For many years, early administration of a prophylactic uterotonic has been included in the package of care known as active management of the third stage of labour (AMTSL), however, this became routine practice without rigorous Assessment. Trials to date have been less than robust as they fail to report data for all maternal and neonatal outcomes with inconsistencies in timing, route and type of uterotonic used. With recent changes in practice regarding delayed cord clamping (DCC) this raises the question of when is the optimal timing for administration of an uterotonic. Most would agree that reliable evidence should guide practice and this ought to include evidence concerning potentials for both benefit and harm so as midwives we need to challenge when the evidence is insubstantial (Aslam, 2000; Brayford, Chambers, Boath and Rogers, 2008; Cluett, 2006; Soltani, 2008). In order to understand the advantages or disadvantages and to ensure evidence based practice this research proposal highlights the need for a randomised controlled trial (RCT) as the timing of uterotonic may have significant implications on the well-being of the women and neonate.

Pamela Spencer first described AMTSL in 1962 but it was not widely adopted until an RCT revealed a significant reduction in the incidence of post-partum haemorrhage (PPH) compared with expectant management (Prendiville, Harding, Elbourne and Stirrat, 1988). However, inconsistencies in practice were prevalent as a survey by Festin et al. (2003) found global variation, especially in the type, timing and route of uterotonic and controlled cord traction (CCT). Despite an international joint policy statement the same year outlining the components of AMTSL (International Confederation of Midwives and the International Federation of Gynaecology and Obstetrics, 2003), similar findings of variations were found in a later European survey (Winter et al., 2007).

Management of the third stage has been the subject of several critical reviews (Gyte, 1994; Hooper et al., 2015; Maughan, Heim and Galazka, 2006; McDonald, 2007; Prendiville, 1996; Soltani, 2008) with Gyte, Hooper et al., McDonald and Soltani all calling for more research and conclusive evidence. In 2006, Gyte further discussed a delayed active management technique suggesting administering the uterotonic after the cord has finished pulsating and been clamped and cut, followed by CCT. However, she does again acknowledge that outcomes have not been fully investigated and calls for more research.

The timing of uterotonic is the focus of a Cochrane review by Soltani, Hutchon and Poulose (2010). They found no significant influence on outcomes such as PPH, retained placenta or the length of third stage of labour from administration of oxytocin either before or after expulsion of the placenta. However, the number of available studies was limited and oxytocin was mainly administered through intravenous infusion (IV). The authors concluded that more studies are needed to examine maternal and neonatal outcomes using consistent approaches. Research by Schorn (2012) found no SD between women who received oxytocin and those who did not in relation to the length of third stage, contraction frequency and intensity. This could suggest that delaying oxytocin would allow physiological transition for low risk women before active management is needed but caution needs to be taken as this was a small sample size and did not measure PPH as an outcome.

The World Health Organisation (WHO) recommended that “All women giving birth should be offered uterotonics during the third stage of labour for the prevention of PPH,” (WHO, 2012, p.4). A later publication revised this to immediately after birth of the baby with emphasis that administration of the uterotonic was the most important component of AMTSL (WHO, 2014). The evidence to change from during third stage to immediately after birth appears to stem from a single clinical trial (Gülmezoglu et al., 2012). They found that CCT had little effect on the risk of severe PPH and conclude effective prevention can be achieved with intramuscular injection of oxytocin at ten international units (10IU) after delivery of the baby. To justify this conclusion I believe their evidence is debatable for several reasons. Firstly, all women received the uterotonic at the same time; there are no comparison groups to juxtapose outcomes following administration at different times. Secondly, if an intravenous line was in situ, oxytocin was diluted in saline and administered intravenously; the findings do not clarify these numbers so it is difficult to determine if route of administration affected outcomes measured. Thirdly, individual sites involved in the trial were allowed to implement uterine massage according to their existing routine and again, the findings do not clarify influence on outcomes.

Guidelines from the National Institute for Health and Care Excellence (NICE) were also expanded in 2014 to say “Administer 10IU oxytocin by intramuscular injection with the birth of the anterior shoulder or immediately after the birth of the baby and before the cord is clamped and cut,” (NICE, 2014, p.1.14.13). When reviewing AMTSL the guideline development group (National Collaborating Centre for Women’s and Children’s Health, 2014) agreed that: PPH, postnatal haemoglobin and the need for a blood transfusion were priority outcomes. They reviewed three studies and concluded that the timing of uterotonic in relation to the timing of cord clamping did not make a significant difference (SD) to priority outcomes so recommended that an uterotonic should be administered at the birth to be in line with current practice. I would question to validity of this assumption as in all the studies the uterotonic was administered as soon as the baby delivered with no comparison groups. The timing of uterotonic was not the focus of the reviews and therefore differential outcomes were not reported on addressing optimal timing of uterotonic. It may be practice to administer at birth but I do not feel its evidence based.

The Royal College of Obstetricians and Gynaecologists (RCOG) acknowledge that, while it has been evidenced and accepted as common practice in the UK to DCC (Westhoff, Cotter and Tolosa, 1996; NICE, 2014), the full influences of prophylactic uterotonic drugs is less well known (RCOG, 2015). While they recognise that administration of intramuscular uterotonics before cord clamping is unlikely to affect placental transfusion they do call for further research to confirm whether there could be other clinically relevant effects (RCOG, 2015).

From my background search I have found that New Zealand appears to be the only country that has actively acknowledged an issue with the evidence as they say:

When cord clamping is delayed the optimal timing for administration of the uterotonic is currently unknown and the effect of the administration of an uterotonic on neonatal health when given prior to clamping of the cord is also unknown. Until further evidence to support practice is collated, it is advised that the uterotonic drug of choice be administered after the cord has been clamped and cut. (New Zealand College of Midwives, 2013, p.3).

Literature review

The objective of this literature review is to assess and investigate available literature studying the timing of oxytocin during the third stage of labour. I searched the Maternity and Infant Care, the Cochrane Library and Pubmed databases and details of the search strategies can be found in appendix 1. Articles were limited to those from developed countries, written in English, with publication date from 1990 to the current day. Abstracts were read and eligibility for inclusion assessed. Any articles that did not describe studies which included a comparison group, investigate oxytocin use in third stage and relate to maternal and/or neonatal outcomes were excluded. Studies using a qualitative design were also excluded as they provide little evidence into the effectiveness of oxytocin; qualitative is better suited when you are exploring perceptions, understanding, opinions and motivations behind a research question (Steen and Roberts, 2011). The search identified seven studies of which four were excluded.

For more information about the included and excluded studies see the characteristics of included and excluded studies in appendix 2 and table 1. A Critical Appraisal Skills Programme (CASP UK, 2013) tool was applied to each of the three studies and full details can be found in appendix 3.

The aims of the three studies were similar with all comparing administration of oxytocin at the beginning of the third stage of labour with after placental delivery. AMTSL comparison was more difficult as all three studies used different techniques or failed to fully disclose details, such as timing of cord clamping. The route of administration and dose used varied between the studies. Attrition was also compared with a rate of 0%, 7.2% and 12.5% respectively (calculated by drop out/total participants x 100). A loss greater than 20% could indicate possibility of bias so is an important factor when assessing quality of evidence (Dumville, 2006).

All three studies were double-blinded epidemiological RCT’s which is seen as the gold standard in conducting research as this removes many sources of bias from the process (De Brún, 2013; NICE, 2017; RCOG, 2015). The objective of an RCT is to ensure that the intervention is the only SD between the control/intervention groups and while study one partially achieved this with no SD in baseline characteristics there was extraneous variables which are determinants for PPH such as prior PPH (Briley et al., 2014) and oxytocin augmentation (Belghiti et al., 2011). Study two also produced maternal and fetal characteristics with SD in gestational age and the use of epidural analgesia; also risk factors for PPH (Sheldon et al., 2014). Study three found no SD in some comparison characteristics of the groups, however, it lacks depth and breadth of analysis to support its conclusions. For example, it concludes that IV oxytocin at the start of third stage is safe and effective but this is based on the only SD of lower mean estimated blood loss (EBL) in the intervention group; there were no cases of PPH in the study. The primary function of oxytocin use is to prevent PPH (WHO, 2012), not to achieve the lowest blood loss as some would argue that physiological blood loss up to 500ml is perfectly normal following childbirth (Gyte, 2006).

It could be argued that only study one is likely to be robust as this successfully determined and achieved a sufficient sample size. Small numbers in RCT’s increase the risk of chance from unknown or confounding factors that may interfere with outcomes (Cluett & Bluff, 2006). Study two could be criticized as it was initially aiming for a larger sample size but was terminated early due to SD in PPH and lack of resources to continue. Study three acknowledges the small sample size but does present the results as pilot data. One does have to question the validity of the latter two and the generalisation to a larger population.

The differing heterogeneous data, such as exclusion criteria, between the studies further challenge implications for practice. All studies excluded multiple gestations and caesarean section births. Study one also excluded antepartum haemorrhage (APH) or bleeding disorders while study three also excluded bleeding disorders but not APH. Within the United Kingdom (UK), third stage of labour does not follow the same process as any of the studies (NICE, 2014) and variables included in the studies would be managed differently. For example, the Trust for the research proposal uses an alternative uterotonic for women with known risk factors for PPH (Dunning & Godwin, 2015). Therefore, by having comprehensive exclusion criteria this proposal aims to reduce bias ensuring results can be generalised for low risk women only.

A summary of outcomes measured can be found in Table 2. Study one found no SD between the control and intervention groups using 20IU oxytocin intravenously, however, study two found a significantly higher incidence of PPH in those receiving 10IU intramuscularly at the time of delivery of the anterior shoulder. While the use of 10IU intramuscularly is comparable to UK, the findings need to be considered with caution due to the smaller sample size. The implications for UK practice are also limited as management of the third stage is significantly different, for example delayed cord clamping and no routine uterine massage or cord draining. Study three found that mean EBL was higher in those receiving 10IU intravenously after placental delivery but it is difficult to substantiate the benefit or harm of this as neither equates to a PPH.

To conclude, this background and literature review has revealed that the state of knowledge is currently unsatisfactory concerning whether early use of Oxytocin on low risk women affects maternal and neonatal outcomes that are different from those who have delayed administration. It would be unwise to accept current evidence without acknowledging the inconsistencies. An overarching theme of a one size fits all comes across which devalues individualised needs for safer, personalised maternity care moving forward  (Cumberlege, 2017). I would have liked to have reviewed more research but found a limited number of appropriate studies. From the studies I have included, these have various constraints which make it difficult for them to offer contribution to the current evidence base for practice.  Whilst it is clear prophylactic oxytocin is beneficial in AMTSL, it is difficult to be confident that early routine administration is better then optimal timing based on the clinical picture and individual needs. Women and neonates could potentially be receiving interventions that increase the risk of morbidity and mortality as oxytocin has potential to influence maternal blood loss, retained placenta, haemoglobin and haematocrit for the women and neonate which has implications for immediate and long term health.

The proposal below offers to promote the development of new understanding in an area with inadequate depth of current knowledge. This can be used to inform clinical practice and justify standards of care based on robust evidence using rigorous methodology and evidence on the well-being of the women and neonate to ensure optimal outcomes.