Labour is traditionally divided into three artificial stages. The first stage consists of the woman in labour experiencing regular, painful contractions, with progressive cervical dilation from 4 cm until fully dilated at 10 cm. The second stage continues from when the woman's cervix is fully dilated until the birth of the baby. The third stage is the period of time between the birth of the baby and the delivery of the placenta and membranes (National Institute for Health and Care Excellence (NICE), 2017). There is always some blood loss during this third stage, and healthcare professionals' care aims to reduce excessive blood loss. In clinical practice, care is managed by two distinct clinical approaches: active management and expectant management, often referred to as physiological management (Begley et al, 2019). This paper reviews evidence on the effectiveness of these approaches and the implications of this research for current practice guidelines for women at low risk of postpartum haemorrhage who choose to birth in a midwife-led unit or home birth setting and want to experience a birth with minimal intervention.
Current use of active and expectant management
Active management aims to accelerate delivery of the placenta to reduce blood loss. A prophylactic uterotonic drug (exogenous oxytocin) is given to accelerate the contractility of the uterus, to cause the placenta to separate from the uterus wall more quickly. Other components of active management include delayed cord clamping, cutting of the cord and controlled cord traction (Royal College of Obstetricians and Gynaecologists (RCOG), 2016; NICE, 2017). Delayed cord clamping between 1 and 3 minutes after birth can have positive effects on the infant such as increasing birth weight, haemoglobin concentration and iron reserves up to 6 months after birth (McDonald et al, 2013). Iron reserves are important, as childhood anaemia is associated with negative effects on child development (Khan, 2018). Delayed cord clamping can also have a positive effect on a baby's neurodevelopment (Mercer et al, 2020). The benefits of delayed cord clamping for at least 30 seconds in babies born preterm may also reduce the number of preterm babies who die before discharge from hospital (Rabe et al, 2019), as well as also having a positive effect on their neurodevelopment (Nasasimhan et al, 2019).
The main principle of expectant management is to support the woman during labour and birth so her body can produce optimal levels of endogenous oxytocin. The healthcare practitioner also watches and waits for signs of placental separation, after which the placenta is delivered spontaneously or with the aid of gravity and maternal pushing (NICE, 2017). Anything that interferes with this oxytocin release by the woman's body will reduce the effectiveness of a physiological third stage of labour (Inch, 1985; Buckley, 2004; 2015; Fry, 2007). Hence, expectant management would not be appropriate in this circumstance. Consequently, expectant management of the third stage of labour is only appropriate for women who have had a normal physiological birth, which is defined by the World Health Organization (WHO, 1997) as one where labour occurs spontaneously and the woman is at low risk of obstetric complications at the start of labour and remains low risk throughout labour and birth. The baby is born spontaneously and in the head-down position between 37 and 42 completed weeks of pregnancy. After birth, the woman and baby are in good condition, meaning that there are no concerns regarding the woman's or the baby's physiological wellbeing after the birth
In practice, although not recommended in clinical guidelines and perhaps not intentional by the practitioner but as a result of clinical circumstances, a mixed management approach often occurs (Harris, 2006; Winter et al, 2007; Begley et al, 2019). This approach combines some of the components of both expectant and active management, but does not completely contain all the components of either (Begley et al, 2019). For example, this mixed approach may involve early uterotonic administration, cord clamping after pulsation ceases and controlled cord traction, or may involve delayed uterotonic administration until cord pulsation ceases, then cord clamping and controlled cord traction.
At present, active management of the third stage of labour is the more common approach in the UK and Ireland, as in most high-income countries (Begley et al, 2019). However, expectant management is sometimes practiced by midwives in the UK, other northern European countries and New Zealand, mainly in midwife-led units and home birth settings (Fry, 2007; Kanikosmy, 2007; Blackburn, 2008; Begley et al, 2009) or by midwives providing continuity of carer models of maternity care (Homer et al, 2017; Sandall et al, 2016).
Midwives who work in midwife-led units or home birth settings and women who choose to birth in these setting are more likely to value minimal intervention (Shallows, 2003; Walsh, 2012). Women who choose to birth in midwife-led units and home birth settings are also more likely to be at low risk of obstetric complications; hence the use of expectant management may be more suitable for these women. In low-income countries, expectant management is more commonly practised for community and home births (Begley et al, 2019).
Studies have shown that when women are offered expectant management as a feasible option, they will choose it (Rogers et al, 1998; Dixon et al, 2009; 2013; Fahy et al, 2010; Begley et al, 2011a; Gottvall et al, 2011; Davis et al, 2012; Laws et al, 2014; Monk et al, 2014; Grigg et al, 2017; Kataoka et al, 2018). Furthermore, the UK National Collaborating Centre for Women and Children's Health (2014) acknowledges that some women may want to experience a birth with minimal intervention and request a physiological third stage of labour.
Review of guidance on management of the third stage
Current national and international guidance on management of the third stage of labour advocates active management (International Confederation of Midwives (ICM) and the International Federation of Gynaecology and Obstetrics (FIGO), 2003; 2006; WHO, 2012; 2018; RCOG, 2016; NICE, 2017; Royal College of Midwives (RCM), 2018). This is because studies on active management have found a reduction in rates of and treatment for postpartum haemorrhage compared with expectant management in women identified as both at high and low risk of postpartum haemorrhage (Prendiville et al, 1988; 2000; Begley, 1990; Thilaganathan et al, 1993; de Groot et al, 1996; Rogers et al, 1998; Begley et al, 2011b; 2015). Postpartum haemorrhage is defined as blood loss of ≥500 ml from the vaginal tract after the birth of the baby (WHO, 2012).
This article examines the robustness of the research evidence underpinning third stage of labour guidelines (WHO, 2012; 2018; RCOG, 2016; NICE, 2017; RCM, 2018). A number of concerns are highlighted about the appropriateness of some aspects of the guidance for women at low risk of postpartum haemorrhage, who choose to birth in a midwifery-led unit or home birth setting who want to experience a labour and birth with minimal intervention. Indications for future research are also suggested.
Research studies informing current guidance
A Cochrane systematic review by Begley et al (2011b) included Begley (1990), Prendiville et al (1988), Rogers et al (1998) and Thilaganathan et al (1993) research studies, that compared active with expectant management for women expected to give birth vaginally within an obstetric-led hospital unit. The findings were that when generalising across all women, irrespective of their risk factors for bleeding, active management reduced maternal blood loss after birth and the incidence of ‘minor postpartum haemorrhage’ (estimated blood loss of 500–1000 ml) or ‘severe postpartum haemorrhage’ (estimated blood loss of ≥1000 ml), compared to expectant management. Active management also reduced the treatment needed for excessive blood loss, leading to a reduction in the use of therapeutic oxytocic drugs, anaemia and blood transfusion. The duration of the third stage of labour was also found to be shorter with active management.
However, for women identified as at low risk of postpartum haemorrhage, Begley et al (2011b) did not identify any statistically significant difference for severe postpartum haemorrhage or incidences of anaemia. This is important, as research studies have found that well-nourished, healthy women are able to compensate for a blood loss of up to 1000 ml (Cunningham and Williams, 2001; Blackburn, 2008; Oishi et al, 2017). Therefore, active management may be of limited value to this group of women. The most recent Cochrane review, Begley et al (2019), also commented that the potential harms of active management are more concerning in women at low risk of postpartum haemorrhage. This was because the potential benefits of reducing severe postpartum haemorrhage through active management in randomised controlled trials were less evident for this group. Begley et al (2019) advocated further studies comparing active with expectant management in women at low risk of postpartum haemorrhage, to confirm if there is a difference in severe bleeding.
Begley et al (2011b; 2015; 2019) found that compared with expectant management, active management showed a statistically significant increase in the need for postnatal analgesia and in women returning to hospital as an outpatient, because of bleeding and a decrease in the baby's birth weight. This reduction in birth weight was possibly caused by the practitioner clamping the umbilical cord early, reducing the volume of placental blood transfusion. Research has shown that babies can gain up to 214g in the first 5 minutes following birth if the cord is left unclamped (Farrar et al, 2010).
Cutting the cord before it stops pulsating has been found to increase the risk of iron deficiency anaemia in term infants (Chararro et al, 2006; Anderson et al, 2011). As a result of these adverse effects, NICE (2017), the RCM (2018) and the WHO (2014; 2018) recommend not clamping and cutting the cord for at least 1 minute after the birth in an actively managed third stage.
The findings of Begley et al (2011b) informed the RCOG (2016) and WHO (2012; 2018) third stage of labour guidelines and recommendations. Begley et al (2015) subsequently updated and replaced the Cochrane review. Their updated 2015 review informed the RCM (2018) third stage of labour practice recommendations. This review has recently been updated and replaced again by Begley et al (2019). Despite conducting more up to date literature searches, no new studies were identified in these updated reviews that compared active and expectant management of the third stage of labour in women identified as at low risk of bleeding or its effects, or in women irrespective of their risk of bleeding. Therefore, in these updated reviews, the recommendations for these women regarding active compared with expectant management remain the same as the Begley et al (2011b) initial review.
The randomised controlled trials by Prendiville et al (1988), Rogers et al (1988) and Thilaganathan et al (1993), which were included in the Begley et al (2011b) Cochrane review, were also used to inform the NICE (2017) guidance regarding active versus expectant management approaches during the third stage of labour for women at low risk of obstetric complications. Another randomised controlled trial by de Groot et al (1996) compared intramuscular oxytocin with a placebo and was also used to inform the recommendation of active management within the NICE (2017) guidelines.
Assessing and evaluating the research evidence
The evidence regarding third stage of labour care has not only informed national and international third stage of labour guidelines and recommendations, but also influenced local maternity guidelines. However, the evidence underpinning these guidelines is not without question, particularly for women at low risk of postpartum haemorrhage who choose to birth in a midwifery-led unit or home birth setting and want to experience a birth with minimal intervention.
Appropriateness of comparisons/evidence
The Cochrane reviews by Begley et al (2011b; 2015; 2019) included four randomised controlled trials, consisting of 4829 women in total, that compared active and expectant management (Prendiville, 1988; Rogers et al, 1998; Begley, 1990; Thilaganathan et al, 1993). Although in all the trials, women were identified as healthy pregnant women expected to give birth vaginally within a hospital obstetric-led unit, only three out of these four trials identified women at the beginning of the study that were classified as being at low risk of bleeding or its effects (Begley, 1990; Rogers et al, 1998; Thilaganathan et al, 1993).
The Begley et al (2011b; 2015; 2019) reviews noted a lack of rigour, although some of the randomised controlled trials included in the reviews did present power calculations and postpartum haemorrhage rates (Prendiville et al, 1988; Rogers et al, 1988; Begley et al, 1990). However, the trials may have been biased towards active management; for example, Prendiville et al (1988) included women identified as at high risk of postpartum haemorrhage who were not suitable for expectant management. This is because expectant management is only appropriate for women who are at low risk of postpartum haemorrhage and have had a normal physiological birth. Furthermore, many of the women in these trials developed risk factors for postpartum haemorrhage during labour. By the time they reached the third stage of labour, they were at high risk of postpartum haemorrhage despite being identified to be of low risk of postpartum haemorrhage on trial entry (Begley, 1990; Rogers et al, 1998). Again, these women should also not have been included in studies comparing active with expectant management, as expectant management was clearly not appropriate for them. In Begley (1990), 27% of women in both active and expectant management groups had their labour induced, accelerated or augmented using synthetic oxytocin, and as a result clearly did not have a normal physiological birth. Additionally, in Prendiville et al (1988), 24% of women who received expectant management had their labour inducted or augmented by oxytocin.
However, Begley (1990) found no significant difference between the use of synthetic oxytocin for labour induced, accelerated or augmented in the active and expectant management group (P=0.9). Additionally, the number of women who received oxytocin to induct or augment their labour was similar in expectant and active management groups in Prendiville et al (1988). Nevertheless, the use of oxytocin in labour can interfere with a woman's physiological production of oxytocin (Buckley, 2009; 2015; Uvnaas Moberg, 2011). Consequently, the use of synthetic oxytocin for labour induction, acceleration or augmentation would be riskier, with regards to the increasing the risk of bleeding during the third stage of labour with expectant management. Therefore, including these women in both groups in these trials is not a fair test of expectant management and skews findings towards active management. NICE (2017) also state that the use of oxytocin in labour is among the risk factors for postpartum haemorrhage.
Begley (1990), Prendiville et al (1988) and Rogers et al (1998) included women who had episiotomies, and although Begley (1990) showed that there was no significant difference in episiotomies performed between active and expectant management groups and Rogers et al (1998) showed that the episiotomy rates in active and expectant management groups were similar (11.6% compared with 12.3%), episiotomies are identified as a risk factor for postpartum haemorrhage (NICE, 2017). This is because of the increased risk of bleeding from the episiotomy site, as the third stage of labour can potentially take longer with expectant compared with active management, consequently delaying repair of the episiotomy site and resulting in increasing blood loss. Episiotomies would be riskier because of the increasing the risk of bleeding during the third stage of labour with expectant management. Including these women who had an episiotomy in both groups is again not a fair test of expectant management and so skews findings to favour active management.
Additionally, de Groot et al (1996) informed NICE (2017) third stage guidance, but did not compare active with expectant third stage of labour management. This study compared intramuscular oxytocin or a placebo and no other component of active or expectant management was reported. This questions the relevance of Begley (1990), Rogers et al (1998), Prendiville et al (1988) and de Groot et al (1996) when examining active versus expectant management of the third stage of labour for women identified as low risk of postpartum haemorrhage who have had a normal physiological birth.
Begley et al (1990) reported the use of intravenous ergometrine as the uterotonic drug for women having active management. The use of ergometrine alone is no longer recommended in current practice for active third stage of labour management, which calls into question the contemporary relevance of this study. The smaller scale trial by Thilaganathan et al (1993), including women at low risk of postpartum haemorrhage, did not present a power calculation or postpartum haemorrhage rates and presented only rounded mean blood loss figures. Therefore, this study may have reported biased results.
The effect of midwives' experience in third stage management
In all these trials, active management of the third stage of labour was routine. As a result, midwives were more experienced in conducting active management. The experience of healthcare professionals in conducting third stage management approaches is important in reducing blood loss during the third stage of labour or shortly after (Dixon et al, 2009; Fahy et al, 2010; Begley et al, 2011a; Davis et al, 2012; Laws et al, 2014; Monk et al, 2014; Grigg et al, 2017). This is evident in Begley (1990) and Rogers et al (1998), who found that midwives who did not routinely use expectant management needed time to become familiar with it. Once midwives in these studies were familiar with expectant management, blood loss during the third stage of labour was reduced. In Begley (1990), the postpartum haemorrhage rate in the expectant management group dropped during the trial from 21% in the pilot study to 12% in the first 4 months, and 7% in the last 6 months of the main study.
Blood loss in these trials was assessed mainly through visual estimation from the healthcare practitioner. It is widely acknowledged that blood loss during the third stage of labour or shortly after birth is difficult to assess accurately and is frequently under- or over-estimated by practitioners (Razvi et al, 2008; Schorn, 2010). Additionally, it was not possible to double blind assessment of blood loss in these trials. Therefore, the estimation of blood loss was open to considerable observer/assessor bias and if active management was the routine third stage of management approach in these studies, this observer/assessor bias may have skewed findings towards active management.
Place of birth and third stage of labour
All of the trials focused on women giving birth in hospital obstetric-led units. Place of birth is important, as more recent cohort studies conducted outside of the UK have shown that women who birthed in midwife-led as opposed to obstetric-led units, experienced reduced blood loss during the third stage of labour or shortly after with expected management as opposed to active third stage management (Fahy et al, 2010; Davis et al, 2012). Additionally, a lower incidence of postpartum haemorrhage has been found in midwife-led units, despite increased expectant management and reduced active management, in comparison to obstetric-led units (Dixon et al, 2009; Laws et al, 2014; Monk et al, 2014; Grigg et al, 2017). The randomised controlled trial by Begley et al (2011a) found that despite an increase in expectant management in a midwife-led unit compared to obstetric-led units, there was no statistically significant difference in estimated mean blood loss during the third stage of labour or shortly after, or in the incidence of postpartum haemorrhage.
Strength of the evidence
The reliability, validity and generalisability of these trials informing NICE (2017), RCOG (2016), WHO (2012; 2018) and RCM (2018) third stage of labour guidance is questionable, particularly for women at low risk of postpartum haemorrhage who choose to birth in midwife-led units or home birth settings. In fact, the National Collaborating Centre for Women's and Children's Health (2014) graded the quality of evidence supporting NICE's (2017) guidelines regarding active compared with expectant management and incidence of postpartum haemorrhage as low (Prendiville et al, 1988; de Groot et al, 1996; Rogers et al, 1998) and severe postpartum haemorrhage as very low (Prendiville et al, 1988; de Groot et al, 1996; Rogers et al, 1998). This was as a result of the risk of bias, inconsistencies and indirectness in the studies.
The latest Cochrane review (Begley et al, 2019) graded the quality of evidence examining the incidence of postpartum haemorrhage as low (Begley, 1990; Rogers et al, 1998) and the quality of evidence examining the incidence of severe postpartum haemorrhage (Begley, 1990; Rogers et al, 1998) and haemoglobin <9 at 24 hours (Thilaganathan et al.1993) as very low. The quality of evidence regarding mean maternal blood loss (Begley, 1990; Thilaganathan et al, 1993; Rogers et al, 1998) and maternal blood transfusions was also graded as low. However, the quality of evidence examining the use of therapeutic uterotonic during the third stage and/or within the first 24 hours was graded as moderate (Begley, 1990; Thilaganathan et al, 1993; Rogers et al, 1998). This was again as a result of the risk of bias, inconsistencies and indirectness in the studies.
Conclusions
Practice guidelines and recommendations by NICE (2017), the RCM (2018), the RCOG (2016) and the WHO (2012; 2018) recommend active management for all women. However, this may not be appropriate for women at low risk of postpartum haemorrhage giving birth in a midwife-led unit or at home who want to experience a birth with minimal intervention. This is because of limitations in the reliability, validity and generalisability of the evidence informing these practice guidelines.
Expectant management may also be a suitable choice for women at low risk of postpartum haemorrhage who choose to give birth in a midwife-led unit or home birth setting and want minimal intervention. However, more research into third stage management practices in midwife-led and home birth settings and midwives' perspectives on these issues are needed. This research could help to clarify the extent to which the apparent risks of expectant management might be a consequence of limited training, confidence or experience of the practitioner or organisational culture, rather than just a consequence of the management approach.
Key Points
- There is always some blood loss during the third stage of labour and care aims to reduce excessive blood loss
- In clinical practice, care during the third stage of labour is managed by two distinct clinical approaches: active and expectant management
- At present, active management of the third stage of labour is the more common approach in the UK, Ireland and most high-income countries
- Expectant management of the third stage of labour is only appropriate for women who have had a normal physiological birth and want to experience birth with minimal intervention
- The reliability, validity and generalisability of the randomised controlled trials informing third stage of labour guidance is questionable, particularly for women at low risk of postpartum haemorrhage who choose to birth in midwife-led units or home birth settings
CPD reflective questions
- Do you think third stage of labour practice guidelines and recommendations present robust evidence regarding third stage of labour management approaches for women at low risk of postpartum haemorrhage who birth in midwife-led units or at home?
- How have you come to this conclusion?
- Do you think current third stage of labour guidelines and recommendations can be applied to women at low risk of postpartum haemorrhage who birth in midwife-led units or at home and want to experience a normal with minimal intervention?
- How have you come to this conclusion?
- What influences your own third stage of labour practice?