Treatment for primary postpartum haemorrhage
This commentary is now outdated and has been replaced by a RHL summary. It is included in RHL for archival purposes only. It may be cited as: Fawcus S. Treatments for primary postpartum haemorrhage: RHL commentary (last revised: 8 May 2007). The WHO Reproductive Health Library; Geneva: World Health Organization.
RHL Commentary by Fawcus S
1. EVIDENCE SUMMARY
The objective of the review was to determine the effectiveness and safety of pharmacological and surgical interventions used to treat postpartum haemorrhage (PPH). The interventions focused on were those used to treat uterine atony—the most common cause of PPH. The authors planned to compare seven interventions (uterotonics, haemostatics, uterine packing or tamponade, vessel ligation, hysterectomy, uterine compression sutures and radiological embolization); each with other treatments, controls or placebo. Main outcomes defined were maternal mortality, severe maternal morbidity and hysterectomy.
There were only three studies included in the review. The first was conducted in South Africa in 2001 (SA 2001 study) (1). It was a randomized single blind trial that compared rectal misoprostol 800 µg with a “standard regime” for the management of PPH, which consisted of intramuscular syntometrine plus oxytocin infusion (defined as > 500 ml of blood loss). The study was stopped prematurely since, compared with the standard regime, there was a significant benefit in the misoprostol arm of the trial with respect to the cessation of bleeding within 20 minutes (6% vs 34%, RR 0.18, 95% CI 0.04- 0.67). Also, in the misoprostol group, there was a significantly less need for additional medical intervention. No difference was demonstrated on the rates of hysterectomy, vessel ligation, maternal mortality and morbidity. This was largely due to the study being too small. The reviewers indicate that this study was flawed in that it was not blinded for the practitioners who may have been biased in their interpretation of the response to treatment. Also the method of assessment of blood loss was not described and appeared subjective.
The other two randomized controlled trials were conducted in South Africa(2004) and the Gambia(20040.(2, 3). These are double-blinded trials comparing misoprostol regimens with placebo for the second line treatment of PPH due to uterine atony. In both studies, management of the third stage of labour included prophylaxis with oxytocin or syntometrine. In women with PPH (defined as > 500 ml of blood loss) in the Gambian study, and “more than usual blood loss” in the South African study, a standard treatment regime of an oxytocin injection with or without ergometrine was initially given. After this women were randomized to the misoprostol or the placebo group, and continuing blood loss measured. Another feature of these studies was that different dosages and routes of misoprostol administration were used: the Gambian study used 200 µg orally plus 400 µg sublingually and the South African study used 200 µg orally, 400 µg sublingually and 400 µg rectally. It was argued that these combinations of different routes administration facilitated rapid and sustained uterotonic action. Both studies had an objective method for estimating blood loss. Results showed a trend towards lower blood loss with misoprostol compared with placebo. However, the results were not statistically significant, possibly due to the standard treatment resulting in less blood loss in the placebo group than expected. It would be useful to document how often ergometrine is included in the standard and rescue treatments, since this agent has the potential for severe side-effects. These studies also documented side-effects in both study arms, demonstrating a significant increase in shivering in the misoprostol arm. The SA study had 3 maternal deaths in the misoprostol arm, but in two, the cause was undiagnosed trauma (cervical and possible scar dehiscence) and not due to intractable uterine atony. Both studies conclude with a recommendation for sufficiently powered multicentred randomized trials to conclusively evaluate the efficacy and safety of misoprostol in the treatment of PPH, after first line oxytocics have failed.
The review was not able to identify any trials on other modalities of treatment for primary PPH.
2. RELEVANCE TO UNDER-RESOURCED SETTINGS
2.1. Magnitude of the problem
PPH is a major cause of maternal mortality in both developed and developing countries. Globally it is estimated that severe PPH occurs in about 11% of women who give a live birth (4). The incidence is thought to be much higher in developing countries where many women do not have access to a skilled attendant at delivery and where active management of the third stage of labour may not be routine. It is estimated that about 14 million women suffer severe blood loss postpartum, and that 1% of these die as a result. A further 12% survive with severe anaemia (4). Adverse sequelae of PPH are often aggravated in under-resourced settings by pre-existing anaemia and HIV/AIDS, meaning that even lesser amounts of blood loss can be fatal. The United Kingdom Confidential Enquiries into Maternal Deaths had a maternal mortality rate due to PPH of 1.0 per 100 000 maternities. During the three years from 2000 to 2002, there were 10 maternal deaths from PPH out of a total of 261 maternal deaths (5). The South African Confidential Enquiries into Maternal Deaths indicated that PPH was the third most common cause of death, accounting for 313 deaths in a total of 3406 in the triennium 2002 to 2004 (6). The South African report, which describes deaths in health facilities, shows that the majority of deaths from PPH (40%) occur at level 1 institutions, such as clinics or district hospitals. In the report, patient/community-related factors, lack of emergency transport, lack of blood for transfusion, lack of appropriately skilled personnel, and substandard treatment were all included among avoidable factors. A community-based study of maternal mortality in Zimbabwe showed that obstetric haemorrhage was a leading cause of maternal mortality in a rural area as compared with an urban area (7). In the same study, it was found that 50% of women dying from PPH did not have access to any treatment before their death—i.e. they died at home or in transit to, or between, institutions (8).
2.2. Applicability of the results
The review refers to only three studies for the treatment of PPH. These studies were all done in developing countries, and there is no reason to suggest that their findings would not be generally applicable to all women with PPH from uterine atony. Misoprostol has a long shelf life and does not require refrigeration and is therefore suitable for use in a resource-poor setting. Another advantage it has over ergot drugs (e.g. ergometrine, syntometrine) is that it does not cause hypertension and can be safely used in women with pre-eclampsia and hypertensive disorders. The 2001 SA study evaluated misoprostol as a first line oxytocic agent for treatment of uterine atony, but the study was not double blinded so, it cannot, at this stage, be concluded that it should become first-line choice for the management of all cases of uterine atony. As a second-line uterotonic agent, misoprostol may have advantages over ergometrine and prostaglandin F2 alpha owing to its fewer major side-effects, but this needs further evaluation. The 2004 studies evaluated this question and suggested that misoprostol reduced blood loss, but were underpowered to be conclusive about efficacy and safety. If misoprostol is conclusively proven to be effective, the question will need to be answered as to whether it should be the first-line or second-line treatment (i.e. following unsuccessful oxytocin infusion) and whether it can be administered safely by midwives at primary-level facilities.
2.3. Implementation of the intervention
If misoprostol is confirmed to be an effective medical treatment for PPH, the implementation of this policy would require rewriting protocols for the management of PPH at each level of care, and training of midwives and doctors in its use. The protocol would need to specify whether misoprostol is the first-line or second-line treatment and staff would need to be educated about its side-effects. Implementation would be feasible in units currently using an intervention such as syntometrine plus oxytocin infusion for treatment of PPH. It would require replacing this standard regime with rectal misoprostol. Rectal administration is a procedure within the scope of practice of all doctors and midwives.
Storage of misoprostol does not require refrigeration, but requires a good control of stock levels. Its unregulated use in the community for termination of pregnancy or induction of labour can lead to dangerous consequences
The implementation of a new regime for treatment of PPH will only be possible and effective if other aspects of the health systems are functional. In many developing countries, maternity services are arranged in tiered systems with midwifery units providing care for low-risk women and the district or regional hospital being staffed by doctors and providing care for women with obstetric complications (9). The rapidity of deterioration and severity of PPH means that women with severe PPH cannot survive long distance referral. Therefore midwives at the primary-level units must have the capacity to provide immediate resuscitation and treatment. Hence, guidelines for treatment of PPH in under-resourced settings must give clear indications as to which treatments can be provided by what cadre of staff at each level of care. Such guidelines need to be in the form of algorithms which include the steps in diagnosing the cause of PPH, sequential modalities of treatment, resuscitative measures and how/when to refer. Some deaths from PPH occur due to health workers persisting with treatment for uterine atony when another cause is responsible for the PPH, such as retained placenta or genital tract lacerations.
Misoprostol is potentially a very useful intervention for treatment of PPH. More research is needed to:
- confirm its efficacy with a double-blind randomized control trial, comparing it with standard treatment (oxytocin + syntometrine). Larger trials are needed to assess the effect of misoprostol on hysterectomy rates and maternal mortality rates, as well as its side-effect profile;
- establish whether it should be used as a first-line or second-line treatment option:
- evaluate the optimal route of administration; and
- evaluate whether it can be used effectively by midwives at primary-level maternity facilities.
The reviewers also conclude that more randomized controlled trials are needed on specific treatments: radiological embolization, haemostatic drugs, tamponade procedures, uterine haemostatic sutures for intractable haemorrhage. Radiological embolization requires tertiary-level facilities and so is not readily implementable in under-resourced settings.
There is a need for an effective temporizing intervention for use at the primary level when first-line measures have been unsuccessful and there is a need to transfer the woman to a higher level facility. Packing and tamponade procedures could be usefully evaluated in this regard. The recent pilot studies on a non-pneumatic Anti-Shock garment for stabilization during transit, requires further in service evaluation(10).More research is needed on appropriate surgical interventions for intractable haemorrhage in district and regional hospitals. It would be useful to compare the B Lynch suture with stepwise devascularization of the uterus as a conservative measure for intractable haemorrhage.
There is also a need for clear ethical guidelines for evaluating treatment for intractable haemorrhage based on evidence from randomized control trials. Placebo-controlled trials are inappropriate. How, and from whom, consent is obtained before randomization needs further examination. If the aim of research is to reduce deaths from PPH, then health systems research will also by required. Such research can investigate the functioning of a health system and identify constraints to effective care. As indicated in the commentary, deaths from PPH are due to problems of access, availability of blood, transport delays and lack of skilled personnel. Interventions should address all of the above. From the perspective of an under-resourced setting, it would be useful to evaluate specific “intervention packages”. A community-based situational assessment of mortality from PPH at home, primary, secondary, and tertiary levels could be performed. This could be followed by the development of a management algorithm which includes the diagnosis of the cause of the PPH, resuscitation, stepwise treatment and referral. After training and implementation of the guideline, a post intervention evaluation could be performed.
- Lokugamage AV, Sullivan KR, Niculescu L, Tigere P, et al. A randomised study comparing rectally administered misoprostol versus syntometrine combined with an oxytocin infusion for the cessation of primary postpartum haemorrhage. Acta obstetrica gynecologica scandinavica 2001;80 (9):(9).835-839.
- Walraven G, Dampha Y, Bittaye B, Sowe M, Hofmeyr J. Misoprostol in the treatment of postpartum haemorrhage in addition to routine management: a placebo randomized controlled trial. British journal of obstetrics and gynaecology 2004;111: 1014–1017.
- Hofmeyr J, Ferreira S, Nikodem C, Mangesi L, et al . Misoprostol for treating postpartum haemorrhage: a randomized controlled trial. BioMed Central Pregnancy and childbirth 2004;4:16.
- Abou-Zahr C . The global burden of maternal death and disability. British medical bulletin 2003;67:1-11.
- Why mothers die . 2000-2002. The report on confidential enquiries into maternal deaths in the United Kingdom. CEMACH, London, RCOG Press 2004;86-93.
- Saving mothers . 2002-2004. The report on confidential enquiries into maternal deaths in South Africa. Third report on confidential enquiries into maternal death in South Africa. Department of Health, South Africa 2006:68-95.
- Mbizvo M, Fawcus S, Lindmark G, Nystrom L and the Maternal Mortality study group . Operational factors of maternal mortality in Zimbabwe. Health policy and planning 1993;8:(4).369–378.
- Fawcus S, Mbizvo M, Lindmark G, Nystrom L and the Maternal Mortality Study Group . A community-based investigation of maternal mortality from obstetric haemorrhage in rural Zimbabwe. Tropical doctor 1997;27:159–163.
- Koblinsky M, Campbell O, Heichelheim J . Organising delivery care. What works for safe motherhood. Bulletin of the World Health Organization 1999;77:(5).399–406.
- Miller S, Hamza S, Bray EH et al. First Aid for obstetric haemorrhage: the pilot study of the non-pneumatic anti-shock garment in Egypt. BJOG 2006;113:424-429.
This document should be cited as: Fawcus S. Treatments for primary postpartum haemorrhage: RHL commentary (last revised: 8 May 2007). The WHO Reproductive Health Library; Geneva: World Health Organization.