doi:10.1111/jog.13285 J. Obstet. Gynaecol. Res.

2017

Vaginal delivery after placental abruption with intrauterine

fetal death: A 20-year single-center experience

Ayami Inoue, Eiji Kondoh, Koh Suginami, Shingo Io, Yoshitsugu Chigusa and Ikuo Konishi
Department of Gynaecology and Obstetrics, Kyoto University, Kyoto, Japan

Abstract
Aim: The aim of this study was to elucidate the feasibility and safety of vaginal delivery (VD) when placental
abruption causes fetal demise.
Methods: We conducted a retrospective study of women who were managed for placental abruption with intra-
uterine fetal death at Kyoto University Hospital during the period from 1995 to 2015.
Results: Sixteen cases were identified during the study period. VD was attempted in 15 cases and was accom-
plished in 14 (93.3%) cases. The median gestational age was 36 (24–39) weeks, and there were eight primiparas.
The median Bishop score on admission was 2.5 (1–9). Eight pregnancies were complicated with pregnancy-
induced hypertension. The median duration of labor was 5 h and 18 min (30 min–12 h 43 min), and the median
amount of hemorrhage was 2503 (445–6808) mL. Fresh frozen plasma (≥ 20 U) and red cell concentrate (≥ 10 U)
were administered in 10 (71%) and 9 (64%) cases, respectively. Two cases required uterine artery embolization
for post-partum hemorrhage, while there was no case of maternal death or hysterectomy. Patients with Bishop
score > 3 (n = 6) experienced shorter-duration deliveries (P = 0.020) and had significantly larger blood loss volume
(P = 0.020) compared to patients with Bishop score ≤ 3. The duration of labor was negatively correlated with the
amount of blood loss (R2 = 0.56, P = 0.039).
Conclusion: After placental abruption with intrauterine fetal death, VD is feasible and safe regardless of gesta-
tional age, parity, cervical maturity, and duration of labor when intensive medical resources are available.
Key words: blood loss, duration of labor, intrauterine fetal death, placental abruption, vaginal delivery.

Introduction underlying cause is the fundamental treatment for DIC,

Placental abruption is defined as a premature separation
of the placenta from the uterine wall and complicates ap-
proximately 1% of all pregnancies.1 Placental abruption
predisposes patients to stillbirth, especially in cases
where more than 50% of the placenta has separated.2
Placental abruption with intrauterine fetal death (IUFD)
frequently causes maternal disseminated intravascular
coagulation (DIC) and can lead to a life-threatening
event. Therefore, intensive management generally in-
cludes administration of fresh frozen plasma (FFP) to
correct coagulopathy. Moreover, as removal of the
delivery of the dead fetus and the placenta is mandatory.
In Western countries, vaginal delivery (VD) is consid-
ered to be preferable and indications for cesarean deliv-
ery are limited to some circumstances, such as
fetopelvic disproportion or a prior classical cesarean de-
livery.3 This is because cesarean delivery has no fetal ad-
vantage and bleeding from surgical incisions is
extremely difficult to control in the presence of DIC.4,5
However, in Japan, cesarean delivery is prevalent just be-
cause prompt delivery of the dead fetus and the placenta
is likely to prevent worsening of maternal DIC.6 Indeed,
the guidelines for obstetrical practice in Japan have

Received: August 8 2016.

Accepted: December 19 2016.
Correspondence: Dr Eiji Kondoh, Department of Gynaecology and Obstetrics, Kyoto University, 54 Shogoin Kawahara-cho, Sakyo-ku, Kyoto
606-8507, Japan. Email: kondo@kuhp.kyoto-u.ac.jp

© 2017 Japan Society of Obstetrics and Gynecology 1

A. Inoue et al.

determined that labor induction or augmentation for VD
or emergency cesarean section (CS) should be chosen af-
ter considering the patient’s condition while simulta-
neously assessing and treating the DIC.7 At our
hospital, VD has been primarily performed for a long
time in cases of placental abruption with IUFD because
CS is thought to increase maternal morbidity without
any fetal advantage. Thus, we aimed to evaluate the fea-
sibility and safety of VD in the management of this par-
ticular condition.
Methods
We conducted a retrospective analysis of all patients
who were diagnosed with placental abruption with
IUFD and managed at Kyoto University Hospital during
the period from January 1995 to December 2015. The
study was approved by the ethics committee of Kyoto
University. The medical records of patients were
reviewed, and clinical and laboratory data were ex-
tracted, including maternal age, obstetric history, Bishop
score on admission (cervical dilation, effacement, consis-
tency, position, and fetal station), the presence of vaginal
hemorrhage and pregnancy-induced hypertension
(PIH), and plasma fibrinogen levels. Management
(mode of delivery, amniotomy, devices or medications
for cervical ripening and labor induction, use of MgSO4,
and blood transfusion) and outcomes (birthweight, du-
ration of labor, and blood loss) of the patients were also
reviewed. As maintenance of arterial oxygen delivery
and the correction/prevention of coagulopathy are es-
sential for the management of patients with placental
abruption with IUFD, blood transfusion has been per-
formed with a target hemoglobin of at least 7 g/dL, with
a fibrinogen level maintaining at least 150 mg/dL, and
with a target platelet count above 50 000 cells/mm3.
The data are expressed as median (range). In order to an-
alyze the influence on duration of labor and amount of
blood loss, the Mann–Whitney U-test, log–rank (Man-
tel–Cox) test, and Spearman’s rank correlation test were
used to determine the significance of differences in vari-
ables, including maternal age, parity, gestational age,
and Bishop score.
Results
Sixteen cases of placental abruption with IUFD were
identified during the study period (Tables 1, 2). VD
was attempted in all but one of the cases, in which emer-
gency CS was performed to save the fetus because fetal
heart beat was observed just before arrival at our hospi-
tal (Case 16). Of the 15 attempted VD cases, fourteen
(93.3%) accomplished VD while one case (Case 15) was
forced to switch to CS due to arrest of labor at 7-cm dila-
tion of the cervix. The following data analysis was there-
fore performed on the 14 women who accomplished VD.
The maternal age was 30.5 (26–37) years. There were six
multiparas, two of whom had previously undergone CS.

Table 1 Characteristics of patients with placental abruption with intrauterine fetal death
Platelet‡
Age Gestational Bishop Vaginal Hemoglobin‡ Fibrinogen‡ (×103cells/
Case (years) Parity age (weeks) score† bleeding† PIH (g/dL) (mg/dL) mm3)

1 34 1 24 1 + 10.5 N/A 185

2 28 0 27 1 + 7.9 50 97
3 26 0 31 2 + 8.3 156 194
4 31 1 (VD) 33 4 + Severe PE 5.9 58 30
5 33 0 34 4 Severe GH 9.8 71 130
6 30 0 35 3 Severe GH 7.2 68 47
7 26 0 35 2 + Severe PE 8.7 77 70
8 32 1 (CD) 37 2 Severe PE 13.2 138 93
9 37 0 37 4 Severe PE 12 139 120
10 30 1 (CD) 38 2 8.4 188 384
11 32 0 38 9 + 10 60 68
12 28 0 38 2 Severe PE 13.6 50 76
13 28 2 (VD) 39 9 + Severe GH 10.7 60 108
14 31 1 (VD) 39 4 8.2 108 164
15 24 0 33 3 + Severe PE 5.2 87 50
16 31 0 37 6 + 9.1 NA 145
†Findings on arrival. ‡Lowest value during hospital stay. CD, cesarean delivery; GH, gestational hypertension; NA, not available; PE, pre-eclamp-
sia; PIH, pregnancy-induced hypertension; VD, vaginal delivery.

2 © 2017 Japan Society of Obstetrics and Gynecology

 Table 2 Management and outcomes of patients with placental abruption with intrauterine fetal death
Devices or Blood transfusion (unit)
Attempted medications for Mode Duration Blood
mode of cervical ripening Use of of Birthweight of labor loss
Case delivery Amniotomy or labor induction MgSO4 delivery (g) (h:min) (mL) FFP RCC PC
1 VD None VD 642 2:41 445 0 0 0
2 VD + TBC, PGF2α VD 1020 12:43 1123 24 20 20

© 2017 Japan Society of Obstetrics and Gynecology

3 VD + TBC, PGF2α VD 1854 9:10 2588 9 14 20
4 VD + TBC, OT VD 1727 1:12 6808 36 25 30
5 VD + OT + VD 1608 4:49 4152† 60 18 30
6 VD + OT + VD 2172 5:40 2418 48 14 20
7 VD + OT + VD 2653 7:00 2653 30 12 15
8 VD + OT + VD 1942 9:42 887 18 4 20
9 VD + OT + VD 2698 4:05 4131† 54 14 45
10 VD + TBC, OT VD 2686 8:57 1157 0 0 0
11 VD OT VD 2895 5:58 2289 26 8 60
12 VD + OT + VD 3480 4:56 3480 57 10 35
13 VD + OT VD 3164 0:30 2814 27 6 0
14 VD + OT VD 3124 3:09 3079 33 10 30
15 VD + TBC, PGF2α, OT + CD 1954 17:00 4614 18 30 35
16 CD None CD 3093 N/A 2350 0 6 0
†Uterine artery embolization was required to control hemorrhage. CD, cesarean delivery; FFP, fresh frozen plasma; MgSO4, magnesium sulfate; NA, not applicable; OT, oxytocin; PC, plate-
let; PGF2α, prostaglandin F2 alpha; RCC, red cell concentrate; TBC, transcervical balloon catheter; VD, vaginal delivery.
Safe vaginal delivery after abruption

3
A. Inoue et al.
The gestational period was 36 (24–39) weeks, and the
Bishop score was 2.5 (1–9) on arrival at the hospital.
On admission, seven cases presented with external
bleeding, while seven had a concealed abruption. Eight
women were complicated with severe PIH, including
three cases of gestational hypertension and five cases of
pre-eclampsia.
When the patients were diagnosed with placental
abruption with IUFD, immediate amniotomy was per-
formed in all cases except for two, who presented with
membranes that had ruptured prior to admission (Cases
1 and 11). Medications (oxytocin or prostaglandin F2 al-
pha) or transcervical balloon catheter were used in 13
cases, and the birthweight was 2413 (642–3480) g. Mag-
nesium sulfate (MgSO4) was used during delivery to
prevent eclampsia in six cases with severe PIH. The du-
ration of labor after admission was 5 h 18 min (30 min–
12 h 43 min). Among relevant variables, the presence of
VD history was associated with shorter time to delivery
of a baby (Table 3, P = 0.0020). In addition, Bishop score
> 3 was linked to shorter duration of labor compared to
Bishop score ≤ 3 (Table 3, Mann–Whitney U-test, P =
0.020; Figure 1a, log–rank test, P = 0.0066).
The amount of blood lost was 2503 (445–6808) mL.
The lowest levels of hemoglobin, fibrinogen, and plate-
lets were 9.3 (5.9–13.6) g/dL, 71 (50–188) mg/dL, and
10.3 × 104 (3.0–38.4) cells/mm3, respectively. Of the 14
cases, 12 (85.7%) required blood transfusion, and blood
products transfused were 11 (0–25) U of red cell concen-
trate (RCC), 28.5 (0–60) U of FFP, and 20 (0–60) U of
platelets. Massive FFP (≥ 20 U) and RCC (≥ 10 U) were
administered in 10 (71%) and 9 (64%) cases, respectively.
Bishop score > 3 was associated with significantly in-
creased blood loss (Table 3, P = 0.020). Patients compli-
cated with PIH presented with greater blood loss, but
the Mann–Whitney analysis did not quite reach statisti-
cal significance (P = 0.059). The subtype of PIH and ad-
ministration of MgSO4 did not significantly affect the
time to delivery or the amount of hemorrhage. The
amount of blood loss was significantly negatively corre-
lated with the duration of labor (Fig. 1b; R2 = 0.56, P =
0.039). The fibrinogen level at admission was not corre-
lated with either the duration of labor or the amount of
blood loss (Spearman’s rank correlation, P = 0.54, P =
0.75). No major changes were observed in indication or
technique of administration of coagulation factors or
blood transfusion during the 20-year study period. The
therapeutic approach to 15 cases with placental abrup-
tion with IUFD remains basically unchanged over 20
years. We have used uterine artery embolization since
2009, and intrauterine balloon tamponade since 2010.
4 © 2017 Japan Society of Obstetrics and Gynecology
Intrauterine balloon tamponade was performed just for
Cases 5 and 9, both of whom further required uterine ar-
tery embolization (UAE) for hemostasis. No candidate
patient was observed for either intrauterine balloon
tamponade or uterine artery embolization before 2009,
and we have managed nine cases since 2009. No hyster-
ectomy was required in any of the cases. There was no
maternal death, and seven cases had subsequent suc-
cessful pregnancy and delivery.
Discussion
In the present study, VD was attempted with success
rates as high as 93% for any case of placental abruption
with IUFD. Generally, a trial of VD is first choice if the fe-
tus is dead or has no chance of survival outside the
uterus. Nevertheless, in Japan, CS remains common for
placental abruption with IUFD, though Kawana et al. re-
ported the steadily decreasing rate of cesarean delivery
from 87.5% in 2002 to 66.7% in 2008 using the national
obstetrics registry of the Japan Society of Obstetrics and
Gynecology.6 In their report, they mentioned the possi-
bility that the mode of delivery is partly determined by
parity and gestational weeks. Indeed, the proportion of
primiparous women was 46% in the CS group (n =
389) and 33.1% in the VD group (n = 117) in 506 cases
of placental abruption with IUFD (database of the
Japan Society of Obstetrics and Gynecology, 2002–
2008). On the other hand, the rate of CS delivery was
91.3% at 28–31 weeks of gestation, while that rate
dropped to 71.5% in pregnancies of 36 weeks or more.6
In the current study, VD was accomplished within 6 h
by almost two-thirds of women regardless of gestational
age, parity, cervical maturity, duration of labor, and con-
tinuous infusion of MgSO4 to prevent eclampsia. This
rapid progress in labor is probably due to early
amniotomy and persistently hypertonic uterus. A trial
of VD is therefore preferred with the exception of some
cases, including a transverse lie or a prior classical CS
delivery.
Serious coagulopathy usually develops in women
with placental abruption with IUFD, and the treatment
of consumptive coagulopathy is key to successful man-
agement, irrespective of the mode of delivery. Indeed,
11 cases (78.6%) had plasma fibrinogen level below 150
mg/dL, and FFP was aggressively administered just af-
ter their arrival. Early amniotomy was routinely per-
formed in order to reduce thromboplastin infusion into
the maternal circulation, which might alleviate accelera-
tion of consumptive coagulopathy.3 Imprudent surgical
 Safe vaginal delivery after abruption

Table 3 Duration of labor and blood loss

Duration of labor (h:min) Blood loss (mL)
Factors (No.) Median Statistics Median Statistics
Maternal age at delivery (years) ≤30 (7) 7:00 0.207 2588 0.603
>30 (7) 4:06 3079
History of vaginal delivery + (4) 1:56 0.002* 2947 0.635
– (10) 6:30 2503
GA at delivery (weeks) <37 (7) 5:40 0.779 2503 0.966
≥37 (7) 4:56 2947
Birthweight (g) <2500 (7) 5:40 0.522 2418 0.450
≥2500 (7) 4:56 2814
Bishop score† ≤3 (8) 7:58 0.020* 1788 0.020*
>3 (6) 3:37 3605
Vaginal bleeding + (7) 4:56 0.871 3079 0.520
– (7) 6:00 2588
Plasma fibrinogen‡ (mg/dL) <100 (8) 5:18 0.741 2734 0.269
≥100 (5) 6:30 1873
PIH + (8) 4:53 0.343 3147 0.059
– (6) 7:29 1723
Type of PIH GH (3) 4:49 0.500 2814 0.857
PE (5) 4:56 3480
Administration of MgSO4 + (6) 5:18 0.650 3067 0.410
– (8) 4:35 2439
†Score at admission. ‡Lowest level. *Statistically significant. GA, gestational age; GH, gestational hypertension; MgSO4, magnesium sulfate; PE,
preeclampsia; PIH, pregnancy-induced hypertension.

intervention with insufficient correction of coagulopathy
could result in hemorrhage, hysterectomy, and maternal
death. Kawana et al. performed a subset analysis of 85
cases with detailed maternal information among 506
cases of placental abruption with IUFD, and found two
cases with maternal deaths, two cases that required hys-
terectomy in the CS group (n = 70), and one case of hys-
terectomy in the VD group (n = 15).6 They state that these
unfortunate events were probably due to lack of vigor-
ous restoration of coagulation factors using FFP. Thus,
consumptive coagulopathy should be corrected before
delivery using massive FFP transfusion.
In the management of placental abruption with IUFD,
maternal safety becomes the primary concern. Blood loss
Figure 1 (a) Time to delivery and Bishop score on admis-
sion. The group consisting of patients with Bishop score
> 3 (n = 6) took significantly less time to deliver as com-
pared to patients with Bishop score ≤ 3 (n = 8). (b) Dura-
tion of labor and blood loss. Duration of labor was
negatively correlated with the amount of blood loss.
in these women usually reaches a considerable amount,
which tends to be concealed in the retroplacental space.
In order to prevent hypovolemic shock, RCC is to be ad-
ministered to maintain adequate hemoglobin levels (≥ 7
g/dL) even if the amount of external bleeding is absent.
To date, the clinical management of abruption with
IUFD has had to rely on knowledge other than that ob-
tained through randomized clinical trials,8 and limited
data are available concerning outcomes associated with
the mode of delivery and time limits to permit VD.6,8–11
There were reportedly no significant differences in ma-
ternal outcomes regarding the amount of blood loss,
the rates of blood transfusion, or the rate of coagulopa-
thy between the VD and CS groups.6 As for time limits
for delivery, some had set arbitrary time limits, such as
6–8 h after the onset of placental abruption,12 while
others disagree with the existence of such time limits.3
Intriguingly, we found that a longer duration of labor
was associated with less blood loss. This unexpected re-
sult is consistent with the report by Kikuchi et al. of nine
cases of placental abruption with IUFD.10 They reported
that the blood loss was less in the slowly progressed
group (<6 h, n = 5) compared with that in the rapidly
progressed group (≥6 h, n = 4) because there was suffi-
cient time for adequate blood transfusion to stabilize
the maternal condition in slowly progressed cases. An-
other possible explanation would be that, in cases with
rapidly progressed labor, uterine contraction and

© 2017 Japan Society of Obstetrics and Gynecology 5

A. Inoue et al.
intrauterine pressure were strong enough to cause dis-
semination of thromboplastins into the maternal circula-
tion, thereby accelerating DIC,3 but this can never be
proven to be true. In our cases, the fibrinogen level at ad-
mission was not associated with either the duration of la-
bor or the amount of blood loss. This may be because we
started FFP transfusion in most cases as soon as the pa-
tients were diagnosed with placental abruption with
IUFD. Furthermore, since hemostasis at the placental im-
plantation site depends largely on myometrial contrac-
tion rather than blood coagulability, post-partum
intractable uterine bleeding may be due to rapid VD,
which could have a risk of deep occult laceration of the
lower uterine segment. Indeed, in the two cases that
required UAE (Cases 5 and 9), both exhibited contrast
medium extravasation from the lower uterine segment
in the dynamic computed tomography (CT) images
(Supplementary Figure 1). This type of arterial bleeding
is almost always difficult to control by augmentation of
uterine contraction alone, and does require intrauterine
balloon tamponade or UAE for hemostasis.13 Although
our findings and those presented by Kikuchi are based
on single-center studies with small sample sizes, we
should be alert to massive post-partum hemorrhage, es-
pecially in case of rapid progress in labor, and it may
safely be said that there is no need for particular time
limits for completion of VD, even in this critical setting.
In conclusion, VD is recommended for women with
placental abruption severe enough to cause IUFD. Al-
though massive blood transfusion is often mandatory
as serious coagulopathy usually develops, VD can be
completed without maternal death or hysterectomy
regardless of gestational age, parity, cervical maturity,
and the duration of labor.
Disclosure
The authors report no conflicts of interest.
References
1. Ananth CV, Oyelese Y, Yeo L, Pradhan A, Vintzileos AM. Pla-
cental abruption in the United States, 1979 through 2001: Tem-
poral trends and potential determinants. Am J Obstet Gynecol
2005; 192: 191–198.
6 © 2017 Japan Society of Obstetrics and Gynecology
2. Ananth CV, Berkowitz GS, Savitz DA, Lapinski RH. Placental
abruption and adverse perinatal outcomes. JAMA 1999; 282:
1646–1651.
3. Cunningham FG, Leveno KJ, Bloom AL et al. Placental abrup-
tion. In: Williams Obstetrics, 24th edn. New York: McGraw-Hill,
2014; 793–799.
4. Oyelese Y, Ananth CV. Placental abruption. Obstet Gynecol
2006; 108: 1005–1016.
5. Hall DR. Abruptio placentae and disseminated intravascular
coagulopathy. Semin Perinatol 2009; 33: 189–195.
6. Kawana Y, Adachi T, Higaki H et al. Maternal indices of placen-
tal abruption with intrauterine fetal death: Comparison be-
tween cesarean section and vaginal delivery. J Jpn Soc Perinat
Neonat Med 2012; 48: 22–26.
7. Minakami H, Maeda T, Fujii T et al. Guidelines for obstetrical
practice in Japan: Japan Society of Obstetrics and Gynecology
(JSOG) and Japan Association of Obstetricians and Gynecolo-
gists (JAOG) 2014 edition. J Obstet Gynaecol Res 2014; 40:
1482–1483.
8. Neilson JP. Interventions for treating placental abruption.
Cochrane Database Syst Rev 2003; (1): CD003247. DOI:10.1002/
14651858.CD003247.
9. Tanigawa T, Miura K, Yoshida A, Nakayama D, Masuzaki H.
Vaginal delivery and cesarean section: Comparison of maternal
outcome in placental abruption with fetal death. Jpn J Obstet
Gynecol Neonat Hematol 2008; 18: S-51–S-52.
10. Kikuchi N, Kobara H, Osada R et al. Management of placental
abruption with intrauterine fetal death: Benefit of vaginal deliv-
ery. J Jpn Soc Perinat Neonat Med 2010; 46: 813–817.
11. Imanaka S, Naruse K, Akasaka J, Shigemitsu A, Iwai K,
Kobayashi H. Vaginal delivery after placental abruption and in-
trauterine fetal death, including failed cases. Int J Gynaecol
Obstet 2014; 126: 180–181.
12. Sher G. Trasylol in the management of abruptio placentae with
consumption coagulopathy and uterine inertia. J Reprod Med
1980; 25: 113–118.
13. Kawamura Y, Kondoh E, Hamanishi J et al. Treatment decision-
making for post-partum hemorrhage using dynamic contrast-
enhanced computed tomography. J Obstet Gynaecol Res 2014;
40: 67–74.
Supporting information
Additional supporting information may be found in the
online version of this article at the publisher’s web-site.
Figure S1. Dynamic contrast-enhanced CT images of
post-partum hemorrhage. CT images showing an extrav-
asation during the arterial phase at the lower uterine seg-
ment (arrow). Arterial bleeding was not controlled by
intrauterine balloon tamponade using multiple Foley
catheters. This case required uterine artery embolization.