UROGYNAECOLOGY

Intrauterine Bakri Balloon and Vaginal

Tamponade Combined with Abdominal
Compression for the Management of
Postpartum Hemorrhage
Yuna Guo, MD;1 Renyi Hua, MD, PhD;1 Shoufang Bian, MD;1 Xianjing Xie, MD;1 Jue Ma, MD;1
Yanqin Cai, MD;1 Suren R. Sooranna, PhD;2 Weiwei Cheng, MD, PhD1
1
International Peace Maternity and Child Health Hospital of China Welfare Institution (IPMCH), Shanghai, China
2
Imperial College Parturition Research Group, Academic Department of Obstetrics & Gynaecology, Imperial College School of Medicine,
Chelsea and Westminster Hospital, London, UK

Abstract Copyright © 2017 The Society of Obstetricians and Gynaecologists of

Objective: This study sought to investigate the effect of Bakri
balloon use and vaginal tamponade combined with abdominal
compression for the management of postpartum hemorrhage
(PPH).
Methods: This retrospective study reviewed cases of PPH in the
International Peace Maternal and Child Health Hospital of China
Welfare Institution in Shanghai, China from January 1, 2010 to
December 31, 2015. A single use of the intrauterine Bakri balloon
was applied in some cases, and additional vaginal tamponade
combined with abdominal compression (double compression) was
applied in other cases. The authors evaluated the effect of these
two methods in the management of PPH.
Results: The Bakri balloon was used in 305 cases of intrauterine
PPH, and the clinical efficacy was 93.26%. One group of study
patients underwent double compression, and these patients had a
better clinical efficacy rate of 96.3% (157 of 163), whereas the
efficacy in cases using the Bakri balloon alone (control group) was
87.3% (124 of 142). The postoperative complication rates of these
two groups were 9.4% and 8.7%, respectively. Uterine arterial
embolization was performed in patients in whom Bakri balloon use
failed. None of the cases resulted in a hysterectomy.
Conclusion: Intrauterine Bakri balloon use combined with vaginal
tamponade and abdominal compression is more effective in the
treatment of PPH compared with Bakri balloon use alone. This
method does not increase postoperative complications. Uterine
atony with placenta previa or implantation may be possible
reasons for noneffectiveness of Bakri balloon use.
Key Words: Intrauterine, Bakri balloon, vaginal tamponade,
compression, postpartum hemorrhage
Corresponding Author: Dr. Weiwei Cheng, International Peace
Maternity and Child Health Hospital of China Welfare Institution
(IPMCH), Shanghai, China. julietsubmit@163.com
Y.G. and R.H. are joint first authors.
Competing interests: None declared.
Received on June 15, 2017
Accepted on August 21, 2017
Canada/La Société des obstétriciens et gynécologues du Canada.
Published by Elsevier Inc. All rights reserved.
J Obstet Gynaecol Can 2017;■■(■■):■■–■■
https://doi.org/10.1016/j.jogc.2017.08.035
INTRODUCTION
P ostpartum hemorrhage (PPH), defined as the loss of
500–1000 mL of blood after the third stage of labour
(for both vaginal delivery and CS),1 causes approximately
27.1% of maternal deaths worldwide,2,3 and it is the main
cause of maternal deaths in developing countries. Uterine
atony, placental factors, injury to soft tissues associated with
the birth canal, or coagulation abnormalities are the most
common reasons for PPH. Uterine atony causes about 70%
of PPH cases and is the most common cause. Uterine
massage and use of a uterotonic may help to reduce bleed-
ing. If this fails, a surgeon normally needs to perform a
further procedure, such as uterine tamponade, B-Lynch or
uterine artery embolization, uterine artery ligation, and even
hysterectomy. The uterine cavity balloon has the double effect
of compression and tamponade, and it is widely used for
treating PPH. In 1992, Bakri4 first reported on the use of
a balloon for the treatment of PPH secondary to placenta
previa.
However, Bakri balloon tamponade is not 100% success-
ful in PPH. Vitthala et al.5 found the success rate of Bakri
balloon tamponade to be 80%. This finding is consistent
with results reported by Gronvall et al.,6 who found the
success rate to be 86% (43 of 60 cases). Other investiga-
tors reported that the success rate was around 75.5% to
90%.5,7–9 Vaginal tamponade is one method of treating PPH

■■ JOGC ■■ 2017 • 1
UROGYNAECOLOGY
that has been commonly used clinically. In this study, we
combined those two methods to evaluate the success rate
of Bakri balloon tamponade in treating PPH.
MATERIALS AND METHODS
We reviewed 78 708 deliveries, among which 35 120 were
CS procedures, from January 1, 2010 to December 31, 2015
in the International Peace Maternal and Child Health Hos-
pital (IPMCH), Shanghai, China. The Ethics Committee of
the IPMCH approved this study. A single use of an intra-
uterine Bakri balloon was applied in some cases, and
additional vaginal tamponade combined with abdominal com-
pression was applied in other cases. We compared the
outcomes of these two methods in the management of PPH.
Diagnosis of Postpartum Hemorrhage and
Postoperative Morbidity: the Procedure before Use
of Tamponade
The Bakri uterine balloon (also called SOS Bakri) was pur-
chased from Cook Medical (Bloomington, IN). The criterion
for use of uterine tamponade was, during CS and after
massage and use of a uterotonic (e.g., oxytocin and
tromethamine), the presence of persistent active uterine
bleeding or an amount of bleeding that reached a volume
of 500 mL. PPH after CS is defined as an amount of bleed-
ing during CS in excess of 1000 mL. Treatment of PPH
includes intraoperative uterine massage and use of
uterotonics, including oxytocin and carboprost tromethamine,
which are the two most commonly used medications in our
hospital. If balloon tamponade fails to control the bleed-
ing, a uterine artery embolization or hysterectomy is normally
performed.
After the first 24 hours of the operation, the patient’s tem-
perature was measured every 4 hours and four times daily
for up to 3 days. Postoperative fever is diagnosed when the
patient’s temperature reaches 38°C on two occasions on the
same day.
Protocol for Uterine and Vaginal Tamponade
Patients were randomly divided into two groups on the basis
of individual doctors’ experience and practices. The control
group was treated with Bakri balloon tamponade within the
uterine cavity. If PPH was in progress during CS and the
bleeding could not be controlled by uterine massage or
uterotonics, a Bakri balloon was placed in the bottom of the
uterus, distally through the cervix and indwelling in the
vagina. Then the balloon was filled with 250 to 350 mL of
saline. A catheter in the Bakri balloon allows the blood in
the uterus to be released. Therefore, the amount of vaginal
bleeding after placement of the balloon could be noted. If
2 • ■■ JOGC ■■ 2017
Figure. Protocol for uterine and vaginal tamponade.
Postpartum hemorrhage during CS
Massage and the use of uterotonic
Active uterine bleeding still persisted or
if the amount of bleeding approached 500 mL
Bakri balloon tamponade inserted Bakri balloon tamponade inserted
within the uterine cavity within the uterine cavity together
with vaginal gauze
there was still active bleeding, then this raised the possibil-
ity of failure of the balloon, and the other methods (e.g.,
uterine artery embolization or hysterectomy) were imple-
mented. If the bleeding was controlled, then the Bakri
balloon was left in the uterine cavity for 24 hours before it
was taken out. In the study group, we used the method of
“double compression”: the Bakri balloon was filled in the
uterine cavity, with vaginal gauze packing. The balloon and
vaginal tampons were removed within 24 hours. The pro-
tocol used is summarized diagrammatically in the Figure.
The dimensions of the vaginal tampon used were about
200 × 40 mm. The tampon was placed in the vagina for 24
hours together with the balloon. Intravenous antibiotics (e.g.,
cephalosporins or clindamycin) were routinely used to prevent
infection during all these procedures.
Statistical Analysis
All statistical analyses were performed with SPSS V.20.0 for
Windows (IBM, Armonk, NY). A value of P < 0.05 was con-
sidered statistically significant.
RESULTS
The amount of blood loss in the control group was
805.63 ± 459.52 mL, and it was 863.21 ± 467.15 mL in the
double compression group (Table 1). In the double com-
pression group, 96.3% cases of PPH were successfully
treated, a higher number than in the control group (87.3%;
P < 0.05). There were no differences in patients’ ages, GAs,
or the rates of CS between the two groups (Table 1). There
was no difference in the percentage of patients with uterine
atony with or without placenta previa or placenta accreta,
nor was there any difference in the incidence of postop-
erative fever in either of the two groups.
There was no difference in the success rate of controlling
uterine atony caused by PPH in both groups (Table 2). In
PPH caused by placental effects and uterine atony, double
 Managing Postpartum Hemorrhage

Table 1. Clinical characteristics of the two groups of patients studieda

Double compression (n = 163) Control (n = 142)
Age 32.02 ± 3.56 33.47 ± 4.25
Multipara, n (%) 31 (19.01) 31 (21.88)
Emergency CS, n (%) 40 (24.54) 31 (21.83)
GA (mean ± SD) 37.30 ± 1.82 37.09 ± 1.53
Uterine atony, n (%) 80 (49.08) 62 (43.66)
Placenta previa and placenta accreta, n (%) 40 (24.54) 58 (40.85)
Uterine atony and placenta previa and/or placenta accreta 43 (26.38) 31 (21.83)
Blood loss (mL) (mean ± SD) 863.21 ± 467.15 805.63 ± 459.52
Bleeding control, n (%) 157 (96.3)b 124 (87.3)
Postoperative disease, n (%) 14 (8.6) 13 (9.2)
SD: standard deviation.
a
The percentage of bleeding control in the double compression group is greater than that in the control group. Data were compared using the chi-square test, and
n refers to the number of patients in each group.
b
A significant difference of P < .01.

compression (100%) was more effective than use of the Bakri
balloon alone (57.14%).
Furthermore, the amount of blood loss in the double com-
pression group (1040 ± 508.90 mL) was greater than in the
control group (776.43 ± 306.99 mL), but this was not sig-
nificantly different. Six cases in the double compression
group with blood loss of 900–2525 mL underwent uterine
arterial embolization, whereas five subjects in the control
group with blood loss of 700 to 1200 mL had uterine and
vaginal tampons.
In the treatment of PPH caused by placental effects, the
blood loss in the double compression group was less than
in the control group. The cure rate of the control group
in management of placental effects in addition to uterine
atony as the causes of PPH is only 57.1% (4 of 7) (Table 3).
In the control group, eight patients had further invasive man-
agement after use of the Bakri balloon; three of these
patients had uterine atony with placenta previa or placenta
accreta, and the other patient had uterine atony. Among all
the patients who required further management, one patient
in the control group had uterine and vaginal tampons, and
the others underwent uterine arterial embolization. In no
patient was a hysterectomy performed.
DISCUSSION
The Bakri balloon is normally used when uterotonics or pros-
taglandins fail to control PPH.10 One study showed that the
rate of maternal mortality increased if patients were not
stable in the first hour after operation.11 In 2009, a report
by the Royal College of Obstetricians and Gynaecologists
recommended using the Bakri balloon in PPH caused by
uterine atony.12 This device was further recommended for
use in PPH for other reasons in 2011 because complica-
tions arising from its use are extremely rare.13
A study published in 2009 showed that the efficacy of the
Bakri balloon was 80% in 18 cases of PPH.5 In other studies,
the efficacy was 87% and 75% in 16 and 23 cases of PPH,
respectively.10,14 In our study, the efficacy of the Bakri balloon
was 93.26%, a rate similar to that reported in the study by
Gao et al.15 (93%). Use of an intrauterine Bakri balloon in
addition to vaginal tamponade combined with abdominal

Table 2. The success rate of bleeding control in different clinical circumstances of the two groups of patientsa
Double compression (n = 163) Control (n = 142)
Uterine atony, n (%) 74/80 (92.5) 57/62 (91.94)
Placenta previa and placenta accreta, n (%) 40/40 (100) 58/58 (100)
Uterine atony and placenta previa and/or placenta accreta, n (%) 43/43 (100)b 16/31 (51.61)
Total, n (%) 157 (96.32) 131 (92.25)
a
In cases of uterine atony and placenta previa and/or placenta accreta, double compression is more efficient than the single use of a Bakri balloon. The chi-square
test was used to compare the data, and n refers to the number of patients in each group.
b
A significant difference of P < .001.

■■ JOGC ■■ 2017 • 3
UROGYNAECOLOGY

Table 3. Management of postpartum hemorrhage arising from different clinical causesa

Double compression Control
Uterine atony Blood loss (mL) (mean ± SD) 1040 ± 508.90 776.43 ± 306.99
Efficacy, n (%) 74/80 (92.5) 57/62 (91.94)
Further treatment 6 cases of uterine arterial embolization 5 cases of uterine and vaginal tampons
Postoperative disease, n (%) 9/80 (11.25) 5/62 (8.06)
Placenta accreta (without Blood loss (mL) (mean ± SD) — 966.67 ± 823.21
uterine atony) Efficacy, n (%) — 28/28 (100)
Further treatment — —
Postoperative disease, n (%) — 5/28 (17.86)
Placenta previa (without Blood loss (mL) (mean ± SD) 554.29 ± 104.64 687.14 ± 259.79
uterine atony) Efficacy, n (%) 40/40 (100) 30/30 (100)
Further treatment — —
Postoperative disease, n (%) — 5/30 (16.67)
Uterine atony and placenta Blood loss (mL) (mean ± SD) 881.25 ± 481.97 1142.86 ± 784.45
previa/placenta accreta Efficacy, n (%) 43/43 (100) b
16/31 (51.61)
Further treatment — 3 cases of uterine arterial embolization
Postoperative disease, n (%) 5/43 (11.63) 5/31 (16.13)
SD: standard deviation.
a
The efficacy of double compression in various types of PPH was generally higher than in the control group. The chi-square test was used, and n refers to the
number of patients in each group.
b
A significant difference of P < .001.

compression is more effective in PPH than using the Bakri
balloon alone (96.5% vs. 87.5%). Some clinicians sug-
gested that vaginal tampons should be used only in patients
with a dilated cervix. Their opinion was that an increase in
the pressure within the uterus would lead to an increased
risk of uterine rupture.16 There was no side effect such as
infection noted in our study. Uterine arterial embolization
was performed in two cases in which double compression
failed to control PPH.
Placental implantation increases the risk of PPH such that
a hysterectomy would be recommended to patients who do
not wish to maintain their fertility. Uterine arterial embo-
lization before hysterectomy would decrease the amount of
bleeding and obviate the need for a blood transfusion.17 We
successfully used double compression to treat PPH in pla-
cental implantation, except for one patient, who had both
uterine atony and placental implantation; this situation re-
sulted in uterine arterial embolization, and hence a
hysterectomy was prevented.
Placenta previa is a single risk factor for PPH.9 Kumru et al.18
found that the rate of PPH related to use of the Bakri balloon
in placenta previa was 88% (22 of 25). These investigators re-
ported that only two patients required further uterine arterial
embolization or the B-Lynch procedure. Hysterectomy was per-
formed in one of these patients. Side effects of the use of the
Bakri balloon were very rare. In a randomized controlled study
that included 51 cases of placenta previa, the Bakri balloon was
placed in every patient in the study group as a preventive
measure. The results showed that this technique did not help
to reduce the use of other uterotonics.19 In our study, among
21 patients with a single risk of placenta previa, the Bakri balloon
was used in 16, and although bleeding was active, the amount
of blood loss did not reach over 1000 mL. The efficacy was
100% in both the double compression (n = 12) and control
groups (n = 4). The efficacy of Bakri balloon use in the other
28 cases with placental factors and PPH was 100% (double com-
pression; 18 of 18) and 81.25% (control group, 13 of 16). In
conclusion, in cases of PPH caused by placenta previa, the use
of double compression is more effective than use of the Bakri
balloon alone.
In recent years, researchers have gradually combined applica-
tion of the Bakri balloon with other compression methods, such
as the B-lynch procedure. The “sandwich” technique is to apply
the balloon within the uterus, by using the B-lynch procedure
with external pressure applied, to restrict the movement of the
fundus, thereby enhancing the effect of packing.9 A study that
included 20 patients showed that the efficacy of Bakri balloon
use, with or without the B-lynch procedure, to reduce severe
PPH was 90%.20 Canonico et al.21 suggested that the Bakri
balloon, in addition to the B-lynch procedure, was effective in
reducing PPH caused by placenta previa. However, this ap-
proach may cause some side effects, such as endometritis, uterine
rupture, and necrosis.22

4 • ■■ JOGC ■■ 2017

Uterine atony in addition to a placental abnormality generally
causes a large amount of blood loss and therefore requires blood
transfusion.23 Our study also found that double compression
was more effective than method used in the control group, es-
pecially in PPH that resulted from placental factors and uterine
atony. Three patients in the control group required further
uterine arterial embolization. The possible reason could be that
patients with uterine atony had active bleeding at the placen-
tal detachment surface, the Bakri balloon alone failed to stop
the bleeding, and double compression could help to improve
the situation. In our study, abdominal compression bandag-
ing can limit movement of the fundus of uterus and the vaginal
tampons and thus prevent the downward movement of the
balloon and compression of the lower uterine segment. This
approach produced a double pressure on the uterus. This double
compression did not result in uterine rupture, uterine necro-
sis, or other complications, nor did it increase postoperative
morbidity. This finding suggests that the increased vaginal gauze
did not increase the chances of infection. Therefore, the method
appears to be clinically safe to use. In addition, abdominal com-
pression bandaging and vaginal tampons are economical and
practical.
CONCLUSION
Our study suggests that in patients with uterine atony and
placental factors, the double pressure option for the treat-
ment of PPH should be recommended.
ACKNOWLEDGEMENTS
This study was supported by grants from the National
Natural Science Foundation of China, No. 81300513.
REFERENCES
1. Asboth G, Phaneuf S, Lopez Bernal AL. Prostaglandin E receptors in
myometrial cells. Acta Physiol Hung 1997;85:39–50.
2. Say L, Chou D, Gemmill A, et al. Global causes of maternal death: a
WHO systematic analysis. Lancet Glob Health 2014;2:e323–33.
3. Hogberg U. The World Health Report 2005: “make every mother and
child count”—including Africans. Scand J Public Health 2005;33:409–11.
4. Bakri YN. Uterine tamponade-drain for hemorrhage secondary to pla-
centa previa-accreta. Int J Gynaecol Obstet 1992;37:302–3.
5. Vitthala S, Tsoumpou I, Anjum ZK, et al. Use of Bakri balloon in post-
partum haemorrhage: a series of 15 cases. Aust N Z J Obstet Gynaecol
2009;49:191–4.
6. Gronvall M, Tikkanen M, Tallberg E, et al. Use of Bakri balloon tam-
ponade in the treatment of postpartum hemorrhage: a series of 50 cases
from a tertiary teaching hospital. Acta Obstet Gynecol Scand
2013;92:433–8.
Managing Postpartum Hemorrhage
7. Yoong W, Ridout A, Memtsa M, et al. Application of uterine compres-
sion suture in association with intrauterine balloon tamponade (“uterine
sandwich”) for postpartum hemorrhage. Acta Obstet Gynecol Scand
2012;91:147–51.
8. Kaya B, Tuten A, Daglar K, et al. Balloon tamponade for the manage-
ment of postpartum uterine hemorrhage. J Perinat Med 2014;42:745–53.
9. Nelson WL, O’Brien JM. The uterine sandwich for persistent uterine
atony: combining the B-Lynch compression suture and an intrauterine
Bakri balloon. Am J Obstet Gynecol 2007;196:e9–10.
10. Condous GS, Arulkumaran S, Symonds I, et al. The “tamponade test” in
the management of massive postpartum hemorrhage. Obstet Gynecol
2003;101:767–72.
11. Lewis G. Saving Mothers’ Lives: the continuing benefits for maternal
health from the United Kingdom (UK) Confidential Enquires into Ma-
ternal Deaths. Semin Perinatol 2012;36:19–26.
12. Royal College of Obstetricians and Gynaecologists. Prevention and man-
agement of postpartum haemorrhage. Green-top guidelines 52. BJOG
2017;124:e106–49. London: Royal College of Obstetricians and
Gynaecologists.
13. Johnston TA, Paterson-Brown S, Guidelines Committee of the Royal
College of Obstetricians and Gynaecologists. Placenta praevia, placenta
praevia accreta and vasa praevia: diagnosis and management. Green-top
guidelines 27. London: Royal College of Obstetricians and
Gynaecologists; 2011.
14. Dabelea V, Schultze PM, McDuffie RS Jr. Intrauterine balloon tampon-
ade in the management of postpartum hemorrhage. Am J Perinatol
2007;24:359–64.
15. Gao Y, Wang Z, Zhang J, et al. [Efficacy and safety of intrauterine Bakri
balloon tamponade in the treatment of postpartum hemorrhage: a multi-
center analysis of 109 cases]. Zhonghua Fu Chan Ke Za Zhi
2014;49:670–5 [in Chinese].
16. Georgiou C. Balloon tamponade in the management of postpartum
haemorrhage: a review. BJOG 2009;116:748–57.
17. Angstmann T, Gard G, Harrington T, et al. Surgical management of pla-
centa accreta: a cohort series and suggested approach. Am J Obstet
Gynecol 2010;202:38, e1-9.
18. Kumru P, Demirci O, Erdogdu E, et al. The Bakri balloon for the man-
agement of postpartum hemorrhage in cases with placenta previa. Eur J
Obstet Gynecol Reprod Biol 2013;167:167–70.
19. Beckmann MM, Chaplin J. Bakri balloon during cesarean delivery for
placenta previa. Int J Gynaecol Obstet 2014;124:118–22.
20. Diemert A, Ortmeyer G, Hollwitz B, et al. The combination of intra-
uterine balloon tamponade and the B-Lynch procedure for the treatment
of severe postpartum hemorrhage. Am J Obstet Gynecol 2012;206:65,
e1-4.
21. Canonico S, Arduini M, Epicoco G, et al. Placenta previa percreta: a case
report of successful management via conservative surgery. Case Rep
Obstet Gynecol 2013;2013:702067.
22. Saman Kumara YV, Marasinghe JP, Condous G, et al. Pregnancy compli-
cated by a uterine fundal defect resulting from a previous B-Lynch
suture. BJOG 2009;116:1815–7.
23. Ferrazzani S, Iadarola R, Perrelli A, et al. Use of an intrauterine inflated
catheter balloon in massive post-partum hemorrhage: a series of 52
cases. J Obstet Gynaecol Res 2014;40:1603–10.
■■ JOGC ■■ 2017 • 5