Management of the third stage of labor

Authors
Frank Silverman, MD
Eran Bornstein, MD
Section Editor
Charles J Lockwood, MD
Deputy Editor
Vanessa A Barss, MD
Last literature review version 18.3: September 2010 |This topic last updated: September 23, 2010

INTRODUCTION — the third stage of labor refers to the interval from the delivery of the baby to the
separation and expulsion of the placenta. The major complication associated with this period is postpartum
hemorrhage (PPH), which is the most common cause of maternal morbidity and mortality in developing
countries. Even in developed countries, although maternal mortality rates are much lower, PPH remains a
major concern [1]. It ranks just behind thromboembolic events and hypertensive disease as a common
cause of maternal death in women whose pregnancies continue beyond 20 weeks.
Some clinicians also recognize a fourth stage of labor, which begins at the delivery of the placenta and
lasts for an arbitrary period of time afterward. This very early postpartum period is associated with maternal
complications from hemorrhage, fluid shifts, and embolism.

THIRD STAGE PHYSIOLOGY

Normal placental separation — Placental separation occurs due to shearing of the placental surface when
the uterus contracts after the infant is delivered. Signs of placental separation include a gush of blood,
lengthening of the umbilical cord, and anterior-cephalad movement of the uterine fundus, which becomes
firmer and globular after the placenta detaches. Placental expulsion follows as a result of a combination of
events including spontaneous uterine contractions, downward pressure from the developing retroplacental
hematoma, and an increase in maternal intraabdominal pressure.
The changes in the uteroplacental unit during placental separation have been classified as [2]:
 Latent phase —the interval from delivery of the fetus to the beginning of myometrial contractions
which lead to placental separation. This is the main determinant of duration of the third stage of
labor.
  Contraction phase —this phase is characterized by contraction of the myometrium at the placental
site.
 Detachment phase — Contraction of the myometrium at the placental site leads to shearing of the
placenta off of the uterus.
 Expulsion phase —the separated placenta is extruded from the uterus into the vagina.
The phases of placental separation have been examined using dynamic real-time sonographic methods [3].
The latent phase is characterized by a thick placenta-free myometrial wall that undergoes periodic
contractions and a thin myometrial wall overlying the placenta-site. When thickening of the placenta-site
wall (from <1 cm to >2 cm) occurs at the end of this phase, basal blood flow ceases between the placenta
and myometrium, thereby preventing excessive maternal blood loss when the placenta separates.
Persistent blood flow from abnormal vascular connections occurs with abnormal placentation, such as
placenta accreta.
Thickening and reduction of the surface area of the placenta-site wall results in shearing of the placental
edge. This process generally begins at the lower pole of the placental margin and progresses along
adjacent sites of placental attachment. A "wave of separation" spreads upwards so that the uppermost part
of the placenta detaches last [4]. At this point, the placenta can be expelled from the uterus.
Myometrial strain related to placental separation has been estimated using sonographically determined
changes in uterine wall thickness and a simplified geometric model of the uterine wall [5]. In this study,
strain measurements were made before the second stage of labor and at placental separation. Placental
detachment was associated with maximal myometrial strain, thus suggesting that separation could not be
achieved before birth since the essentially incompressible uterine contents (fetus, amniotic fluid) restrains
myometrial contraction and prevents the achievement of maximal strain.
Duration — The duration of the third stage of labor is important because the prevalence of postpartum
hemorrhage increases as the duration lengthens [6,7]. There is no universally accepted criterion for the
normal length of the third stage. Two large series of consecutive deliveries showed that the average length
was five to six minutes, and that 90 percent of placentas were delivered within 15 minutes and 97 percent
were delivered within 30 minutes of birth [7,8].
Influence of gestational age — Gestational age is the major factor influencing the length of the third stage:
preterm deliveries are associated with a longer third stage of labor than term deliveries [7-10]. As an
example, one study showed that compared to term pregnancy, the frequency of retained placenta (2.0
percent overall) was markedly increased among gestations ≤26 weeks (odds ratio [OR] 20.8, 95% CI 17.1-
25.4) and, to a lesser degree, among those less than 37 weeks (OR 3.0, 95% CI 2.6-3.5) [7].
This relationship has also been demonstrated among women undergoing midtrimester termination of
pregnancy. A study of 96 women administered prostaglandin E2 vaginal suppositories for pregnancy
termination noted that the third stage was ≤15 minutes in 18 patients (19 percent) and ≤30 minutes in 36
patients (38 percent); however, 40 of 96 (42 percent) had not spontaneously delivered the placenta within
two hours of fetal expulsion [11]. Retained placenta was most likely when the fetus was at a very early
gestational age (ie, 15 to 16 weeks).

MANAGEMENT
Expectant versus active management —the optimal approach to management of the third stage of labor is
controversial. The two acceptable approaches are:
 Expectant management — Expectant or physiologic management refers to spontaneous delivery of
the placenta without the use of uterotonic agents or cord traction.
 Active management — Active management generally consists of early cord clamping, controlled
cord traction, uterine massage, and administration of an uterotonic agent prior to placenta
separation.
The benefits of active management were illustrated in a systematic review of five randomized controlled
trials comparing active versus expectant management of the third stage of labor [ 12]. Active management
was defined as administration of a prophylactic uterotonic agent before delivery of the placenta; typically
with early cord clamping/cutting and controlled traction of the umbilical cord.
Active management was associated with reduced risks of maternal blood loss (weighted mean difference
-79.33 mL, 95% CI -94.29 to -64.37); postpartum hemorrhage of more than 500 mL (RR 0.38, 95% CI 0.32-
0.46); and prolonged third stage of labor (weighted mean difference -9.77 minutes, 95% CI -10.00 to -9.53)
[12]. The authors concluded that "active management should be the routine management of choice for
women expecting to deliver a baby by vaginal delivery in a maternity hospital."
If active management is undertaken, uterotonic agents should not be given until after delivery of the
anterior fetal shoulder to ensure that shoulder dystocia is not exacerbated. The clinician should also be
certain that there is no undiagnosed and undelivered twin.
Despite evidence supporting protocols of active management of the third stage of labor, the components of
these protocols are significantly underutilized in the United States and Latin America [13,14]. However,
multifaceted behavioral intervention (including interactive workshops, hands-on training, clinical reminders,
and feedbacks) can increase adherence [15].
Drugs used for active management
Oxytocin — Administration of oxytocin is a major component of standard management of the third stage of
labor. The value of oxytocin was shown in a systematic review of seven studies (n = 3000 patients) that
compared prophylactic administration of oxytocin to no uterotonic agents in the third stage of labor [ 16].
Patients treated with oxytocin demonstrated a significant reduction in mean blood loss and postpartum
hemorrhage (RR for blood loss > 500 mL 0.50; 95% CI 0.43-0.59). However, no difference in the length of
the third stage was detected.
The optimum timing of oxytocin administration is unclear [17]. It has been given before placental separation
to expedite the process and after placental expulsion to enhance contraction of the uterus and reduce the
volume of blood loss. Most randomized trials have administered the drug prior to placental separation;
there are fewer data on administration after placental expulsion.
 A survey of Canadian obstetricians reported 52 percent of respondents administered oxytocin prior
to delivery of the placenta [18]. A similar questionnaire given to 1500 obstetricians in Texas found
that only 15 percent administered oxytocics prior to delivery of the placenta, whereas 92 percent
did so after placental expulsion [19].
 A trial comparing the two regimens included 1486 women who were randomly assigned to receive
oxytocin (20 units in 500 mL of normal saline) upon delivery of the anterior shoulder or after
delivery of the placenta [20]. A modified Brandt maneuver (an abdominal hand secures the uterine
fundus to prevent uterine inversion while the other hand exerts sustained downward traction on the
umbilical cord) was performed in all patients. There were no significant differences between the
two groups in blood loss or retained placenta.
Route and dose — Oxytocin is usually infused into a maternal vein. As discussed above, the dose and
duration of intravenous oxytocin as a prophylactic agent varies considerably among institutions. A solution
of 10 to 20 units of oxytocin in 500 or 1000 mL 0.9 percent saline is commonly used [ 20-22]. It is our
practice to infuse 20 units of oxytocin in one liter of fluid at a rate of 125 mL/hour following the delivery of
the placenta. If there is no PPH and the mother is stable in the postpartum unit, the infusion is usually
discontinued upon completion of the infusion.
Other routes for administering oxytocin include umbilical vein injection (20 units) or intramuscular
administration of up to 10 units. The majority of data suggests that both of these approaches are similarly
effective [23-25]. Intravenous bolus (1 to 10 units) has also been used [21,22,26,27]. However, the safety of
intravenous bolus of oxytocin has been questioned due to reports of significant hypotension that may lead
to cardiovascular collapse. Even a 2.5 unit bolus of oxytocin can produce adverse effects, including
tachycardia and hypotension, in preeclamptic patients [28].
 Intraumbilical vein injection versus placebo — A randomized, double-blind trial assessed the
effectiveness of intraumbilical injection of oxytocin on the duration of the third stage [29]. In this
trial, 412 women were randomly assigned to receive intraumbilical injection of 30 mL of saline or 20
units of oxytocin in 26 mL of saline. Both groups also received prophylactic intramuscular injection
of 10 units oxytocin. Compared to the saline control group, the intraumbilical vein oxytocin group
had significantly less blood loss (195 versus 288 mL), a shorter duration of the third stage (4.5
versus 7.9 minutes), and fewer placentas undelivered after 15 minutes (0 versus 4 percent).

A smaller trial reported similar findings [30]. However, umbilical vein injection may increase the
incidence of fetomaternal transfusion, which is a concern [23].
 Intravenous slow infusion versus intraumbilical vein injection — Another trial randomly assigned
women in the third stage to receive 20 units of oxytocin via an intravenous infusion or into the
umbilical vein [23]. There was no significant difference between the two groups in either blood loss
or the duration of the third stage. However, an increased incidence of fetomaternal transfusion, as
measured by maternal serum alpha-fetoprotein levels, was observed in the intraumbilical injection
group.
 Intravenous bolus versus slow infusion — A double-blind trial randomly assigned 201 women to
receive either an intravenous bolus of oxytocin (10 units push) or an infusion (10 units in 500 mL
saline at 125 mL/h) at delivery of the anterior shoulder [21]. Bolus injection did not cause adverse
hemodynamic changes. Compared to the bolus group, the dilute oxytocin infusion group had
greater mean estimated blood loss (423.7 mL versus 358.1 mL), increased use of additional
oxytocics (35 versus 22 percent) and a greater drop in hemoglobin from admission to postpartum
(1.74 g/dL versus 1.14 g/dL). The minimum effective bolus dose of oxytocin is unclear, but may be
3 units or less [31].

 As discussed above, other trials have questioned the safety of bolus infusion. At least two trials in
women undergoing cesarean delivery reported bolus injection of oxytocin was associated with
 ischemic electrocardiogram changes [32,33]. Rapid intravenous bolus of oxytocin may also result
in significant hypotension and has been reported to cause cardiac arrest [ 34]. The confidential
Enquiry into Maternal Deaths in the United Kingdom reported significant hemodynamic changes
from administering 10 units of oxytocin as an intravenous bolus and concluded that it could have
contributed to the deaths of two women who were already hemodynamically unstable [35]. In
addition, a randomized trial that compared blood loss and hemodynamic changes in 30 women
undergoing scheduled cesarean delivery demonstrated that 5 units of oxytocin injected slowly (over
5 minutes) minimized the cardiovascular side effects of a bolus dose without compromising the
therapeutic benefits [36].
Based on these data, we recommend slow intravenous infusion of oxytocin rather than bolus injection.
Ergot alkaloids — Ergot alkaloids can be administered as single agent therapy, most commonly as
methylergonovine 0.2 mg intramuscularly; its use is contraindicated in women with hypertension, history of
migraine, or Raynaud's phenomenon.
Ergot preparations are associated with more side effects than oxytocin because they act systemically on
smooth muscle, while oxytocin is specific for uterine smooth muscle. However, they are longer lasting than
oxytocin and produce more tetanic contractions, thus they are useful in prophylaxis and treatment of
postpartum hemorrhage. A systematic review found ergot alkaloids used alone in the third stage of labor
significantly decreased mean blood loss (weighted mean difference -83 mL, 95% CI -99 to -67 mL) and the
risk of PPH of at least 500 mL (RR 0.38, 95% CI 0.21-0.69) [37]. A major disadvantage of these agents is
that the frequency of vomiting, blood pressure elevation, and pain requiring analgesia was significantly
increased in the treatment group. The injectable ergometrine is very unstable when stored unrefrigerated or
exposed to light, which limits its use in rural areas of developing countries.
Several investigators have compared the prophylactic use of ergot alkaloids to oxytocin. A systematic
review of six trials comparing these two treatment modalities involved 2800 women and detected no
significant difference in blood loss between the two groups [16]. However, the use of oxytocin was
associated with fewer manual removals of the placenta and side effects. Additionally, adding oxytocin to
ergometrine was no more effective than ergometrine alone.
Ergometrine-oxytocin — The combined ergometrine-oxytocin preparation (ie, Syntometrine: 5 units
oxytocin plus 0.5 mg ergometrine, not available in the United States) has been successfully used in the
active management of the third stage of labor. A systematic review of six trials with a total of 9332 women
determined that the use of ergometrine-oxytocin was associated with a small but statistically significant
reduction in the risk of postpartum hemorrhage of 500 to 1000 mL compared with oxytocin alone (OR 0.82,
95% CI 0.71-0.95) [38]. However, the two groups did not differ in the risk of postpartum hemorrhage of
>1000 mL, or in the duration of the third stage of labor. Maternal side effects, such as significant nausea,
vomiting, and hypertension in previously normotensive women, were more commonly associated with the
use of ergometrine-oxytocin than with oxytocin alone. Information regarding neonatal complications was
limited, but did not show any difference between the two groups.
Prostaglandins — Prostaglandins have been widely used in obstetrical and gynecological practice for
cervical ripening, termination of pregnancy (in conjunction with mifepristone), and for induction of labor. In
the management of the third stage of labor, they have mainly been used for intractable PPH as a last resort
when other medical interventions did not suffice [39]. They have not been useful initially in active
management of the third stage of labor [40-43]. (See "Overview of postpartum hemorrhage".)
A systematic review found that oral or sublingual misoprostol was more effective than placebo or no
treatment in reducing the frequency of severe PPH and blood transfusion (RR 0.31; 95% CI 0.1 to 0.94; five
oral misoprostol studies including 3519 women) [39]. However, when compared to other uterotonics, oral
misoprostol was associated with increased risk of severe PPH, use of additional uterotonics, a significant
increase in shivering and fever, but fewer blood transfusions. There were few studies comparing injectable
prostaglandins with conventional uterotonics. The review concluded that neither intramuscular
prostaglandins, nor misoprostol (oral or rectal) given prophylactically, were preferable to conventional
injectable uterotonics (eg, oxytocin, ergometrine) for management of the third stage of labor.
Use of misoprostol also has several advantages: it is inexpensive, can be administered orally, and does not
require refrigeration. Thus, use of misoprostol provides an advantage in resource-poor countries where
drugs that must be refrigerated or require needles for injection/intravenous administration may pose a
problem [44-48]. The efficacy of misoprostol in this setting was illustrated in a randomized, placebo
controlled trial in which 1620 women in rural India were given either oral misoprostol (600 mcg) or placebo
within five minutes of cord clamping [49]. Use of misoprostol was associated with significant reductions in
acute PPH (blood loss greater than 500 mL decreased from 12.0 to 6.4 percent, RR 0.53, 95% CI 0.39-
0.74), severe PPH, and postpartum blood loss within two hours of delivery. One case of PPH was
prevented for every 18 women treated.
Oxytocin agonists — Carbetocin, a long-acting synthetic oxytocin agonist, has similar pharmacologic
properties to those of natural oxytocin. It binds to smooth muscle receptors of the uterus and causes
rhythmic contractions, increases the frequency of contractions, and increases uterine tone. It is in use in
many countries (but not the United States) for prevention of uterine atony and hemorrhage [ 50-52]. A
potential advantage of carbetocin over oxytocin is its longer duration of action.
A systematic review including four studies (n = 1037 women) compared the risk of PPH in patients
receiving prophylactic oxytocin or carbetocin (100 micrograms as an intravenous bolus) in the third stage of
labor [53]. No differences in PPH were detected between the two drugs for women who underwent
cesarean delivery, or for those with risk factors for PPH who underwent vaginal delivery. The use of
carbetocin was associated with a reduction in the need for therapeutic uterotonic agents (RR 0.44, 95 % CI
0.25-0.78) compared to oxytocin in patients who underwent cesarean delivery, but not in those who
underwent vaginal delivery. Additionally, it was associated with reduced need for uterine massage in
patients who were delivered by cesarean or a vaginal delivery. A few studies reported a non-significant
trend towards greater hemoglobin drop in the oxytocin group.
Carbetocin has been reported to be as effective as Syntometrine (oxytocin 5 units and ergometrine 500
mcg), a preparation widely used in the Commonwealth, with fewer side effects [52].
Carbetocin 100 mcg may be given by a single slow intravenous injection. The toxicity spectrum is similar to
that of oxytocin. Therefore, it seems reasonable to use this drug as an alternative to oxytocin in countries
where it is available, as it is easy to administer and has a long duration of action.
Tranexamic acid — A meta-analysis of two randomized trials found that administration of tranexamic
acid (0.5 or 1 gram intravenously) significantly decreased mean postpartum blood loss and the risk of
postpartum hemorrhage [54]. In one trial, the drug was given immediately after vaginal delivery and, in the
other, it was given 10 minutes prior to incision for cesarean birth. The trials were of unclear quality and
looked at only a few outcomes. Further investigation of the safety and efficacy of this approach is needed,
given the increased risk of venous and arterial thrombosis with use of this drug.
Nonpharmacologic maneuvers — Active management of the third stage of labor consists of non-
pharmacological maneuvers, such as careful cord traction and uterine massage [55].

COMPLICATIONS — The major complications of the third stage of labor are:

 Hemorrhage
 Retained placenta
 Uterine inversion
Postpartum hemorrhage — Causes of postpartum hemorrhage after vaginal delivery are listed in the table
(table 1). An overview of the causes and medical and surgical treatment of this disorder is provided
separately. (See "Overview of postpartum hemorrhage".)
As discussed above, the prevalence of postpartum hemorrhage increases as the duration of the third stage
of labor lengthens. This was illustrated in a study including almost 13,000 consecutive patients in which the
incidence of postpartum hemorrhage remained constant over the first 30 minutes after birth, then rose,
reaching a plateau at 75 minutes [8]. In another study, the frequency of postpartum hemorrhage peaked by
40 minutes after birth, regardless of gestational age [7]. In a third study, the best predictor of postpartum
hemorrhage using a receiver operating characteristic curve was a third stage >18 minutes [56].
Retained placenta — Retained placenta is variously defined as a placenta that has not been expelled by 30
to 60 minutes after delivery of the baby [57]. It occurs in 1:100 to 1:200 deliveries and is an important cause
of PPH. A retained or partially detached placenta interferes with normal uterine contraction and retraction,
which leads to bleeding. For this reason, a retained placenta should be removed, usually manually, in an
operating room, under anesthesia or with conscious sedation. Although there is a small risk of uterine
trauma or introducing infection; these risks should be weighed against the risks of bleeding and infection
with prolonged expectant management.
Sonographic scanning is appropriate when the third stage of labor is prolonged and the etiology is
uncertain. If the placenta is already detached, as demonstrated sonographically, further traction on the
umbilical cord is warranted. If the placenta is partially or totally adherent, one may perform a manual
removal with or without ultrasonic guidance, if bleeding is not brisk. The possibility of placenta accreta
should be considered, especially in women with risk factors. Diagnosis and management of placenta
accreta are discussed separately. (See "Diagnosis and management of placenta accreta".)
Manual removal — Manual removal is performed by using one hand to follow the path of the umbilical cord
through the vagina, cervix, and lower uterine segment, while the other hand holds the uterine fundus. If the
placenta is free in the lower segment, it is removed. If it has not separated, then the fingers are used to
gently develop a space between the placenta and uterus and shear off the placenta. The placenta and
membranes are then grasped and removed manually. There are no data from randomized trials to support
or refute a benefit from prophylactic administration of antibiotics upon manual removal of the placenta [ 58].
General anesthesia may be required to relax the uterus and because manual removal is painful.
Intravascular injection of nitroglycerin (50 to 200 mcg) may be administered to relax the uterus [59-61]. This
often facilitates removal of the placenta without manual exploration and avoids the need for general
anesthesia. Blood pressure should be monitored carefully, as a drop in blood pressure always occurs,
although most patients do not become hypotensive. Trendelenburg position may help relieve symptoms
from the drop in blood pressure. Intravenous narcotics can also be administered for analgesia, if required.

Pharmacologic interventions
Intraumbilical vein oxytocin — Intraumbilical injection of a solution of oxytocin in saline has been used as
an alternative to manual removal in stable patients, but pooled data from randomized trials show it has little
to no efficacy in reducing the need for manual placental removal.
A systematic review including 12 randomized trials with a total of 1288 women found that umbilical vein
injection of oxytocin plus saline solution did not result in a clinically significant reduction in need for manual
removal of a retained placenta (RR 0.91; 95% CI 0.82-1.00) [62].
The largest trial included in the meta-analysis was a well-designed, multi-center international trial (Release
trial) that randomly assigned 577 women with retained placenta for at least 30 minutes to receive an
intraumbilical vein injection of 50 IU oxytocin or water in 30 mL saline via an umbilical vein catheter [ 62].
The umbilical cord was clamped after the injection to prevent backflow and manual removal was performed
if the placenta did not deliver with gentle cord traction 30 minutes after the injection. The authors found no
difference between groups in need for manual removal of the placenta (61.3 versus 62.1 percent with water
alone, RR 0.98, 95% CI 0.97-1.12) and no significant differences in frequency of adverse events.
Prostaglandins — Although promising, more data on the safety and efficacy of prostaglandins in
management of retained placenta are needed before this therapy can be recommended:
 Misoprostol versus oxytocin versus placebo— A three-arm randomized trial compared the need for
manual removal of the placenta in women with retained placenta who were managed by umbilical
vein injection of: (1) 50 units oxytocin (in 30 mL saline); or (2) 800 mcg misoprostol (in 30 mL
saline); or (3) 30 mL saline alone [63]. Use of misoprostol significantly reduced the need for manual
removal of the retained placenta, whereas oxytocin was no more effective than saline alone.
 Sulprostene versus placebo— A double blind trial randomly assigned women with retained
placentas to treatment with sulprostone 250 mcg, a prostaglandin E-2 derivative, or placebo
administered intravenously over 30 minutes [64]. Significantly more patients in the sulprostone
group expelled the placenta without need for manual extraction (13/24 versus 4/26).
Uterine inversion — Uterine inversion is an obstetrical emergency because it is often accompanied by
profuse hemorrhage. An attempt should be made to manually reposition the uterus; administration of
pharmacologic agents to assist in uterine relaxation may facilitate this maneuver. If manual repositioning is
unsuccessful, then surgical intervention is necessary. After the uterus has been restored to its normal
position, uterotonic agents are given to prevent recurrent uterine inversion. Management of uterine
inversion is discussed in more detail separately. (See "Puerperal uterine inversion".)

SUMMARY AND RECOMMENDATIONS

 There is no universally accepted criterion for the normal length of the third stage. The average
length is five to six minutes, 90 percent of placentas are delivered within 15 minutes and 97
percent are delivered within 30 minutes of birth. However, preterm deliveries are associated with a
longer third stage of labor than term deliveries. (See 'Duration' above.)
 Administration of a uterotonic agent is recommended during the third stage to reduce maternal
blood loss. (See 'Expectant versus active management' above.)
 There are insufficient data to determine whether administration of the uterotonic agent is more
effective if it is given after delivery of the anterior shoulder but before separation of the placenta or
immediately upon expulsion of the placenta. We employ a relatively rapid infusion of intravenous
oxytocin (20 units in 1000 mL crystalloid) upon delivery of the anterior shoulder (to limit the risk of
shoulder dystocia). After delivery of the baby and when signs of placental separation are evident,
we employ the Brandt maneuver and put gentle traction on the umbilical cord to facilitate delivery
of the placenta. We perform manual uterine massage and continue the oxytocin infusion after
delivery of the placenta. (See 'Oxytocin' above.)
 We recommend oxytocin or the combined ergometrine-oxytocin preparation (ie, syntometrine, not
available in the United States) for the uterotonic agent. When compared to oxytocin alone,
ergometrine-oxytocin appears to be associated with a small reduction in the risk of postpartum
hemorrhage of 500 to 1000 mL, but an increase in side effects (vomiting, hypertension). (See
'Management' above.)
 Oxytocin can be infused (20 units in 1000 mL crystalloid) into a maternal vein or injected into the
umbilical vein (20 units), both routes of administration are effective. Although bolus intravenous
injection (up to 10 units) was reported to be safe and more effective than slow infusion in one well-
designed trial, several other studies have demonstrated ischemic electrocardiogram changes,
cardiovascular collapse and even cardiac arrest after bolus administration. Therefore, we
 recommend avoiding rapid bolus injection of oxytocin. In the absence of venous access,
intramuscular injection (up to 10 units) is reasonable. (See 'Oxytocin' above.)
 Retained placenta is variously defined as a placenta that has not been expelled by 30 to 60
minutes after delivery of the baby. We recommend manual removal of a placenta that has been
retained for at least 30 minutes to prevent excessive bleeding and infection. The procedure is done
in an operating room, and ultrasound guidance is suggested. Intravenous injection of
nitroglycerin (50 to 200 mcg) may facilitate the procedure by relaxing the uterus. (See 'Retained
placenta' above.)

http://www.uptodate.com