Misoprostol for the prevention and treatment of postpartum haemorrhage

Misoprostol for postpartum haemorrhage: a global issue 

The use of misoprostol for preventing or treating postpartum haemorrhage (PPH) has become a high-profile issue globally. Apart form the usual clinical and public health considerations, the subject has been complicated by sensitivities surrounding the off-label use of misoprostol during pregnancy, dissociation of the original patent-holding company from the drug evaluation process and the global imperative to reduce maternal mortality as a matter of urgency.1 This chapter presents an overview of the topic, addressing all these issues.

Postpartum haemorrhage: pathopysiology 

Placental bed haemostasis following childbirth is a remarkable physiological process. During pregnancy, trophoblast invasion of the maternal spiral arterioles creates wide-diameter, non-contractile vessels, ensuring a high volume blood flow to the placenta. After the birth of the baby, the placenta separates from the uterine wall and the severed placental blood vessel lumina are compressed by extrinsic pressure from the surrounding mesh of myometrial fibres. Central to this process is the efficient and sustained contraction of the myometrium, under the influence of endogenous uterotonic hormones.

This physiological process is imperfect. Under the relatively ideal circumstances of low-risk women enrolled in randomized clinical trials of active versus expectant management of the third stage of labour, in the ‘physiological’ (natural third stage) group estimated blood loss in excess of 1000mL occurred in 83 of 3158 women (2.6%).2 With routine clinical intervention, including administration of a uterotonic drug and controlled traction on the clamped umbilical cord, this number was reduced to 27 of 3126 women (0.9%) [relative risk (RR) 0.33, 95% confidence interval (CI) 0.21 to 0.51]. Clinical estimation has been found to underestimate blood loss by at least 50%.

Uterotonic drugs 

Effective uterotonic drugs have existed for many years. Systematic review of randomized trials found that prophylactic administration of oxytocin reduced severe PPH from 83/1136 (7%) to 48/1107 (4.3%) (RR 0.61, 95% CI 0.44 to 0.87).3 Prophylactic use of intravenous ergot alkaloids reduced severe PPH from 11/724 (1.5%) to 1/705 (0.14%) (RR 0.09, 95% CI 0.01 to 0.72).4 In randomized trials comparing oxytocin–ergometrine with oxytocin alone, oxytocin–ergometrine was marginally more effective than oxytocin alone but with more side-effects.5

Postpartum haemorrhage: disease burden 

The disease burden associated with imperfect postpartum haemostasis is immense.6 PPH is a major cause of maternal mortality in low-income countries. Rates as high as 40 maternal deaths per 100,000 births in parts of sub-Saharan Africa7 are in stark contrast to the approximately 1 in 100,000 births in the United Kingdom. The potential to save mothers' lives with medical interventions for haemorrhage is thus considerable.8 No significant reduction in rates of PPH has been reported by industrialized countries in recent times.9

In contrast to the considerable body of randomized trials of preventive measures, there is remarkably little information from randomized trials concerning the treatment of PPH. Uterotonics are generally used empirically for treatment, extrapolating from their demonstrated effectiveness in reducing blood loss when used prophylactically.

Although PPH can be caused by factors such as genital tract injury, infection, retained products of conception and coagulopathy, the most important cause is uterine atony. Thus, a crucial aspect of both prevention and treatment of PPH is uterotonic therapy. The most commonly used agents are injectable oxytocin and/or ergometrine. Why should we be considering the use of misoprostol for the prevention or treatment of PPH when well-established, effective uterotonics already exist?

Attempts to reduce deaths from PPH worldwide are complicated by the fact that: (1) many deaths occur in out-of-hospital settings; (2) death can occur within a short space of time, before transfer to a health facility is possible; and (3) the primary methods of prevention and treatment have depended on injectable uterotonics, which are often not available at a community level.

Misoprostol 

Misoprostol is a methyl ester (a synthetic analogue) of natural prostaglandin E1 additionally methylated at C16. It is well absorbed from the stomach, appearing in the circulation within 90 seconds, and undergoes rapid de-esterification to its biologically active metabolite, misoprostolic acid (MPA). It is thermostable and is relatively inexpensive. Misoprostol has been shown to stimulate uterine contractility in early pregnancy10 and at term.11 Administered orally or vaginally, it is an effective agent for the induction of abortion12 and of labour13, although it is not without problems and risks.14, 15, 16

The development of misoprostol as a drug for use during pregnancy is unusual in that from the outset the patent-holding company distanced itself from the process. Most developments have occurred through the opportunistic efforts of clinicians, researchers, consumers and other advocates.

Pharmacokinetic studies 

Zieman et al randomized ten pregnant women undergoing first-trimester abortions and ten non-pregnant women to oral or vaginal administration of misoprostol 400μg.18 Danielsson et al studied plasma misoprostol levels after administration of 200μg (six women, data not included in Table 1) or 400μg misoprostol (12 women) orally or vaginally in the first trimester of pregnancy.19 Tang et al compared the pharmacokinetic parameters of four different routes: sublingual, oral, vaginal and vaginal with addition of water (data not included) in 40 pregnant women undergoing suction termination of pregnancy.20 Abdel-Aleem et al measured serum and colostrum misoprostol acid levels after misoprostol 600μg orally in the postpartum period.21 Misoprostol acid appeared in the serum at 2 minutes (92pg/mL) and reached peak levels after 20 minutes (345+269pg/mL), falling to low levels by 2 hours (28+15pg/mL). Misoprostol appeared in the colostrum in small amounts (21+13pg/mL). Khan et al randomized 20 postpartum women to rectal or oral administration of misoprostol 600μg.22 Andolina et al studied 20 postpartum women given 400μg postpartum as tablet or crushed in methylcellulose capsules (capsule data not included).23

Table 1. Serum misoprostol values and time course expressed as mean (standard deviation) following misoprostol administration by various routes.

	Author [ref. no.]	Dose (n)	Route of administration		Rectal	Vaginal
			Oral	Sublingual
	Time to peak concentration (minutes)
	Zieman18	400μg (20)	34 (17)			80 (27)
	Danielsson19	400μg (12)	30			60–120
	Tang20	400μg (40)	28 (15)	26 (12)		‘Longer’
	A-Aleem21	600μg (20)	20
	Khan22	600μg (20)	18 (8.8)		40.5 (16)
	Peak concentration (pg/mL)
	Zieman18	400μg (20)	227 (124)			165 (86)
	Danielsson19	400μg (12)	+280*			+65*
	Tang20	400μg (40)	±288 (144)	575 (251)		±125 (54)
	A-Aleem21	600μg (20)	345 (269)
	Khan22	600μg (20)	328 (103)		184 (64.5)
	Andolina23	400μg (10)	381
	Area under curve to 4hours (pg/hours/mL)
	Zieman18	400μg (20)	273 (110)			503 (297)
	Khan22	600μg (20)	190 (125)		311 (141)
	Area under curve to 6 hpirs (pg/hours/mL)
	Zieman18	400μg (20)	300 (103)**			957 (542)**
	Tang20	400μg (40)	403 (152)	744 (291)		434 (183)

* Estimated from published graph. ** 10 non-pregnant women only.

The data in Table 1 suggest a more rapid time to peak concentration with oral and sublingual than with vaginal or rectal administration, highest peak concentrations with sublingual, and greatest area under the time curve with sublingual and vaginal administration.

Physiological studies 

Chong et al measured intrauterine pressures in 57 healthy women after delivery.24 After 30 minutes of baseline recording, the women sequentially received syntometrine 1mL or misoprostol 200, 400, 500, 600 or 800μg (10 per group except the last group, which contained 7 women). The seventh woman receiving 800μg developed life-threatening hyperthermia. The onset of increased uterine activity was significantly slower with misoprostol than with syntometrine. There were no significant differences between groups for the increase in uterine activity or the duration of action. Danielsson et al studied intrauterine pressure in 30 women after administration of 200 or 400μg misoprostol orally or vaginally in the first trimester of pregnancy.19 Time to onset and to maximum tonus, respectively, were as follows: 400μg orally: 7.8+3.0 minutes and 25.5+5.0 minutes; 400μg vaginally: 20.9+5.3 minutes and 46.3+20.7 minutes. The initial increase in tonus was more pronounced after oral than after vaginal administration. Aronsson et al studied intrauterine pressure in 32 women at 8 to 11 weeks' gestation, for 4 hours following misoprostol 400μg orally, vaginally or sublingually.25 The time to increased tonus was shorter with oral (7.8) and sublingual (10.7–11.5) than with the vaginal route (19.4 minutes). The time to maximum tonus was also shorter (39.5, 47.1–51.7 and 62.2 minutes, respectively). Increase in uterine activity after 2 hours was greater for the sublingual and vaginal than the oral route.

Labour induction studies 

To evaluate the relative clinical efficacy of misoprostol by various routes, we systematically reviewed randomized clinical trials in which similar dosages of misoprostol were administered by different routes for labour induction at term. Outcomes related to the uterine activity response to misoprostol were compared. Data were combined using the Review Manager (RevMan) computer program, version 4.2 for Windows (The Nordic Cochrane Centre, Copenhagen) and a fixed effects model (there was no significant heterogeneity).

In three studies, involving a total of 587 women, failed vaginal delivery in 24 hours was less with the vaginal than the oral route (RR 0.60, 95% CI 0.49 to 0.73). In four studies (548 women), induction or randomization to delivery time was also less (weighted mean difference −4.8, 95% CI −6.4 to −3.2 hours). In one study of 100 women, failed vaginal delivery in 24 hours was less with the sublingual than the oral route (RR 0.56, 95% CI 0.37 to 0.84), as was the induction to delivery time (mean difference −8.3, 95% CI −15.4 to −1.2 hours).26 Overall, there was consistently greater clinical efficacy, at equivalent doses, with the vaginal and buccal/sublingual routes than with the oral route. In one small study, similar efficacy for labour induction was found when misoprostol was administered vaginally and rectally.27

Cervical ripening study 

In a prospective, randomized clinical trial, 150 women received 400μg misoprostol 3 hours prior to suction evacuation of the uterus by the sublingual, oral or vaginal route (50 women per group).28 The basal cervical dilatation prior to evacuation was higher with the sublingual (9.9±2.1mm; P<0.001) than the oral (8.2±2.6mm, 4.9±1.7 minutes) or vaginal routes (7.6±2.6mm, 5.2±1.8 minutes).

Discussion 

There is no doubt that misoprostol has strong uterotonic effects. In the context of PPH, the important question is whether a route and dose can be found with rapid absorption, a clinically significant effect and acceptable side effects. To date, the pharmacokinetic data indicate that oral and sublingual routes have the advantage of rapid onset of action, whereas the sublingual, vaginal and rectal routes have the advantage of prolonged activity and greater bioavailability. The increased bioavailability of non-oral routes is thought to be contributed to by the avoidance of the first pass intestinal–hepatic circulation. As clearance of the drug is likely to be rapid irrespective of the route of administration, the prolonged activity of the vaginal and sublingual routes is presumably due to continued absorption over an extended period of time. Duration of absorption is therefore likely to correlate with the retention of the tablet in the respective site over time. This could be a limitation of the sublingual route in an unstable or unconscious patient. The vaginal route is unlikely to be suitable for management of PPH because of the difficulty of ensuring retention in the vagina and absorption in the presence of vaginal bleeding.

The physiological studies reviewed showed more rapid onset of uterine activity with oral and sublingual than with vaginal administration of misoprostol, and more sustained effects with the sublingual and vaginal routes. In the oral groups, no statistically significant difference in uterine contractility was demonstrated for dosages ranging from 200 to 800μg.

Clinical trials during pregnancy have shown that, at equivalent dosage, the vaginal and sublingual routes produce greater clinical efficacy than the oral route, and the rectal route is possibly similar to the vaginal route. Theoretical considerations therefore favour the sublingual route for further clinical evaluation.

Effectiveness 

To determine the relative effectiveness of 600μg misoprostol versus smaller dosages (e.g. 400μg), we reviewed trials comparing different dosages. In view of the lack of adequate data on blood loss >1000mL from direct randomized comparisons, we supplemented the direct data with adjusted indirect comparisons. In this method, comparisons of each of the two dosages with a similar control group (placebo or other uterotonic) from different trials are combined to provide an estimate of the relative effectiveness of the two dosages.48 The method has been shown to usually but not always agree with direct comparisons in 44 meta-analyses.49 Three-way comparisons of 600μg versus 400μg versus placebo or another uterotonic were included only in the direct comparisons.

Studies of the rectal route were excluded from the dosage studies because of uncertain absorption via this route.

For the direct and indirect comparisons the primary outcome measure was blood loss ≥1000mL. As the effectiveness outcome was blood loss measured within 1 hour, additional doses of misoprostol given after 1hour were ignored. There was adequate data on pyrexia ≥38°C from the direct data. Adjusted indirect comparisons were made for this outcome to validate the method in this study population.

All but one of the trials were conducted in settings using routine active management of the third stage of labour. To determine whether the results were affected by the use of active management, a sensitivity analysis was performed excluding the trial in which active management was not practised.35

Meta-analysis of direct and adjusted indirect data from randomized trials showed no evidence of benefit of 600μg over 400μg in terms of the outcome blood loss ≥1000mL (RR 1.02, 95% CI 0.71 to 1.48; Table 3). A sensitivity analysis excluding the trial in which active management was not practised35 produced similar results (data not shown).

Table 3. Meta-analysis of direct and adjusted indirect comparisons of misoprostol 600μg versus 400μg for risk of blood loss≥1000mL.

	Postpartum haemorrhage	Estimate RR	Lower	Upper
	Direct 600 vs. 400	0.841	0.501	1.411
	Indirect 600 vs. 400 (placebo comparisons)	1.420	0.790	2.555
	Indirect 600 vs. 400 (uterotonic comparisons)	0.931	0.599	1.446
	All	1.023	0.707	1.481

RR, relative risk, with lower and upper 95% confidence intervals.

Favourable factors 

Unfavourable factors 

•At the individual level, it is not yet certain that the benefits of misoprostol outweigh the risks.

•The lowest effective dose has not yet been determined.

•The popularity of misoprostol – because it is innovative and easy to administer – can result in its use instead of oxytocin, which is known to be more effective and safer, even where oxytocin should be available.

•At a public health level, programmes to introduce misoprostol might detract from efforts to make more effective injectable uterotonics as widely available as possible.

•Widespread use of misoprostol at delivery can result in occasional inadvertent or ill-advised use of misoprostol before birth to induce or augment labour, or in error. A single dose as low as 100μg (half a tablet) has been associated with uterine rupture in a nulliparous woman. The risk of uterine rupture appears increased when both misoprostol and oxytocin are used.53 For women with previous caesarean section, the use even of doses as small as 25μg vaginally has been associated with high rates of uterine rupture.54

•Inadvertent use of misoprostol before the birth of an undiagnosed second twin might also result in uterine rupture.

Conclusions 

The theoretical evidence reviewed suggests that the sublingual route is likely to be the most suitable for further research on the use of misoprostol in the third stage of labour. This review found a trend towards increased maternal deaths in randomized trials of misoprostol 600μg or more versus placebo or other uterotonics. Misoprostol was less effective than conventional uterotonics in the prevention of PPH. Compared with placebo, severe blood loss ≥1000mL appeared to be reduced with both 600μg and 400μg misoprostol. Compared with both placebo and other uterotonics, pyrexia (≥38°C or as defined by trial authors) was increased about three-fold with 400–500μg and about six-fold with 600μg.

In settings in which uterotonics are not currently available or in use, the first consideration should be to implement oxytocin use (possibly in single-use plastic delivery systems) as this is the gold-standard uterotonic for the prevention of PPH. If oxytocin use is not possible, misoprostol should be considered.

Given the fact that prophylactic use of misoprostol in the third stage of labour involves administration to large numbers of low-risk women in varied settings in which side effects are a major concern, the results presented support the use of a dosage not exceeding 400μg for the prevention of PPH if misoprostol is to be used. This policy would not exclude the possibility of administering an additional 200μg to women who develop PPH, should future research show a clear benefit for the higher dose.

Because of the enormous potential benefits of an effective, orally active third stage uterotonic, and the likelihood that misoprostol will be used on a large scale worldwide, further research is essential to measure the possible beneficial and harmful effects of misoprostol more accurately.

World Health Organization recommendations 

The WHO held a technical consultation on the prevention of PPH in Geneva between 18 and 20 October 2006, to discuss the various issues related to prevention of PPH and develop guidelines. This is available online at: http://whqlibdoc.who.int/hq/2007/WHO_MPS_07.06_eng.pdf.

The recommendations relating to the use of misoprostol were:

•In the context of active management of the third stage of labour, skilled attendants should offer oxytocin for the prevention of PPH in preference to oral misoprostol (600μg). (Strong recommendation, high-quality evidence.)

•In the context of the active management of third stage of labour, skilled attendants should not offer sublingual(strong recommendation, very-low-quality evidence) or rectal misoprostol (strong recommendation, low-quality evidence) for the prevention of PPH in preference to oxytocin

•In the absence of active management of third stage of labour, a uterotonic drug (oxytocin or misoprostol) should be offered by a health worker trained in its use for the prevention of PPH (strong recommendation, moderate quality). For misoprostol, this recommendation places a high value on the potential benefits of avoiding PPH and ease of administration of an oral drug in settings in which other care is not available, but notes there is only one study. The only trial relevant to this recommendation used 600μg of misoprostol. The efficacy of lower doses has not been evaluated. There is still uncertainity about the lowest effective dose and optimal route of administration.

•Key discussion points: misoprostol has unpleasant side effects, which are dose related. A dose of 400μg has been shown to be effective in preventing PPH but has not been compared directly with 600μg. Most trials have used 600μg because the largest trial by WHO has used that dosage. It may be prudent to use the lowest effective dose to avoid undesirable side effects but this has to be determined based on further trials.

Practice points

Preventing PPH

•Routine active management of the third stage of labour is recommended.

•Oxytocin 10 units is more effective than misoprostol for routine use in the third stage of labour, and has fewer side effects.

•Data on the effectiveness of misoprostol versus placebo are conflicting.

•The use of misoprostol has been recommended by WHO in situations in which active management of third stage of labour is not practised.

•Most trials have empirically chosen 600μg orally for preventing PPH. Limited data, and direct and adjusted indirect comparisons, suggest that 400μg has similar efficacy with fewer side effects.

Treating postpartum haemorrhage

•Ergometrine is somewhat more effective than oxytocin but with more risks and side-effects. This profile supports its use in the treatment of PPH rather than for routine prophylaxis.

•Limited evidence suggests that misoprostol can reduce blood loss when used in addition to the conventional treatment of PPH.

•There is to date no evidence that the use of misoprostol for the prevention or treatment of PPH reduces mortality. In randomized trials conducted to date, there is a trend to increased mortality.