear risk of bias.

Delayed cord clamping (DCC) versus early cord clamping (ECC) both with immediate neonatal care after cord clamping (25 studies, 3100 babies and their mothers)

DCC probably reduces the number of babies who die before discharge compared with ECC (average risk ratio (aRR) 0.73, 95% confidence interval (CI) 0.54 to 0.98, 20 studies, 2680 babies (moderate certainty)).

No studies reported on 'Death or neurodevelopmental impairment' in the early years'.

DCC may make little or no difference to the number of babies with severe intraventricular haemorrhage (IVH grades 3 and 4) (aRR 0.94, 95% CI 0.63 to 1.39, 10 studies, 2058 babies, low certainty) but slightly reduces the number of babies with any grade IVH (aRR 0.83, 95% CI 0.70 to 0.99, 15 studies, 2333 babies, high certainty).

DCC has little or no effect on chronic lung disease (CLD) (aRR 1.04, 95% CI 0.94 to 1.14, 6 studies, 1644 babies, high certainty).

Due to insufficient data, we were unable to form conclusions regarding periventricular leukomalacia (PVL) (aRR 0.58, 95% CI 0.26 to 1.30, 4 studies, 1544 babies, low certainty) or maternal blood loss of 500 mL or greater (aRR 1.14, 95% CI 0.07 to 17.63, 2 studies, 180 women, very low certainty).

We identified no important heterogeneity in subgroup or sensitivity analyses.

Delayed cord clamping (DCC) with immediate neonatal care with cord intact versus early cord clamping (ECC) (one study, 276 babies and their mothers)

There are insufficient data to be confident in our findings, but DCC with immediate neonatal care with cord intact may reduce the number of babies who die before discharge, although the data are also compatible with a slight increase in mortality, compared with ECC (aRR 0.47, 95% CI 0.20 to 1.11, 1 study, 270 babies, low certainty). DCC may also reduce the number of babies who die or have neurodevelopmental impairment in early years (aRR 0.61, 95% CI 0.39 to 0.96, 1 study, 218 babies, low certainty). There may be little or no difference in: severe IVH; all grades IVH; PVL; CLD; maternal blood loss ≥ 500 mL, assessed as low certainty mainly due to serious imprecision.

Delayed cord clamping (DCC) with immediate neonatal care after cord clamping versus umbilical cord milking (UCM) (three studies, 322 babies and their mothers) and UCM versus early cord clamping (ECC) (11 studies, 1183 babies and their mothers)

There are insufficient data for reliable conclusions about the comparative effects of UCM compared with delayed or early clamping (mostly low or very low certainty).