Intended for healthcare professionals

Practice Clinical updates

Diagnosis and management of postpartum haemorrhage

BMJ 2017; 358 doi: (Published 27 September 2017) Cite this as: BMJ 2017;358:j3875
cropped thumbnail of infographic

Infographic available

A visual summary of suggested approaches for the management of primary and secondary postpartum haemorrahage

  1. Edwin Chandraharan, consultant obstetrician and gynaecologist, lead clinician of labour ward, and honorary senior lecturer1 2,
  2. Archana Krishna, specialist registrar in obstetrics and gynaecology1
  1. 1St George’s University Hospitals NHS Foundation Trust, London SW17 0QT, UK
  2. 2St George’s University of London, London, UK
  1. Correspondence to: E Chandraharan edchandi{at}

What you need to know

  • Postpartum haemorrhage remains the second leading direct cause of maternal deaths in the UK and the leading cause of maternal mortality in the world

  • Poor uterine tone accounts for about 80% of all cases of primary postpartum haemorrhage, whereas endometritis is the commonest cause of secondary postpartum haemorrhage presenting up to 12 weeks after delivery

  • Tranexamic acid is recommended for all women with atonic and traumatic postpartum haemorrhage as well as for ongoing haemorrhage during a caesarean section

  • Refer women with secondary postpartum haemorrhage after birth for ultrasonography to exclude retained products of conception or endometritis

  • Start broad spectrum antibiotics in women with secondary postpartum haemorrhage due to endometritis

Postpartum haemorrhage is a major cause of death during pregnancy and early motherhood, accounting for 25% of maternal deaths worldwide,1 and is the second leading direct cause of maternal deaths in the UK.2 It is defined as blood loss of more than 500 mL from the female genital tract after delivery of the fetus (or >1000 mL after a caesarean section). Primary postpartum haemorrhage occurs within the first 24 hours of delivery, whereas secondary postpartum haemorrhage occurs between 24 hours and 12 weeks after delivery and is less common.1 For every maternal death due to postpartum haemorrhage, there are at least 10 “near-misses.” Serious maternal morbidities include multiorgan failure, multiple blood transfusion, and peripartum hysterectomy.1 There have been recent advances in the management of postpartum haemorrhage secondary to coagulopathy and abnormal invasion of the placenta.3

This review highlights the causes, diagnosis, and management of postpartum haemorrhage and is aimed at those involved in obstetric and postnatal care.

Sources and selection criteria

We searched PubMed using the key words “postpartum haemorrhage,” “balloon tamponade,” “uterine compression suture,” “embolization,” and “obstetric haemorrhage” from 1981 to 2016. We also considered review articles from 2001 to 2016 including systematic and narrative reviews as well as the most recent Cochrane systematic review on treatment for primary postpartum haemorrhage.21 Priority was given to randomised controlled trials and four large prospective series.

What are the mechanisms of and risk factors for postpartum haemorrhage?

Atonic postpartum haemorrhage secondary to a poor tone of the uterine muscle accounts for approximately 80% of all women with excessive bleeding from the genital tract within 24 hours of delivery.4 Women who have had prolonged labour, multiple pregnancy, polyhydramnios, a large fetus, obesity, or pyrexia during labour are all at increased risk.5 Rare causes of primary postpartum haemorrhage include uterine inversion, placenta percreta (fig 1), as well as extra-genital bleeding. The commonest cause of secondary postpartum haemorrhage is endometritis.


Fig 1 Placenta percreta invading the uterine myometrium and the serosa

It is estimated that over 85% of women who have a vaginal birth will sustain some degree of perineal trauma, and, of these, 60-70% will need to be sutured.6 7 An episiotomy itself can increase the risk of postpartum haemorrhage by up to fivefold.8 Emergency caesarean sections were associated with an approximately threefold increase in postpartum haemorrhage compared with elective caesarean sections or spontaneous vaginal births.9

A population based, cohort nested, case-control study has concluded that, after adjustment for all potential confounders, intrapartum use of oxytocin was associated with a significantly higher risk of severe postpartum haemorrhage (adjusted odds ratio 1.8 (95% confidence interval 1.3 to 2.6)) in women who did not receive prophylactic oxytocin after delivery; the odds ratio for haemorrhage increased from 1 to 5 according to the level of oxytocin exposure.10 It is postulated that this is due to excessive uterine contractions and resultant lactic acidosis in the uterine muscle as well as prolonged labour, when oxytocin is used to augment labour.

How is primary postpartum haemorrhage assessed and diagnosed?

Examination and resuscitation

Attempt to identify the cause in women who are bleeding immediately after birth in parallel with resuscitation. The infographic suggests an approach to management adapted from the Royal College of Obstetricians and Gynaecologists Green-top Guideline on postpartum haemorrhage.5

Further assessment

It is important to perform a systematic examination immediately after birth11 so that specific treatments can be instituted (see infographic).

Any delay in achieving haemostasis after birth can result in major loss of maternal blood volume, leading to hypotension, hypoxia, and acidosis. The blood flow to the uterus at term (that is, >37 weeks of gestation) is approximately 1000 mL of blood every minute, and a fetus at term receives about 200 mL/kg/minute from the placenta.12 Estimating the amount of blood that has been actually lost in postpartum haemorrhage visually is prone to error.13

An obstetric shock index (that is, pulse rate divided by systolic blood pressure) of >1 has been shown to be associated with substantial postpartum haemorrhage and the need for intensive resuscitation and blood transfusion.14 15 In our clinical opinion, an obstetric shock index of >1 would indicate the need for immediate action to ensure haemodynamic stability.

A management algorithm called HAEMOSTASIS has been proposed to aid a systematic and stepwise management of postpartum haemorrhage (box 1),11 and a recent retrospective observational study has suggested that the use of this algorithm has helped improve outcomes and reduced the likelihood of peripartum hysterectomy.16 “HAEMO” refers to the immediate measures to be taken to arrest haemorrhage, while “STASIS” represents the more advanced measures (box 1).

Box 1: HAEMOSTASIS algorithm for management of postpartum haemorrhage16

  • H—Ask for help and hands on uterus (uterine massage)

  • AAssess (that is, ABC) and resuscitate (that is, intravenous fluids)

  • EEstablish aetiology, ensure availability of blood, and ecbolics (drugs that induce contractions of the uterus, oxytocin or ergometrine)

  • MMassage the uterus

  • OOxytocin infusion (10 U/hour) or intramuscular prostaglandins (250 μg)

  • SShift to theatre, with aortic compression, bimanual compression, or anti-shock garment (for low resource settings before transfer to a tertiary centre) as appropriate

  • TTamponade by balloon or uterine packing after exclusion of retained tissue and trauma. Administer intravenous tranexamic acid (1 g)

  • AApply compression sutures on the uterus (B-Lynch or modified technique)

  • SSystematic pelvic devascularisation (uterine, ovarian, quadruple. or internal iliac)

  • IInterventional radiology and, if appropriate, uterine artery embolisation

  • SSubtotal or total abdominal hysterectomy

Intra-abdominal bleeding secondary to an extragenital cause such as the rupture of the liver or spleen is uncommon, but may be more likely in women with severe pre-eclampsia due to rupture of the hepatic capsule. If a woman remains unresponsive to treatment or if the amount of visible blood loss is less than the observed haemodynamic instability, we recommend an ultrasound scan to investigate a possible intra-abdominal cause.

How is postpartum haemorrhage managed?

It is important to identify both antepartum and postpartum risk factors that might predispose women to postpartum bleeding—such as placenta praevia, uterine fibroids, and retained placenta—as outlined in the infographic. Planning ahead involves input from a multidisciplinary team to ensure that experienced clinicians are present at the birth and there are suitable blood products available.

Active management of the third stage of labour is the process by which expulsion of the placenta and membranes is achieved after delivery by uterine massage, controlled cord traction, and the use of oxytocin as well as other drugs.

A randomised controlled trial showed that active management of the third stage alone reduces the incidence of primary postpartum haemorrhage by 70% compared with physiological management alone.17 Clinical guidelines on postpartum haemorrhage continue to recommend mechanical methods such as bimanual compression or emptying the urinary bladder based on consensus of professional opinion.5

What drugs are used in the medical management of primary postpartum haemorrhage?

Oxytocin is the most commonly used drug in the medical management of postpartum haemorrhage. Other drugs are listed in box 2.

Box 2: Drugs used in treatment of postpartum haemorrhage (source: BNF Online June 2017)

First line drugs
  • Oxytocin (octapeptide which is secreted by the supraoptic and paraventricular nuclei of the hypothalamus and is stored in the posterior pituitary gland)

    • Mode of action—Myometrial contraction and retraction; increases basal uterine tone

    • Side effects—Nausea, vomiting, headache

  • Ergometrine (ergot alkaloid)

    • First line drug in developing countries

    • Mode of action—Arterial vasoconstriction and myometrial contraction

    • Side effects—Vomiting, headache, hypertension, chest pain, palpitations, bradycardia, Raynaud’s syndrome, pulmonary oedema18

Second line drugs
  • Tranexamic acid

    • Mode of action—Antifibrinolytic which prevents the breakdown of preformed blood clot and therefore stabilises the clot

    • Side effects—Hypotension, diarrhoea, thromboembolic events

    • Recent Cochrane review of 10 randomised controlled trials (RCTs) reported that blood losses >400 mL or >500 mL and >1000 mL were less common in women who received tranexamic acid compared with placebo or no intervention (risk ratios 0.52 (95% confidence interval 0.42 to 0.63) and 0.40 (0.23 to 0.71), respectively)19

  • Misoprostol (prostaglandin analogue)20

    • Mode of action—Myometrial contraction

    • Side effects—Diarrhoea, rash, dizziness, vomiting21

    • Not found to be effective after administration of oxytocin21 and may increase adverse effects22

  • Prostaglandins F

    • Mode of action—Myometrial contraction

    • Side effects—Bronchospasm, cardiovascular system collapse, dyspnoea, hypertension, vomiting, pulmonary oedema

    • No robust evidence of effectiveness

  • Carbetocin (synthetic oxytocin analogue)

    • Mode of action—Myometrial contraction

    • Side effects—Diarrhoea, hypotension

    • Cochrane review of 11 RCTs concluded that use of carbetocin statistically significantly reduced the need for therapeutic uterotonics (risk ratio 0.62 (0.44 to 0.88) compared with oxytocin for women who underwent caesarean section but not for vaginal delivery.23 There was no robust evidence to suggest that carbetocin was better than oxytocin in reducing postpartum haemorrhage, and its cost effectiveness remains unclear

  • Syntometrine (combination of 5 units of oxytocin and 0.5 mg of ergometrine)

    • Mode of action—Myometrial contraction

    • Side effects—Nausea, vomiting, diarrhoea

    • Cochrane review of 4 RCTs that compared carbetocin and syntometrine showed a lower mean blood loss in women who received carbetocin (mean difference −48.84 mL (95% CI −94.82 to −2.85 mL)23

The incidence of postpartum hypertension has been found to be significantly lower in women who received carbetocin compared with those who received syntometrine.23 Therefore, ergometrine and syntometrine should be avoided in women with hypertension and pre-eclampsia to avoid the risk of stroke.

Although, injectable prostaglandins (prostaglandin F and its synthetic analogue carboprost tromethamine) have been used as an adjunct to oxytocin in the management of atonic postpartum haemorrhage, they have not been subjected to any randomised controlled trials. Despite the lack of robust scientific evidence, most clinical guidelines recommend the use of injectable prostaglandins in the management algorithm, up to eight doses 15 minutes apart of 250 μg given intramuscularly.5 Use prostaglandins with caution in patients with bronchial asthma as it is not recommended as an intra-myometrial injection.5

A recent Cochrane systematic review of 10 randomised controlled trials reported that, compared with placebo, treatment with tranexamic acid (1 g intravenous) reduced blood loss in women with atonic postpartum haemorrhage,21 and therefore tranexamic acid is recommended for atonic and traumatic postpartum haemorrhage as well as ongoing haemorrhage during a caesarean section.5

What is the role of a uterine tamponade balloon?

A recent systematic review, which included 241 women, has concluded that the insertion of a uterine tamponade balloon to control bleeding is effective in 97% of cases of postpartum haemorrhage.24 The authors suggested that it would be logical to use this least invasive, easiest, and most rapid approach as the first step in the management of intractable postpartum haemorrhage after the failure of pharmacological treatment.

However, this systematic review did not include any randomised controlled trials and therefore the conclusion should be interpreted with caution.

What are the surgical treatment options for managing primary postpartum haemorrhage?

Some women may need examination under anaesthesia to repair tears or trauma to the genital tract and the uterus, evacuation of retained products, insertion of balloon for tamponade, or drainage of a haematoma.25 Surgery may also be needed in case of bleeding that is not responsive to volume resuscitation and pharmacological treatment.

Some women may need an exploratory laparotomy in order to insert compression sutures and to manage uncommon extra-genital sources of bleeding such as rupture of splenic artery aneurysm.26 27

Women who remain unresponsive to resuscitation despite measures aimed at controlling the ongoing haemorrhage may need a hysterectomy as a lifesaving measure.5 In centres where blood and blood products are not readily available, hysterectomy may be considered earlier to avoid dilutional coagulopathy secondary to the excessive transfusion of intravenous fluids as well as the loss of platelets, fibrinogen, and clotting factors (that is, the “washout phenomenon”).

How is secondary postpartum haemorrhage managed?

As secondary postpartum haemorrhage occurs anytime between 24 hours and 12 weeks after delivery (most commonly between day 7 and day 14), it is important that clinicians working in community health settings are able to diagnose and manage it (see box 3).

Box 3: Tips for non-specialists

  • Secondary postpartum haemorrhage occurs after the first 24 hours following childbirth and is most commonly due to endometritis or retained products of conception

  • Consider secondary postpartum haemorrhage in women who have excessive bleeding (that is, passage of clots or continuous bleeding) which is more than the normal lochia after childbirth

  • Check the woman’s temperature and exclude uterine tenderness, offensive vaginal discharge, or failure of uterine involution

  • Refer to an obstetrician for an early clinical assessment and ultrasound scanning to exclude retained products of conception

History and assessment

Consider the possibility of retained products of conception or endometritis in women with a history of manual removal of placenta or prolonged rupture of membranes, prolonged labour, or pyrexia during labour. Refer these women for an ultrasound scan to exclude retained products of conception. Start broad spectrum antibiotics in women with secondary postpartum haemorrhage due to endometritis. Retained products of conception may require surgical evacuation.5

What’s new in the management of postpartum haemorrhage?

Tranexamic acid

The international, randomised, double blind, placebo controlled WOMAN trial reported that tranexamic acid reduces death due to bleeding in women with postpartum haemorrhage with no adverse effects.27 This is consistent with the findings in surgery and trauma (CRASH-2 trial).28 Therefore, based on the recent WOMAN trial, tranexamic acid is recommended in the routine management of postpartum haemorrhage unless there are specific contra-indications.

Blood and blood products

A recent randomised controlled trial (PROPPR trial, 2015) reported that initial infusion with plasma, platelets, and red blood cells in a 1:1:1 ratio did not improve overall survival compared with a 1:1:2 ratio in patients who had had or were at risk of massive blood loss.29 However, in additional analyses, more patients in the 1:1:1 group were reported to achieve “anatomic” haemostasis (objective assessment by the surgeon indicating that bleeding within the surgical field was controlled and no further haemostatic interventions were anticipated), and fewer patients may have died due to exsanguination by 24 hours. A 1:1 transfusion ratio of plasma to red blood cells is recommended, especially if bleeding is not under control.5 In acute emergencies where group-specific cross-matched blood is not readily available, consider O− “un-cross-matched” blood.

Abnormal invasion of the placenta

Abnormal and deep invasion of the placenta into the uterine myometrium (fig 1) is associated with serious maternal morbidity and mortality secondary to massive obstetric haemorrhage.30 It has traditionally been managed with peripartum hysterectomy or intentional retention of placenta. However, a new conservative surgical technique called the triple P procedure has been described31 with good outcomes.32

Education into practice

  • If a woman presents with vaginal bleeding up to 12 weeks after delivery in the community, do you palpate her abdomen for uterine size, tone, and tenderness? The uterus should not be palpable per abdomen by day 14; a palpable uterus at this stage should make you suspect endometritis or retained products of conception.

  • How can you ensure that women who have a primary postpartum haemorrhage on their discharge summary receive specific follow-up to assess their bleeding, their haemoglobin levels, and monitor oral iron supplementation? Do you have a local pathway in place to support this, and if not, can you create one?

Educational resources

How patients were involved in the creation of this article

No patients were involved in the creation of this article.


  • Contributors: EC and AK designed the manuscript and AK performed the literature search. EC and AK co-wrote the manuscript, and EC is the guarantor and takes overall responsibility for the manuscript.

  • Competing interests: We have read and understood BMJ policy on declaration of interests and have no relevant interests to declare.

  • Provenance and peer review: Commissioned; externally peer reviewed.


View Abstract

Log in

Log in through your institution


* For online subscription