Misoprostol for the management of postpartum haemorrhage

For decades, oxytocin and ergometrine have been the treatments of choice for postpartum haemorrhage caused by ineffective uterine contraction (uterine atony). Although both drugs are effective, oxytocin is more widely used because it has fewer side effects and can be used safely in women with hypertension and pre-eclampsia. However, in their usual form both drugs can be given only by injection, and both require refrigeration. They are therefore of limited availability and benefit in low resource settings, especially in rural areas. Misoprostol, an orally active and heat stable prostaglandin E₁ analogue, has therefore emerged as a popular alternative. Until a year ago, there was limited evidence for its ability to treat postpartum haemorrhage.1 2 However proponents have argued that it should be “parachuted in” to high risk areas despite the lack of evidence.3 This, in part, has been responsible for its inclusion in multiple guidelines on postpartum haemorrhage both in rich and poor settings (despite the call in a systematic review for more studies2).

Since the systematic review of the treatment of postpartum haemorrhage was last updated in 2007,2 three large double blind randomised trials have been published.4 5 6 Few research teams had been able to carry out a randomised trial of treatment for this condition, but Gynuity Health Projects and the World Health Organization, with backing from the Bill and Melinda Gates Foundation, were able to recruit more than 80 000 women in 14 centres worldwide to three trials to define the role of misoprostol. Although these trials are impressive, they have gone largely unnoticed by many maternity care workers.

The first compared 800 µg sublingual misoprostol with 40 IU oxytocin given in a litre of intravenous solution over 15 minutes for the treatment of postpartum haemorrhage in women who had not received oxytocin prophylaxis.4 The study recruited 9348 subjects; 10% of them were diagnosed with postpartum haemorrhage (around 700 ml of blood loss) and received the study treatments. Further bleeding of at least 300 ml (1 L total) occurred in 30% of the women given misoprostol and in only 17% of women given oxytocin (relative risk 1.78, 95% confidence interval 1.40 to 2.26). Misoprostol was associated with more side effects—“intolerable shivering” was seen in 11% of women receiving misoprostol compared with less than 1% of women taking oxytocin (55.2, 7.70 to 397).

The second trial used the same protocol but in women who had received routine prophylaxis with oxytocin.5 Evidence of the benefit of prophylaxis with oxytocin was overwhelming—only 3% (809/31 055) of women bled compared with 10% in the trial above where no prophylaxis was available. In this second trial, additional blood loss of 300 ml or more after treatment was similar in the two groups (34% v 31%; 1.12, 0.92 to 1.37), whereas blood loss of more than 1 L after treatment occurred in 11 (3%) women managed with misoprostol and three (1%) women given oxytocin (3.62, 1.02 to 12.89). Intolerable shivering occurred in 4% and less than 1% of women treated with misoprostol and oxytocin respectively (16.8, 2.25 to 125). These findings suggest that 800 µg sublingual misoprostol is a possible alternative to 40 IU intravenous oxytocin for the management of postpartum haemorrhage after prophylactic oxytocin, but that it does have more side effects.

The third study assessed the effect of using misoprostol in addition to conventional injectable uterotonics to treat postpartum haemorrhage.6 The study compared 600 µg sublingual misoprostol to placebo in 1422 women who were being treated with 10 IU intramuscular or slow intravenous oxytocin for the treatment of postpartum haemorrhage. The study found no significant difference between the two treatment groups in the proportion with blood loss of 500 ml or more within 60 minutes (14% in both treatment groups; 1.02, 0.79 to 1.32) or blood loss greater than 1000 ml (1% in both treatment groups; 1.02, 0.41 to 2.55). Consistent with the other trials, side effects were more common with misoprostol than with placebo.

Following on from WHO studies nearly 10 years ago showing that misoprostol was less effective than oxytocin for prophylaxis,7 the results of these studies were disappointing for misoprostol enthusiasts. Not only is it less effective than oxytocin, but it has more side effects and no adjunctive effect if the woman has already been given oxytocin. The only comfort is that detailed examination of the data, along with the excellent outcomes for the participants, suggests that misoprostol is better than nothing.

So is there any remaining role for misoprostol in the management of postpartum haemorrhage? In settings where oxytocin is freely available it should be used instead of misoprostol for prophylaxis. And although the two drugs have similar efficacy after oxytocin prophylaxis, there is no benefit of providing a second drug that is commonly more expensive, has more side effects, and has no additional effect.

In rural low resource settings, however, where injectable oxytocics are rarely available, misoprostol is an important weapon in the fight against postpartum haemorrhage related mortality. Its heat stability and ease of use mean that all midwives and doctors in these settings should carry a stock. Furthermore, recent observational studies in women having home births in Nepal and Afghanistan suggest that giving misoprostol to women antenatally for self administration immediately after delivery may be a safe and effective strategy.8 9 A large placebo controlled randomised trial is now under way to test this hypothesis. If true, this would provide an effective self administered treatment for the first time to those women most at risk of death from postpartum haemorrhage, and it could help reduce maternal mortality worldwide.