Included studies

Two studies were included for this review, 1 randomised controlled trial (RCT) (Biradar 2021) and 1 Cochrane systematic review (Oladapo 2020), which included 7 RCTs (Adnan 2018, Charles 2019, Dagdeviren 2016, Durocher 2019, Neri-Mejia 2016, Oguz 2014, Sangkhomkhamhang 2015).

Both studies compared intravenous (IV) administration of 10 units oxytocin to intramuscular (IM) injection of 10 units oxytocin. The route of administration of IV oxytocin varied: 5 included RCTs in the Cochrane systematic review administered an IV oxytocin bolus injection (Adnan 2018, Charles 2019, Neri-Mejia 2016, Oguz 2014, Sangkhomkhamhang 2015) and 3 included RCTs in the Cochrane systematic review administered an IV oxytocin slow infusion (Charles 2019, Dagdeviren 2016, Durocher 2019). One RCT administered a combined IV oxytocin bolus injection and IV oxytocin infusion (Biradar 2021).

The included studies are summarised in Table 2.

See the literature search strategy in appendix B and study selection flow chart in appendix C.

Excluded studies

Studies not included in this review are listed, and reasons for their exclusion are provided in appendix J.

Included studies

A systematic review of the economic literature was conducted but no economic studies were identified which were applicable to this review question.

The outcomes that matter most

The aim of this review was to determine the effectiveness of intravenous administration of oxytocin compared with intramuscular administration in the third stage of labour to prevent excessive bleeding after the birth of the baby. The committee therefore agreed that maternal admission to an ITU or high-dependency area, primary PPH and severe PPH were critical outcomes which would directly capture the effectiveness and safety of the intervention.

The committee agreed that need for manual removal of the placenta was an important outcome which would determine the effectiveness of the mechanism of action of the intervention as oxytocin helps to prevent blood loss after birth of the baby by stimulating the uterus to contract, closing the blood vessels to the placenta and helping the placenta to separate. The committee agreed that the need for additional uterotonics during the third stage or within the first 48 hours should be included as an important outcome as they wanted to know whether IV oxytocin would reduce the need for additional interventions in the third stage compared to IM oxytocin.

The committee included side effects during the third stage or within the first 24 hours as an important outcome as oxytocin administered intravenously as a bolus injection has been associated with hypotension and tachycardia. The committee agreed that women’s experience of labour and birth should be included as an important outcome to capture the acceptability of the intervention and any wider impacts on the woman associated with the differential route of administration of oxytocin.

The quality of the evidence

The quality of the evidence ranged from high to low, with most of the evidence being of low quality. Most of the included RCTs were of high quality, but some outcomes were downgraded for risk of bias where evidence came from studies at risk of selective reporting bias or where outcome assessors were not blinded. Evidence was not downgraded in cases where participants or study personnel were not blinded as the outcomes were measured objectively and the intervention was administered once, so the risk of bias due to deviations from the intervention was low. Outcomes were downgraded for imprecision in some cases where 95% confidence intervals (CIs) crossed the minimally important differences, indicating the effect size was small.

Benefits and harms

The committee discussed the evidence and used this alongside their expert opinion and clinical knowledge to make recommendations. Based on overall effect estimates, the committee agreed there was consistent evidence of benefits of IV oxytocin compared to IM oxytocin in the third stage of labour in terms of maternal admission to ITU, primary PPH, severe PPH, need for manual removal of placenta and need for additional uterotonics. The committee considered the outcomes stratified by type of administration and noted that IV bolus injection had important benefits in terms of primary PPH, severe PPH and need for manual removal of placenta compared to IM injection, whereas there was no difference or no evidence of important difference for IV infusion. There was no important difference in terms of side effects, with the data contributing to this outcome only from studies administering IV injection compared to IM injection.

The committee discussed the evidence from two large RCTs comparing IV oxytocin injection to IM oxytocin stratified by women who have had and have not had oxytocin in the first stage of labour. The committee acknowledged that the stratified analysis did not substantially change the effect estimate in terms of primary PPH, however, in terms of severe PPH, the effect estimate was larger in favour of IV injection for women who had had oxytocin compared to the non-stratified effect estimate and conversely there was no important difference found for women who had not had oxytocin in the first stage of labour. Based on this evidence and insufficient data to stratify other critical and important outcomes by women who had and had not had oxytocin in the first stage of labour, the committee agreed that the evidence supported a recommendation for oxytocin to be administered by IV injection in the third stage for women who had had oxytocin in the first stage of labour, but not for women who had not previously received oxytocin.

The committee also discussed the feasibility and acceptability of administering oxytocin by IV injection, as IM injections are quick to administer and do not require an IV access to be sited using a cannula. The committee also discussed that to reduce the possible haemodynamic effects of IV oxytocin it should be administered slowly (over a few minutes), and a slow IV injection would therefore be more time-consuming for the midwife to administer compared to an IM injection. The committee raised concerns that IV administration in midwifery-led settings may be less acceptable as inserting cannulas during third stage of labour may be less commonplace than in obstetric units and it increases the medicalisation of birth which women in these settings may wish to avoid. The committee were also concerned by the need for 2 midwives to be present during the third stage in order to check and administer the slow IV injection while also ensuring that the baby was safe and well, as in their experience this may not be feasible in all settings due to resource constraints. These considerations confirmed the committee’s evidence-based recommendation to offer oxytocin by slow IV injection only to women who had had oxytocin during labour and therefore would have existing IV access. The committee noted that, although 2 midwives may still be needed during the third stage to check and administer the IV oxytocin, it would remove the need to insert and check a cannula during the third stage. The committee noted the lack of evidence in terms of women’s experience of labour and birth, but agreed that, in their experience, women would prefer to avoid having an intramuscular injection which can be painful, if IV access was already established and another option was as safe and effective.

Cost effectiveness and resource use

The committee noted that there was no difference in the acquisition costs to the NHS of oxytocin whether given by IM or IV injection. The committee recognised that putting in IV access solely for the purposes of administering IV oxytocin would be resource intensive, requiring 2 midwives to be present to insert the access, check it and administer a flush and then the oxytocin. The committee also discussed the related practical difficulties of recommending IV oxytocin for all women with current levels of staffing. Therefore, they confirmed that the recommendations should restrict the use of IV oxytocin to women who had already received oxytocin in labour and who therefore have IV access is situ.