The third stage of labour begins at the time the newborn is delivered and continues until the placenta and membranes are expelled. Immediately after birth, the newborn remains attached to the mother and placenta via the umbilical cord. The blood volume transferred from the placenta to the newborn during this time is known as placental transfusion. Active management of the third stage of labour is widely practiced and involves a package of interventions aimed at reducing the incidence of postpartum haemorrhage (PPH) (Prendiville et al, 2009), which is a leading cause of maternal morbidity (Lewis, 2011).
The package of interventions include: routine administration of prophylactic uterotonic drugs, early cord clamping (ECC) and cutting, and controlled cord traction (National Institute for Health and Care Excellence (NICE), 2007), although the timing of each component varies. Active management can lead to a reduced risk of PPH, but it is crucial to establish which individual component reduces the risk or if indeed the full ‘package’ is necessary (Soltani et al, 2010; McDonald et al, 2013).
The International Confederation of Midwives and the International Federation of Gynaecology and Obstetrics (ICM/FIGO) (2006) and, more recently, the World Health Organization (WHO) (2012) updated their guidelines on PPH prevention. The above organisations refer to the benefits of delayed cord clamping (DCC), and the WHO (2012) strongly recommends DCC to be performed 1–3 minutes after birth, while simultaneously initiating newborn care. They also strongly recommend that ECC (<1 minute after birth) be only carried out if the newborn is asphyxiated and in need of immediate resuscitation, although the quality of evidence underpinning these recommendations was described as moderate. Weeks et al (2013) postulate that the optimal management of asphyxia caused by cord compression during labour is cord decompression by delivery and DCC to allow oxygenated placental blood transfusion to the newborn However, due to mandatory transfer of any compromised newborn, to a designated resuscitation area, no research, to date, has been possible. Even though to many, while it may appear counterintuitive to sever this lifeline, the lack of high quality research evidence, means that at present, the status quo prevails and mandatory transfer to areas of resuscitation is seen as the safest option.
There is evidence to suggest that the effects of ECC and DCC differ between term and preterm infants. A systematic review conducted by Rabe et al (2012) sought to examine the effects of ECC and DCC on 738 newborn babies and their mothers, who gave birth between 24 and 37 weeks gestation.
The authors concluded that while ECC allows the newborn to be transferred to the neonatologist, DCC enables increased placental transfusion found to be linked with a smaller need for transfusions due to anaemia, less intraventricular haemorrhage (all grades) and a lower risk of developing necrotising entercoloitis. However, there were insufficient data for reliable conclusions about the comparative effects on the primary outcomes of this review and more research is needed into this area. This review will therefore focus on newborn and maternal outcomes of term infants.
Supporting delayed cord clamping
As midwives, we must practise using the best available evidence (Nursing and Midwifery Council (NMC), 2008). Hence, this evaluation seeks to explore the evidence surrounding the timing of umbilical cord clamping with a view to recommend and implement DCC as part of everyday, routine care. In effect, fundamentally changing the view towards ECC, which has persisted in spite of the increasing evidence in favour of DCC.
Farrar et al (2011) suggest that in term births, placental transfusion can contribute between one-quarter and one-third of the total neonatal blood volume; and up to 60% more red blood cells (Mercer, 2001; McDonald, 2007; Palethorpe et al, 2012). Farrar et al (2011) believe that the duration of placental transfusion is important, as it is the period of transition between fetal and newborn circulation. In utero, only 8% of blood volume goes to the lungs. As these inflate and expand, the newborn may need to redirect blood flow to support the increased respiratory circulation. DCC is thought to lead to more efficient cardiopulmonary adaptation and greater perfusion of organs (Zaramella et al, 2008; Mercer et al, 2010). Once the newborn circulation becomes established, the additional blood volume is absorbed, the red cells broken down and the iron component of haemoglobin is absorbed and stored for future growth and development (Farrar et al, 2011).
Adverse effects of early clamping
Some hypothesise that ECC has detrimental effects on the newborn (Ceriani Cernandes et al, 2006; Mercer et al, 2008). It has been argued that ECC reduces the necessary blood volume substantially enough to cause hypovolemic damage by redirecting blood and impeding capillary perfusion, which results in inflammation and a increased risk of infection in vulnerable newborns (Jahazi et al, 2008; van Rheenen, 2011). In addition to a reduced blood volume, there is also a decrease in red cell mass, iron levels and a loss of hematopoietic stem cells (Ende and Reddi, 2006), therebyincreasing the newborn's risk of developing several blood disorders and type 2 diabetes (Mercer, 2001; Mercer and Skovgaard, 2002; Chaparro et al, 2006).
ECC has been linked with an increased occurrence of feto-maternal transfusion, as a larger blood volume remains in the placenta. Since the introduction of Rhesus D immunoglobulin prophylaxis, the risk appears to be minimal for those with Rhesus negative blood type. However, there appears to be a paucity of research regarding this topic, since the initial findings suggest that there may be a reduction in feto-maternal transfusion if cord clamping is delayed (Lapido, 1972). If the increased risk is associated with the volume of blood left in the placenta, then by leaving the cord unclamped on the maternal side and allowing free drainage, it would reduce the retained placental blood volume, and potentially decrease the risk of feto-maternal transfusion.
Over-transfusion in delayed cord clamping?
A concern of many midwives and obstetricians is that DCC can lead to over-transfusion of the newborn infant, something which can be further exacerbate if a uterotonic drug is administered prior to cord clamping. The concept of over-transfusion appears to be based on secondary analysis of a study conducted by Saigal and Usher (1977). They identified more newborns who were symptomatic for polycythaemia in the DCC group, but only two of 11 newborns had a haematocrit level above the clinically significant level of 70%—strongly suggesting that for the majority of babies, this polycythaemia is benign. The results may have been confounded as the sample consisted of preterm newborns, who are known to have an increased risk of developing jaundice (Darcy, 2009; Wallenstein and Bhutani, 2013). Furthermore, hypertension in the mother, which is known to increase the risk of developing polycythaemia in newborns, was not addressed, reducing the reliability and validity of the results of Saigal and Usher's (1977) study. Yao et al (1968) demonstrated that newborns receive a maximum of 90 ml/kg blood, following administration of a uterotonic drug administration, which is the normal physiological volume, regardless of when clamping occurs. This has been verified by more contemporary research (Ceriani Cernandes, 2006; Hutton and Hassan, 2007; Farrar et al, 2011).
A recent prospective observational study conducted in the UK by Farrar et al (2011) aimed to measure placental transfusion in term births, when the umbilical cord was left intact. The study also sought to examine the effect of administering a uterotonic drug (oxytocin) either before or after cord clamping. A small sample of 26 newborns, a mixture of vaginal (n=13) and caesarean births (n=13), had their weights recorded every 2 seconds after being wrapped post-delivery. Placental transfusion was then calculated by conversion of weight to volume (1 ml of blood=1.05 g). There was no statistical difference in placental transfusion duration or volume between vaginal and caesarean births. The data appeared to show that the time at which net placental flow ceased for most newborns was at 2 minutes of age. However for some newborns, placental blood flow through the umbilical cord continued for up to 5 minutes. As the study sample was small, caution should be taken when interpreting the findings. The volume of placental transfusion did not appear to be influenced by the use of intramuscular oxytocin, whether it had been given before or after the cord had been clamped.
The mean difference in weight overall was 116g (95% confidence interval (CI), 72–160 g) using B-spline (a generalisation of the Bezier curve); however, by inspecting the graphs the mean difference in weight was 87g (95% CI, 64–111 g). The authors had originally planned to examine the data from direct graph inspection, but due to artefacts that occurred when wrapping the newborn to maintain temperature and then placing them on the scales, the start weight became difficult to determine. In an attempt to overcome this problem they used B-spline to smooth the data. An advantage of this approach is that it uses the early weight measurements to estimate the probable start weight, but these early weight measurements are ignored on inspection of the graph, which views the starting weight as the weight when the artefact ceases. Direct inspection of the graph is likely to underestimate true placental transfusion, as the start weight is often measured 20 seconds or more after birth resulting in some transfusion already occurring before weighing is commenced, but B-spline may overestimate if smoothing is incomplete. Although there are differences in the mean weight difference with both these approaches, the data consistently demonstrates that placental transfusion is likely to contribute a significant proportion of the blood volume of the newborn, and the true value is somewhere between the two estimates.
Further evidence for implementing delayed cord clamping
Hutton and Hassan (2007) conducted a systematic review and meta-analysis of controlled trials to provide an evaluation of the potential benefits and drawbacks of DCC versus ECC in term newborns. The review included 15 studies, with data collected from 1912 newborns considered suitable for metaanalysis, of an original sample of over 2000. Eight of the studies were randomised controlled trials (RCTs) and seven were controlled trials (CTs), conducted in several countries with contrasting perinatal mortality rates. The studies considered term infants born vaginally or by caesarean section, approximately 96 and 4%, respectively. ECC was classified as occurring within 60 seconds of birth, while DCC was classified as from 1 minute or after the cessation of cord pulsation. In order to be considered, the studies had to meet the inclusion criteria with studies reporting on at least one pre-specified outcome, including: incidence of jaundice, the need for phototherapy or an increase in haematocrit level to over 65%, which would be consistent with polycythaemia.
Polycythaemia
Newborns in the DCC group were found to be at higher risk of developing polycythaemia for up to 48 hours (RR, 3.82; 95% CI, 1.11–13.21), but none were symptomatic, needed treatment or admission to the neonatal unit (benign polycythaemia). After conducting a sensitivity analysis, which only included high-quality studies, no statistical differences were found between the two groups, providing yet more evidence for the implementation of DCC.
The results also showed the benefits of DCC at ages 2–6 months, as these babies had improved haematologic status, measured as haematocrit (weighted mean difference (WMD), 3.70%; 95% CI, 2.00–5.00%). In addition, improved iron status was observed, measured as ferritin concentration (WMD, 17.89; 95% CI, 16.58–19.21) and stored iron (WMD, 19.90; 95% CI, 7.67–32.13) and a reduction in the risk of developing childhood anaemia (relative risk (RR), 0.53; 95% CI, 0.40–0.70).
Iron status
Even though the Hutton and Hassan (2007) trial included studies from developing countries, many of which have high rates of maternal anaemia, it is important to note that these findings are significant for NHS Trusts in the UK. These organisations provide care for women from diverse cultural, ethnic and socioeconomic backgrounds as well as teenage mothers, all of whom are at increased risk of developing iron-deficiency anaemia. This review, in line with others (Emhamed et al, 2004; Ceriani Cernandes et al, 2006; Van Rheenen and Brabin, 2006; Andersson et al, 2011), shows that DCC reduces the incidence of the newborns developing anaemia even if they are born to anaemic mothers, presenting clear, well-founded evidence for the increased use of DCC to become a routine part of the third stage of labour.
The Hutton and Hassan (2007) systematic review and meta-analysis had well defined inclusion criteria and clear aims of outcomes against which their study was examined. The authors carried out extensive searches of both published and unpublished work that was not restricted by language barriers, thereby avoiding possible publication bias. Hutton and Hassan also applied techniques to their review process that included the use of standardised forms, which further reduced error and improved the accuracy of their findings (Rees, 2011). Similar methodological and statistical efforts were taken to ensure robust data and greater confidence in the findings of the study.
A more recent RCT conducted in Sweden by Andersson et al (2011) sought to examine the impact of ECC and DCC on the iron status of newborn infants up to 4 months of age. The sample consisted of 400 babies from low risk pregnancies who were born at full term. During the antenatal period, women who met the inclusion criteria were given information about the trial. Those who met the criteria for inclusion were again informed of the trial, when admitted in labour by the midwife attending. This written consent, wherever possible, was obtained from both parents thereby meeting the ethical principle of informed consent. When comparing the baseline data, there were no significant differences found between the ECC or DCC groups, with respect to maternal characteristics or newborn baseline data.
Even though participants were not randomly selected for the trial, they were randomly allocated an intervention group using a number generator programme, which reduces bias. The design of the study prevented both mother and clinician performing the intervention from being blinded, making the impact of lack of blinding difficult to assess, an issue in many trials (Ceriani Cernandes et al, 2006; van Rheenen et al, 2007; Al-Tawil et al, 2012). However, attempts were made to reduce detection bias by blinding the physicians performing newborn examinations and those collecting and analysing blood samples.
It was routine practice to perform ECC in the hospital where the Andersson et al (2011) trial was conducted, so implementing DCC would be considered as deviating from the norm and this may have introduced bias in the way that the intervention was implemented, an issue highlighted in the Cochrane review conducted by McDonald et al (2013). Although an outcome not measured in this study, it is reasonable to assume that the knowledge of allocation could have had a subconscious effect, which might have altered subjective outcomes such as the estimation of maternal blood loss.
In the Andersson et al (2011) trial, the ECC group had a significantly higher prevalence of iron deficiency when compared with the DCC group, 6 and 1%, respectively. In order to increase the accuracy of their findings the log ratio of soluble transferrin receptor to ferritin (Malope et al, 2001) was calculated. A significant difference in the groups was found, indicating that DCC leads to a superior iron status, which has a direct impact on infant health and wellbeing. Low iron levels in infants can affect brain development and correlates with various cognitive and behaviour problems (Lozoff et al, 2006; Beard, 2007; Thomas et al, 2009). Szajewska et al's (2010) meta-analysis shows the relationship between non-anaemic iron deficiency and neurodevelopment, with significant improvements in psychomotor development in non-anaemic infants given iron supplements.
As iron is stored mainly in bone marrow and the liver, measurements using blood or serum can only be considered indirect estimates. As a consequence, such measurements have several limitations including: sensitivity to inflammation, poor sensitivity for detecting mild iron deficiency (haemoglobin), mean cell volume, mean cell haemoglobin concentration and a lack of reliable reference intervals (Domellöf, 2007). By including a range of iron status indicators, in addition to ferritin, the Andersson et al (2011) trial offers seemingly strong evidence pertaining to iron status and DCC, making the results more valid.
Infant diet also affects iron stores throughout infancy (Chantry et al, 2007) and prolonged exclusive breastfeeding may increase the risk of iron deficiency (Fewtrell et al, 2011), although is not possible to control for the potentially modifying effects of breastfeeding or iron-fortified formula. Andersson et al (2011) suggest that it is easier to prevent iron deficiency through umbilical cord clamping policies rather than improvements in dietary intakes of the whole population, as it is a free method and is particularly beneficial where there is limited access to drugs and obstetric care (Eichenbaum-Pikser and Zaslaff, 2009).
A limitation of the Andersson et al (2011) study, as with many pertaining to cord clamping, was that it was not made possible to measure the amount of blood transfused from the placenta to the newborn. This leads to the presumptuous conclusions that the greater time allowed for placental transfusion, the greater the volume transfused, resulting in conclusions being drawn from time allocated rather than actual volume transfused. However, there were several indications that there was indeed increased placental transfusion in the DCC group with a higher mean birth weight of 96g, in line with the Farrar et al (2011) trial, less blood retained in the placenta, and higher packed cell volumes and haemoglobin levels at 2 days postnatal.
In addition, the inclusion criteria limited the trial to those who were deemed to have low-risk pregnancies and term births. This is not an accurate representation of the population causing the findings to be biased and be considered to lack generalisability as other perinatal risk factors including maternal diabetes, intrauterine growth restriction and preeclampsia were not considered.
Height of newborn in relation to placenta
An unaccounted for, and potentially confounding and limiting factor in the Andersson et al (2011) study was that the exact level the newborn was held was not measured, nor for how long it was held at that height. This could unknowingly confound results and reduce reliability and validity, as the level the neonate is held in relation to the placenta can affect speed and amount of placental transfusion (Erickson-Owens, 2011; Mercer and Erickson-Owens, 2012). This is a common area for improvement in many studies, as standardisation in trials, and even throughout all trials would potentially generate more generalisable, reliable results with greater rigour, by controlling the effect of gravity and therefore minimising its confounding effect.
Previous findings from studies have suggested that lifting or lowering the newborn by 40 cm from the introitus will affect placental transfusion, but if the newborn is only raised or lowered by 10–20 cm then both duration and volume of placental transfusion is not substantially affected (Gunther, 1957; Yao and Lind, 1969).
Although small, the study by Farrar et al (2011) supports these findings, suggesting that raising the newborn onto the mothers-abdomen does not impact placental transfusion. This would also enable skin-to-skin contact, which is good for neonatal thermoregulation, bonding and the initiation of breastfeeding, all of which are beneficial to both mother and newborn.
Palethorpe et al (2012) sought to evaluate studies in relation to the effect of positioning and the influence of gravity on placental transfusion; however, no studies were identified for comparison and they concluded that there was a need for well designed RCT to be conducted to ascertain the effects of gravity on placental transfusion.
Conclusion
DCC consistently improves both short and longterm haematological values and the iron status of newborn infants, which has been found to improve neurological and cognitive development. In a population where there is an increasing incidence of iron deficiency anaemia, DCC provides an easy, inexpensive and appropriate method to reduce anaemia in the newborn. DCC allows adequate time for full placental transfusion, supporting the transition from intrauterine to extra-uterine life, reducing hypovolemic damage and its related effects. DCC can be facilitated in conjunction with skin-to-skin contact with the mother, enabling bonding, breastfeeding and assisting with newborn thermoregulation. As midwives, we should practice using the best available evidence, striving for the best outcomes in an effort to do no harm. As cord clamping is an intervention, it is the right of women to be fully informed as to their choices in this respect and all health professionals should seek to gain consent before ECC is initiated in preference to DCC; and accordingly, documented in the maternity record. With such compelling evidence in mind, aa review of local and national protocols is most definitely warrented and DCC should become part of routine practice without any furtherdelay.