Use of oral misoprostol in the prevention of postpartum haemorrhage


Correspondence: Dr H. El-Refaey, Department of Obstetrics and Gynaecology, University College Hospital, London WC1 6AU, UK.


Objective To investigate the use of the oral prostaglandin E1 analogue, misoprostol in the prevention of postpartum haemorrhage.

Design A prospective observational study.

Setting A university teaching hospital.

Participants Two hundred and thirty-seven consecutive women undergoing vaginal delivery.

Methods All the women were given 600 μg oral misoprostol just after delivery.

Main outcome measures Rates of postpartum haemorrhage; need for therapeutic oxytocic drugs; retained placenta and length of the third stage of labour.

Results Postpartum haemorrhage occurred in 6% of the women; the need for therapeutic oxytocics in 5%, retained placenta in 2% and the median length of the third stage was 5 min. Vomiting and diarrhoea in the first hour after delivery occurred in 8% and 3% respectively and shivering in 60%.

Conclusions Misoprostol may be effective in the prevention of postpartum haemorrhage, and has few side effects. A double blind randomised trial is required.


The third stage is perhaps the most dangerous part of labour for the mother, the main risk being postpartum haemorrhage. There is now good evidence that prophylaxis with oxytocic drugs in the third stage of labour is effective in preventing postpartum haemorrhage1. The intramuscular administration of Syntometrine® (a combination of 5 IU oxytocin and 0.5 mg ergometrine; Sandoz Pharmaceuticals, Camberley, Surrey, UK) is now routine in the developed world.

However, the use of Syntometrine is associated with several problems. It is contraindicated in women with hypertension in pregnancy which may affect about one in seven women2, frequently causes nausea and vomiting1, and has to be given by intramuscular injection requiring a sterile needle and syringe—an important consideration in developing countries. Finally, because oxytocic agents are not stable at high ambient temperatures, they require special storage conditions. When Syntometrine is stored for prolonged periods the temperature must be maintained between 2°C and 8°C, and it must be protected from light3,4. Studies designed to simulate the storage conditions commonly found in tropical countries found that a variety of brands of ergometrine lost 21% to 27% of their potency after one month, and over 90% after one year of storage exposed to light and at 21°C to 25°C5. These storage requirements are an important hurdle to the widespread use of oxytocics in the developing world. The WHO has been trying to overcome this barrier for some time by examining the possibility of synthesising new drugs which are able to withstand high temperatures, but its efforts so far have met with little success4.

Prostaglandins are known to be useful in the treatment of severe postpartum haemorrhage6, and in this situation may be superior to oxytocin and ergometrine4. Prostaglandins do not cause hypertension. They may therefore be useful in the prevention of postpartum haemorrhage.

Misoprostol is a prostaglandin E1 analogue marketed for oral use. It does not require special storage conditions and has a shelf life of several years7,8. Its safety has been established in studies over the past 10 years for the prevention and management of peptic ulcer9. More recently, it has been shown to be a potent uterotonic agent and has been investigated in the induction of abortion, cervical priming and induction of labour, used either alone or combined with mifepristone10–14. Absorption of misoprostol is very rapid, being detected in the circulation within 2 min of its oral ingestion15. Its effect on the early pregnant uterus has also been shown to be very rapid16, and it does not cause hypertension17,18. The aim of this study was to investigate whether misoprostol could prevent postpartum haemorrhage.


This was a prospective observational study, the main objective of which was to ascertain the rate of postpartum haemorrhage in a group of women given oral misoprostol at the end of the second stage of labour. Postpartum haemorrhage was defined as an estimated blood loss of geqslant R: gt-or-equal, slanted 500 mL measured by the midwife. Secondary outcomes included severe postpartum haemorrhage (> 1000 mL); secondary postpartum haemorrhage; need for blood transfusion; and need for further oxytocic drugs. We also recorded the length of the third stage, the rate of manual removal of placenta, and the need for subsequent evacuation of the uterus. In those women who had an instrumental delivery or blood loss of geqslant R: gt-or-equal, slanted 500 mL, the change in haemoglobin concentration and packed cell volume (compared with values obtained in late pregnancy) was calculated. Potential adverse effects of misoprostol were also recorded: vomiting, diarrhoea, shivering and hypertension (defined as a diastolic blood pressure geqslant R: gt-or-equal, slanted 100 mmHg or a systolic blood pressure geqslant R: gt-or-equal, slanted 150 mmHg). These outcome measures were documented by the midwife on a form which included demographic information and information concerning the woman's labour and delivery.

Exclusion criteria included placenta praevia; multiple pregnancy; intrauterine death; gestational age less than 32 weeks; women with a history of postpartum haemorrhage; women in their sixth pregnancy or more; caesarean section; and pre-eclampsia.

Immediately after the birth of the infant and clamping of the cord, the women were asked to swallow 600 μg misoprostol. Syntometrine was not given, but the midwives were instructed to do so if they felt that there was a clinical indication. Otherwise the delivery of the placenta was managed actively according to hospital policy. Blood loss was estimated clinically by the attending midwife or obstetrician. Two days after delivery a sample of venous blood for measurement of haemoglobin and haematocrit was obtained from women who had had an instrumental delivery or a blood loss geqslant R: gt-or-equal, slanted 500 mL. A record was kept of the last measured blood pressure and temperature before birth and one hour after delivery.

The study protocol was reviewed and approved by the local ethics committee. Information sheets explaining the aims of the study were distributed to potential participants in antenatal classes, antenatal clinics, and the day assessment area. On admission to the labour ward women were reminded of the study and written informed consent was obtained if they still wished to participate.

Statistical methods

Categorical variables are summarised as numbers and percentages, whereas continuous variables that are normally distributed are presented as means and standard deviations. Continuous variables that are not normally distributed are presented as medians and interquartile ranges (IQR). The paired t test was used to compare values of systolic blood pressure, diastolic blood pressure, haemoglobin, haematocrit and temperature before and after delivery. Association between two categorical variables was assessed using the χ2 test.


Two hundred and thirty-seven consecutive women were recruited to the study. Their characteristics are summarised in Table 1. The median length of the third stage of labour was 5 min (IQR 4–7). Thirteen patients (6%) had blood loss geqslant R: gt-or-equal, slanted 500 mL, and no woman had a blood loss of geqslant R: gt-or-equal, slanted 1000 mL. Secondary postpartum haemorrhage was not reported in any case. The median blood loss was 200 mL (IQR 150–300 mL). Four women (2%) required a manual removal of the placenta, and four had a postnatal haemoglobin concentration < 9 g/dL. Two women required a blood transfusion: one for a broad ligament haematoma which was managed conservatively and the other during manual removal of the retained placenta.

Table 1.  Demographic characteristics and labour variables. Values are given as n (%) or mean [SD], as appropriate.
 n = 237
Demographic characteristics 
Maternal age (years)29.1 [5.8]
Maternal weight (kg)76.3 [11.5]
Maternal height (cm)163.2 [6.7]
Primigravidae113 (48)
Gestational age at delivery (weeks)39.7 [1.4]
Labour variables 
Spontaneous onset190 (80)
Augmentation of labour72 (30)
Epidural analgesia112 (47)
Narcotic analgesia90 (38)
Episiotomy52 (22)
First and second degree tear106 (45)
Instrumental vaginal delivery67 (28)
Length of 1 st stage (h)6.7 [3.9]
Length of 2nd stage (min)66 (60)
Birthweight (kg)3.30 [0.50]

The effects of misoprostol on the third stage are given in Table 2. Gastrointestinal side effects were infrequent. Vomiting occurred in 19 women (8%) during the first hour after delivery and loose stool in seven women (3%) during the first 24 h after delivery. These symptoms were usually mild and did not require any treatment. Shivering occurred in 148 women (62%). This was self-limiting, starting approximately 20 min after swallowing the tablets and lasting for 10–15 min. There were no cases of infection and no patient required surgical evacuation of the uterus. Table 3 shows the differences in systolic and diastolic blood pressure, haemoglobin, haematocrit and temperature before and after delivery. Temperature increased significantly by 0–5°C (SE 0.05, P= 0.001).

Table 2.  Effect of misoprostol on the third stage and its complications. Values are given as n (%) or median [interquartile range]. BP = blood pressure.
Blood loss > 500 mL13 (6)
Blood loss >1 000 mL0 (0)
Blood loss (mL)200 [150-300]
Manual removal of placenta4 (2)
Required syntometrine12 (5)
Systolic BP > 1 50 (mmHg)2 (1)
Diastolic BP > 100 (mmHg)0 (0)
Haemoglobin < 9 (g/dL)4 (2)
Third stage length (min)5 [4-7]
Third stage length > 30 min1 (0-5)
Vomiting19 (8)
Diarrhoea7 (3)
Shivering148 (62)
Table 3.  Maternal blood pressure (mmHg), haemoglobin (g/dL) and temperature measurements (°C) before and after delivery. Values are given as mean (SD) or mean [SE]. Key as for Table 2.
 nBefore deliveryAfter deliveryDifferenceP
Systolic BP (mm Hg)184119.3 (13)118.4 (12)−1.0[0.9]0.25
Diastolic BP (mm Hg)18473.8 (8)73.0 (8)−0.8[0.6]0.20
Haemoglobin (g/dL)7611.3 (0.9)11.0 (1.3)−0.3[0.15]0.06
Haematocrit730.34 (0.03)0.33 (0.04)−0.01[0.005]0.047
Temperature (°C)18236.6 (0.4)37.1 (0.7)−0.5[0.05]0.001


This study shows that with misoprostol the rate of postpartum haemorrhage is low. The rates of postpartum haemorrhage (6%), need for further therapeutic oxytocics (5%), and the length of the third stage of labour (median 5 min) are lower than those reported when the third stage is managed physiologically and comparable to results obtained with Syntometrine19. Systematic reviews of prophylactic administration of oxytocic drugs in the third stage of labour show that the rate of postpartum haemorrhage is reduced, from about 18% to 5%4,19,20. In addition, the need for therapeutic oxytocic drugs is reduced from 30% to 6%4,19,20, and the length of the third stage reduced from 15 min to 5 min4,19,20. Although our study was not a randomised trial, its results suggest that misoprostol may be an effective prophylactic oxytocic drug in the third stage of labour, a hypothesis which should be tested by a double-blind randomised trial to compare oral misoprostol and intramuscular Syntometrine.

If misoprostol and Syntometrine are similarly effective in the prevention of postpartum haemorrhage, the lack of hypertension with misoprostol may be an advantage. Syntometrine was found to increase the incidence of hypertension (diastolic blood pressure more than 100 mmHg) by a factor of five compared with oxytocin1. Inadvertent administration of Syntometrine rather than oxytocin to women with pre-eclampsia has been reported as a precipitating factor in postpartum eclampsia21. Maternal death from cardiac arrest and intracerebral haemorrhage22 have been attributed to the vasoconstriction caused by ergometrine as have nonfatal instances of cardiac arrest and myocardial infarction23. Because our observational study was preliminary we excluded women with pre-eclampsia, but it may be that misoprostol can be used safely in these women. This must be the subject of future studies.

Another advantage of oral misoprostol may be its lack of unpleasant side effects. The incidence of vomiting in our study is about half that when Syntometrine is used1,20. We identified a side effect of misoprostol that has not been described previously. Shivering occurred in 62% of the women and was generally regarded as a nuisance rather than serious morbidity. In the induction of abortion larger dosed of misoprostol were not associated with this side effect24,25. Shivering is a recognised symptom after normal delivery and is more common with epidural anaesthesia, occurring in 33% to 60% of deliveries26,27. The incidence after normal delivery without epidural anaesthesia is about 10%28. In our study there was no significant difference in shivering between the women who had epidural anaesthesia and those who had not. Further work is needed to assess whether shivering is due to misoprostol; this work is in progress in our department.

The prophylactic use of oxytocics in the third stage of labour is of particular relevance to obstetric practice in the third world where atonic postpartum haemorrhage is common due to high multiparity and prolonged labour. Death due to postpartum haemorrhage accounts for 17% to 40% of the maternal mortality in some parts of the world29,30. The availability of an oral and thermostable preparation for routine management of the third stage of labour may have considerable benefits in preventing postpartum haemorrhage, and perhaps reduce maternal morbidity and mortality in developing countries. We urgently need a double-blind randomised trial of oral misoprostol and intramuscular Syntometrine in the prevention of postpartum haemorrhage.