A C TA Obstetricia et Gynecologica

AOGS M A I N R E SE A RC H A R TI C LE

No reduction of manual removal after misoprostol for

retained placenta: a double-blind, randomized trial
GIEL VAN STRALEN1, MARIEKE VEENHOF1, CAS HOLLEBOOM3 & JOS VAN ROOSMALEN1,2
1
Department of Obstetrics, Leiden University Medical Centre, Leiden, 2Department of Medical Humanities, EMGO Institute
for Health and Care Research, VU University Medical Center, Amsterdam, and 3Department of Obstetrics and Gynecology,
Bronovo Hospital, Den Haag, the Netherlands

Key words Abstract

Manual removal, misoprostol, postpartum
hemorrhage, prostaglandin E1, retained
placenta
Correspondence
Giel van Stralen, Leiden University Medical
Center, Department of Obstetrics K6-27,
P.O. Box 9600, 2300 RC Leiden, the
Netherlands. E-mail: gielvs@gmail.com
Conflict of interest
The authors have stated explicitly that there
are no conflicts of interest in connection with
this article.
Please cite this article as: van Stralen G,
Veenhof M, Holleboom C, van Roosmalen J.
No reduction of manual removal after
misoprostol for retained placenta: a double-
blind, randomized trial. Acta Obstet Gynecol
Scand 2013; 92:398–403.
Received: 7 June 2012
Accepted: 6 December 2012
Objective. To test the effect of 800 lg of misoprostol orally on the prevention
of manual removal of retained placenta. Design. Multicenter, double-
blinded, placebo-controlled, randomized trial. Setting. One university and one
non-university teaching hospital in the Netherlands. Sample. 99 women with
retained placenta (longer than 60 min after childbirth) in the absence of post-
partum hemorrhage. Methods. Eligible women were administered either 800 lg
of misoprostol or placebo orally. Main outcome measures. Number of manual
removals of retained placenta and amount of blood loss. Results. Manual
removal of retained placenta was performed in 50% of the women who
received misoprostol and in 55% who received placebo (relative risk 0.91, 95%
confidence interval 0.62–1.34). No difference in the amount of blood loss (970
vs. 1120 mL; p = 0.34) was observed between the two groups. Conclu-
sions. Administration of 800 lg of oral misoprostol, one hour after childbirth,
does not seem to reduce the number of manual removals of retained placentas.
The time elapsing results in the delivery of 50% of the retained placentas at
the expense of an increased risk of postpartum hemorrhage.
Abbreviations: PPH, postpartum hemorrhage.

DOI: 10.1111/aogs.12065

1/100 000 live births in western countries. In low-income

Introduction
The necessity of finding alternative measures in the man-
agement of retained placenta is well stated by Nardin: “If
a retained placenta is left untreated, there is a high risk of
maternal death. However, manual removal of the placenta
is an invasive procedure with its own serious complica-
tions of hemorrhage, infection or genital tract trauma”
(1). The conclusion of this Cochrane review emphasizes
the importance of reducing the number of manually
removed retained placentas.
Retained placenta (the placenta has not been delivered
within one hour after birth of the newborn) complicates
delivery in 0.5–3% of all vaginal births (1–3). Mortality is
countries, however, case fatality rates as high as 3–6% are
reported (2). Although 5–10 units of oxytocin (intra-
Key Message
Administration of 800 lg of oral misoprostol one
hour after childbirth does not seem to reduce the
number of manual removals of retained placentas.
The time that elapses leads to delivery of 50% of the
initially retained placentas, but at the expense of an
increased risk of hemorrhage.

ª 2013 The Authors

398 Acta Obstetricia et Gynecologica Scandinavica ª 2013 Nordic Federation of Societies of Obstetrics and Gynecology 92 (2013) 398–403
G. van Stralen et al.
muscularly or intravenously) can shorten the duration of
the third stage of labor, it does not prevent retained
placenta (4). Currently there are no scientific data from
randomized controlled trials on misoprostol in the man-
agement of retained placenta. In a pilot study the effect
of misoprostol in treating retained placenta seemed
promising (5).
With this randomized placebo controlled trial we
wanted to test if 800 lg of misoprostol (prostaglandin
E1) could be a safe alternative for manual removal of
retained placenta. Its favorable characteristics (low price
and heat tolerance) could be of value in both high- and
low-income countries. We chose the oral route of admin-
istration because of the favorable pharmacokinetics of
misoprostol after oral intake (6,7).
Material and methods
We conducted a double-blind, multicenter randomized
trial over the period February 2008 to September 2011 in
Bronovo Hospital, the Hague (a non-university teaching
hospital), and Leiden University Medical Centre (LUMC),
both in the Netherlands. These hospitals represent a
mixed risk and high-risk population of 90% Caucasian
origin, including women with retained placenta after
home delivery.
Data on all women with retained placenta, in the
absence of excessive blood loss (as judged by the attend-
ing obstetrician), who were at least 25 weeks pregnant,
were collected. Minimum participation age was 18 years
and all women had to be fluent in the Dutch language
verbally and in writing. In the case of postpartum hemor-
rhage (PPH), defined as loss of more than 1000 mL blood
within 60 min after birth of the newborn, the patient was
not considered eligible and standard management of PPH
with retained placenta was advised. Women were equally
randomized (1:1) to receive misoprostol or placebo. Eligi-
ble women who refused participation or women who were
erroneously not included, were identified by searching the
electronic databases of the hospitals for women with a
third stage of labor longer than 75 min.
During the first 60 min after birth of the newborn,
women were treated in accordance with the local hospital
protocol. When the third stage had lasted up to 45 min,
women were informed about the study and asked for
consent both verbally and in writing. At 60 min after
delivery of the newborn a final attempt was made to deli-
ver the placenta by controlled cord traction. Only if this
attempt failed was the woman included in our trial.
After inclusion, a blank envelope was drawn containing
study medication and a case record form. The Leiden
University Medical Center pharmacist prepared the enve-
lope and study medication. The study medication was
ª 2013 The Authors
Acta Obstetricia et Gynecologica Scandinavica ª 2013 Nordic Federation of Societies of Obstetrics and Gynecology 92 (2013) 398–403
Management of retained placenta
provided in a single number coded container, which con-
tained two blank pills containing either misoprostol
800 lg (in total) or placebo. After the final attempt to
deliver the placenta, the blank pills were dissolved in
water. The obtained solution was administered orally to
the mother between 60 and 65 min after birth of the neo-
nate. At 30 and 45 min after administration of the medi-
cation, or in the case of clinical signs of placental
separation, an attempt to deliver the placenta was made.
If the final attempt to deliver the placenta (45 min after
administration of the study medication) failed, manual
removal was performed.
During two hours after administering the study medi-
cation, women were asked to report complaints of nau-
sea, vomiting, abdominal pain, headache, dyspepsia,
shaking and dizziness. These side effects of misoprostol
were noted in the case record form. We choose two hours
to record these side effects as the half-life (T½) of
misoprostol is 20–40 min. The Tmax of misoprostol is
30 min (8). All women had a clinical review examination
six to eight weeks postpartum, or earlier in case of symp-
toms, to identify postpartum endometritis, late hemor-
rhage and placental remnants.
Blood loss was measured according to local practice,
i.e. collecting and weighing all blood, including in swabs
and drapes. In case of excessive blood loss after inclusion
and taking the study medication the same management
was advised, but all patients remained in the study on an
intention-to-treat principle. It was possible to retrieve
information about whether misoprostol or placebo had
been administered by opening an enclosed “safety enve-
lope”. This envelope was only opened in the case of clini-
cal necessity (judged by the attending obstetrician).
Two proportions power analysis was performed based
on results from our pilot study (5) and a rough estima-
tion of spontaneous delivery of the placenta due to the
time elapsed. We expected successful treatment (preven-
tion of manual removal) in 80% (8/10) of the cases
(misoprostol), where spontaneous birth of the placenta
was expected in 50% of the control group (placebo). A
statistically significant effect of the treatment was sup-
posed to be reached after inclusion of 40 women in each
group. Considering these calculations we included 100
women.
Statistical Package for the Social Sciences 20.0 (SPSS
Inc., Chicago, IL, USA) was used to analyze the acquired
data. Two-sided statistical significance for dichotomous
variables was calculated using the chi-squared test. Results
are presented as relative risks with 95% confidence inter-
vals. Differences between group means were analyzed with
the independent samples t-test. Cut-off values for statisti-
cal significance were set on p-values <0.05. Statisticians
were not blinded.
399
Management of retained placenta
Primary outcome was number of manual removals,
and secondary outcomes were amount of blood loss,
blood transfusion, PPH, interval between delivery of the
baby, and delivery of the placenta.
This study was approved by the Dutch CCMO (Cent-
rale Commissie Mensgebonden Onderzoek (Central
Counsel Human Research) under number: NL
15196.058.06 and registered with the European Union
Drug Regulating Authorities Clinical Trials (EudraCT)
under number 2006-006371-20.
Results
In the study period, 156 women were eligible for inclu-
sion in our study. Complete data were available for 99
women (Figure 1). Twenty women refused participation
and 36 were not included for logistical reasons. The two
study groups were comparable as regards age, previous
PPH, previous manual removal of retained placenta and
previous cesarean section. There were more nulliparous
women in the misoprostol group (35/48 vs. 23/51,
p = 0.003). Gestational age and birthweight were similar
in both groups (Table 1). On average, study medication
was administered 67.6 min (range 41–105) after birth of
the baby in the misoprostol group and 67.3 min (range
55–104) in the placebo group.
There was no significant difference in the proportions
of manual removal of placentas, blood loss or interval
between administration of the medication and delivery of
the placenta between the misoprostol and the placebo
group (Table 2). None of the women reported symptoms
associated with placental remnants six weeks postpartum.
Follow-up data were available for 38 (78%) women in the
misoprostol group and 44 (86%) women in the placebo
group.
Figure 1. Flowchart inclusions.
400 Acta Obstetricia et Gynecologica Scandinavica ª 2013 Nordic Federation of Societies of Obstetrics and Gynecology 92 (2013) 398–403
G. van Stralen et al.
Of the 20 women who refused to participate, 17 under-
went manual removal of the placenta with an average
blood loss of 1095 mL (range 200–2500 mL). Average age
(34.5 years), gestational age (38+5 weeks) and birthweight
(3200 g) were comparable to the study groups. There
were 13 nulliparous women in this group. The average
duration of the third stage was 112 min (range 75–
154 min). In the “missed inclusions” group (n = 36),
average blood loss (832 mL, range 150–2000 mL), age
(34 years), gestational age (38+4 weeks) and birthweight
(3048 g) were also comparable to the study group’s.
There were 15 nulliparous women in this group. The
average duration of the third stage was 105 min (range
75–174 min).
During the introductory phase of the study, seven
“safety envelopes” were opened (four misoprostol and
three placebo), not because of PPH but for breastfeeding
reasons. Initially, in the case of breastfeeding after ran-
domization for misoprostol, the first milk produced was
discarded because of potential passage of misoprostol into
the breast milk. After a literature search, breastfeeding
after misoprostol was considered safe and no further
envelopes were opened for this reason (9,10). The ethics
committee approved the addendum.
Available data on side-effects of the study medication
are presented in Table 3. Known side-effects of misopros-
tol such as nausea, vomiting and dyspepsia were recorded
more often in the misoprostol group. Shivering, a less
reported side effect, occurred significantly more often in
the misoprostol group.
Discussion
Our expectation that elapsing time would facilitate spon-
taneous birth of around 50% of the retained placentas
was confirmed in the placebo group. Unfortunately,
misoprostol did not live up to our expectations and per-
formed only marginally better than placebo. Unexpect-
edly, there were more nulliparous women in the
misoprostol group. Other variables in Table 1 do not sug-
gest irregularities in the randomization process.
Our inclusion criteria resulted in a selection of women
with retained placenta in the absence of PPH. Although
waiting for the placenta to be born for more than one
hour seems relatively safe (average total blood loss in the
two groups fluctuated around 1000 mL), 38% (miso-
prostol) and 47% (placebo) of the women suffered a PPH
of more than 1000 mL. With these results in mind, we
advise not changing the current management of the third
stage and to strive for delivery of the placenta within one
hour. Shivering was the most reported side-effect of
misoprostol. Unexpectedly, we found no significant differ-
ence in reported abdominal pain, nausea or vomiting
ª 2013 The Authors
G. van Stralen et al. Management of retained placenta

Table 1. Patient characteristics of misoprostol and placebo group.

Misoprostol group (n = 48) Placebo group (n = 51) p-Value

Mean age  SD, years (range) 32  4.15 (21–40) 33  4.42 (20–41) 0.74
Nulliparous, n 35 23 0.003*
Prior cesarean section, n 1 3 0.34
Prior PPH, n 4 2 0.36
Prior MRRP, n 2 3 0.70
Mean gestational age SD, weeks+days (range) 39+2  10.74 days (35+4–41+6) 39+3  13.48 days (34+2–42+0) 0.84
Mean birthweight SD, g (range) 3 270  540 (1637–4315) 3 312  521 (1980–4345) 0.70
Mean interval birth neonate–medication  SD, min (range) 67  13.3 (41–105) 68  11.3 (55–104)

PPH, postpartum hemorrhage; MRRP, manual removal of retained placenta; SD, standard deviation.
Chi-squared test for difference between dichotomous variables, Independent samples t-test for difference between group means.
*Statistically significant.

Table 2. Primary and secondary outcomes of misoprostol vs. placebo.

Misoprostol group (n = 48) Placebo group (n = 51) Relative risk (95% confidence interval)

Manual removal 24 (50%) 28 (55%) 0.91 (0.62–1.34)

Blood transfusion 6 (13%) 9 (18%) 0.83 (0.52–1.32)
PPH, >1000 mL 18 (38%) 24 (47%) 0.83 (0.57–1.21)

p-Value

Average blood loss, mL, 970  771 (200–3000) 1 120  949 (200–5000) 0.39
SD (range)
Average interval medication/birth 59  33 66  39 0.35
placenta, min, SD

PPH, postpartum hemorrhage; SD, standard deviation.

Chi-squared test for difference between dichotomous variables, Independent samples t-test for difference between group means.

Table 3. Available data on side effects of oral misoprostol compared hospitals, over 90% of the obstetricians wait 60 min or
with placebo. longer in the absence of bleeding (11).
Misoprostol Placebo Much effort has been gone into finding alternative
group (n = 48) group (n = 51) p-Value measures for management of retained placenta and asso-
ciated PPH (Table 4) but there are few randomized trials,
Nausea 6/42 1/32 0.10
with a small number of participants. Nitroglycerine orally,
Vomiting 3/42 0/32 0.12
Abdominal pain 8/42 8/32 0.54
intravenously or intra-umbilically, has been reported as
Headache 1/42 1/32 0.54 either highly successful (12,13) or ineffective (14). Intra-
Dizziness 5/42 3/32 0.73 venous sulprostone, described by Van Beekhuizen,
Dyspepsia 4/42 0/32 0.07 reduced the number of manually removed placentas by
Shivering 15/42 1/28 0.001* 49% (15). Because of the similar design of Van Beekhui-
Chi-squared test for difference between dichotomous variables.
zen’s and our study, it is interesting to compare our
*Statistically significant. results with theirs. Sulprostone and misoprostol appear
equally good for achieving spontaneous expulsion of the
retained placenta (52 vs. 50%). However, there was a
compared with placebo, although there was a trend for noticeable difference (18 vs. 45%) between the two
these effects to occur more often in the misoprostol group. placebo groups in spontaneous expulsion.
The definition of retained placenta is heterogeneous. Oxytocin, ergometrine, misoprostol and placental
Authors use the term if the placenta has not been deliv- drainage have been studied for the management of
ered within 20–60 min after the birth of the newborn. retained placenta (1,13,16–20). Surprisingly, in the search
The difference is not just in the wording: in Spain, 90% for a good alternative for manual removal of retained
of obstetricians decide to perform manual removal of placenta, both uterotonics and tocolytics have been exam-
the retained placenta within 30 min, whereas in Dutch ined.

ª 2013 The Authors

Acta Obstetricia et Gynecologica Scandinavica ª 2013 Nordic Federation of Societies of Obstetrics and Gynecology 92 (2013) 398–403 401
Management of retained placenta G. van Stralen et al.

Table 4. Management of retained placenta; literature overview.

Diagnosis of
retained
Author (Ref) (Year) Study type placenta (min) Agent (route) Dose Success (n)

Chedraui (12) (2003) Cohort >30 Nitroglycerin (i.v.) Max. 200 lg 30/32
Bullarbo (13) (2005) RCT >40 Nitroglycerin (oral) vs. placebo 1 mg 12/12 vs. 1/12*
Van Beekhuizen (15) RCT cohort >60 Sulprostone i.v. vs. placebo 250 lg in 30 min 28/77 vs. 4/26*
(2005)
Rogers (18) (2007) RCT >30 Oxytocin vs. misoprostol vs. placebo 50 U 4/20 vs.
(all umbilical vein) 800 lg 12/21* vs. 6/13
in 30 mL saline
Van Stralen (5) (2007) Cohort >60 Misoprostol (rectal) 800 lg 7/10
Visalyapuba (14) (2011) RCT >30 Nitroglycerin (i.v.) vs. placebo 200 lg 3/20 vs. 4/20
Lim (16) (2011) RCT >20 Oxytocin (umbilical vein) vs. 100 U 21/30 vs. 10/31*
controlled cord traction
Harara (17) (2011) RCT >30 Oxytocin vs. ergometrine vs. misoprostol 20 U 19/26 vs. 17/27 vs.
(all umbilical vein) 0.2 mg 20/25*
800 lg

i.v., intravenous; RCT, randomized controlled trial.

*Statistically significant.

Our definition of retained placenta (one final attempt
to deliver the placenta 60 min after birth of the baby)
was accurate, resulting in a cohort of women who had a
retained placenta in the absence of PPH. Unfortunately,
one-third of the eligible women were not randomized for
logistical reasons (missed inclusions) or because of a refu-
sal to participate. Women were asked for consent only if
retained placenta without PPH was likely. We chose this
timing because of the low incidence of retained placenta
the absence of PPH. Asking consent in an earlier stage
would have exposed our whole population to the study
information, knowing there is only a 1–3% chance of
inclusion.
We checked the databases of both hospitals for women
with a third stage of more than 75 min, who delivered in
the study period, to identify possible “missed inclusions”.
Data on the outcome of these women are similar to
the study group, which makes selection bias unlikely.
Although the study medication was blinded, gastro-intes-
tinal complaints and shivering are side-effects of miso-
prostol that could have indicated which type of study
medication had been administered. It is unlikely that this
interfered with the final results.
Conclusion
Among women with retained placenta in the absence of
overt bleeding, oral misoprostol did not reduce the number
of manual removals of retained placenta, although further
research with more women is needed. Patience (allowing
time) facilitated spontaneous delivery of around 50% of
retained placentas in this specific group. Considering the
associated high risk for PPH, we do not advise changing
the policy for the timing of manual removal of placenta.
Acknowledgments
We thank Clara Kolster-Bijdevaate and Marjolein Verhart
for help with this study, and Marjon de Boer for review-
ing the manuscript.
Funding
No special funding.
References
1. Nardin JM, Weeks A, Carroli G. Umbilical vein injection
for management of retained placenta (review). Cochrane
Database Syst Rev. 2011;5:CD001337.
2. Weeks AD, Mirembe FM. Editorial. The retained placenta
—new insights into an old problem. Eur J Obstet Gynecol
Reprod Biol. 2002;102:109–10.
3. World Health Organization. The prevention and
management of postpartum haemorrhage: report of a
technical working group. 3–6 July 1989. Geneva: WHO, 1989.
4. Nordstrom L, Fogelstam K, Fridman G, Larsson A,
Rydhstroem H. Routine oxytocin in the third stage of
labour: a placebo controlled randomised trial. Br J Obstet
Gynaecol. 1997;104:781–6.
5. Van Stralen G, van Roosmalen J. Regarding Roger MS,
Yen PM, Wong S. Avoiding manual removal of placenta:
evaluation of intra-umbilical injection of uterotonics using
the Pipingas technique for management of adherent
placenta. Acta Obstet Gynecol Scand. 2007;86:764.

ª 2013 The Authors

402 Acta Obstetricia et Gynecologica Scandinavica ª 2013 Nordic Federation of Societies of Obstetrics and Gynecology 92 (2013) 398–403
G. van Stralen et al.
6. Zieman M, Fong SK, Benowitz NL, Banskter D,
Darney PD. Absorption kinetics of misoprostol with
oral or vaginal administration. Obstet Gynecol.
1997;90:88–92.
7. Chong YS, Chua S, Shen L, Arulkumaran S. Does the
route of administration of misoprostol make a difference?
The uterotonic effect and side effects of misoprostol given
by different routes after vaginal delivery. Eur J Obstet
Gynecol Reprod Biol. 2004;113:191–8.
8. Farmacotherapeutisch kompas. http://www.cvzkompassen.
nl; July 2008.
9. Vogel D, Burkhardt T, Rentsch K, Schweer H, Watzer B,
Zimmermann R, et al. Misoprostol versus
methylergometrine: pharmacokinetics in human milk. Am
J Obstet Gynecol. 2004;191:2168–73.
10. Abdel Aleem H. The pharmacokinetics of the
prostaglandin E1 analogue misoprostol in plasma and
colostrum after postpartum oral administration. Eur J
Obstet Gynecol 2003;108:25–8.
11. Deneux-Tharaux C, MacFarlane A, Winter C, Zang WH,
Alexander S, Bouvier-Colle MH, the EUPHRATES group.
Policies for manual removal of placenta at vaginal delivery:
variations in timing within Europe. Br J Obstet Gynaecol.
2009;116:119–24.
12. Chedraui PA, Insuati DF. Intravenous nitroglycerin in the
management of retained placenta. Gynecol Obstet Invest.
2003;56:61–4.
13. Bullarbo M, Tjugum J, Ekerhovd E. Sublingual
nitroglycerin for management of retained placenta. Int J
Gynecol Obstet. 2005;91:288–32.
ª 2013 The Authors
Acta Obstetricia et Gynecologica Scandinavica ª 2013 Nordic Federation of Societies of Obstetrics and Gynecology 92 (2013) 398–403
Management of retained placenta
14. Visaloyaputra S, Prechapanich J, Suwanvichai S, Yimyam
S, Permpolprasert L, Suksopee P. Intravenous nitroglycerin
for controlled cord traction in the management of retained
placenta. Int J Gynecol Obstet. 2011;112:103–6.
15. Van Beekhuizen HJ, De Groot ANJA, De Boo T, Burger D,
Jansen N, Lotgering F. Sulprostone reduces the need for
manual removal of the placenta in patients with retained
placenta: a randomized controlled trial. Am J Obstet
Gynecol. 2006;194:226–50.
16. Lim PS, Singh S, Lee A, Yassin MAJM. Umbilical vein
oxytocin in the management of retained placenta: an
alternative to manual removal of placenta? Arch Gynecol
Obstet. 2011;248:1073–9.
17. Harara R, Hanafi S, Zidan MSA, Alberry M.
Intraumbilical injection of three uterotonics in the
management of retained placenta. J Obstet Gynecol Res.
2011;9:1203–7.
18. Rogers MS, Yuen PM, Wong S. Avoiding manual removal
of placenta: evaluation of intra-umbilical injection of
uterotonics using the Pipingas technique for management
of adherent placenta. Acta Obstet Gynecol Scand.
2007;86:48–54.
19. Abdel-Aleem H, Abdel-Aleem MA, Shaaban OM. Tocolysis
for management of retained placenta. Cochrane Database
Syst Rev. 2011;1:CD007708.
20. Soltani H, Poulose TA, Hutchon DR. Placental cord
drainage after vaginal delivery as part of the management
of the third stage of labour (review). Cochrane Database
Syst Rev. 2011;9:CD004665.
403