Journal of Obstetrics and Gynaecology

ISSN: 0144-3615 (Print) 1364-6893 (Online) Journal homepage: http://www.tandfonline.com/loi/ijog20

Abdomino-pelvic packing to control severe

haemorrhage following caesarean hysterectomy

S. GHOURAB

To cite this article: S. GHOURAB (1999) Abdomino-pelvic packing to control severe

haemorrhage following caesarean hysterectomy, Journal of Obstetrics and Gynaecology, 19:2,
155-158

To link to this article: http://dx.doi.org/10.1080/01443619965480

Published online: 02 Jul 2009.

Submit your article to this journal

Article views: 27

View related articles

Citing articles: 1 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=ijog20

Download by: [University of Sheffield] Date: 05 November 2015, At: 20:16

 Journal of Obstetrics and Gynaecology (1999) Vol. 19, No. 2, 155± 158
OBSTETRICS
Abdomino-pelvic packing to control severe
haemorrhage following caesarean hysterectomy
S. GHOURAB, L. AL-NUAIM, A. AL-JABARI, A. AL-MESHARI, M. S. MUSTAFA,
Z. ABOTALIB and M. AL-SALMAN
Departments of Obstetrics and Gynaecology, and Surgery, King Khalid University Hospital,
Riyadh, Saudi Arabia
Summary
Surgically uncontrollable peri-operative obstetric haemor-
rhage associated with coagulopathy, developed in ® ve women
Downloaded by [University of Sheffield] at 20:16 05 November 2015
who were managed by emergency caesarean hysterectomy.
All women had a morbidly adherent anterior placenta praevia
and a previous low er segm ent caesarean section scar.
C onventional m edical and surgical therapy to con trol
bleeding from pelvic and abdominal raw surfaces were unsuc-
cessful. Abdomino-pelvic packing was performed with 10± 12
dry laparotomy pads applied ® rmly over bleeding sites. The
abdomen was closed after observation of the cessation of
bleeding for 5± 10 minutes. Following correction of coagula-
tion and haemodynamic disorders relaparotomy for pack
rem oval was perform ed 34± 48 hours later. O ne patient
developed small bowel obstruction on the 5th post-operative
day, how ever, there w as no long term gynaecological
morbidity in any of the cases. Abdomino-pelvic packing
achieved complete haemostasis in all of the ® ve women which
we believe m ay have been impossible using atternative
measures.
Introduction
M assive obstetric haemorrhage is an extreme em er-
gency condition. Official guidelines exist regarding the
conservative m anagement of the condition (Cham be-
rain and Patel, 1992 ; Report on Con® dential Enquiry
into M aternal Death in the United K ingdom , 1994).
One of the major causes of obstetric haemorrhage is a
morbidly adherent placenta particularly if associated
w ith previous caesarean sections. The hazards of
massive obstetric haemorrhage in association with the
combination of placenta praevia, morbidly adherent
placenta and caesarean delivery is con® rmed in various
studies (Read et al., 1980 ; Taylor et al., 1994 ; Clark et
al., 1985). The incidence of m aternal mortality and
morbidity rises steeply with the degree of placental
invasion, the am ount of blood loss, the associated
maternal complication and delay in the diagnosis and
management. A lthough the availability of effective
oxytocics together with m odern blood banking facili-
ties m inimized the danger of obstetric haemorrhage,
the m anagement of intractable bleeding in the pres-
ence of coagulopathy remains a formidable task and
the risk of maternal mortality remains great (Zelop et
al., 1993).
Five patients with surgically uncontrollable periop-
erative obstetric haem orrhage treated with abdomino-
Correspondence to: Dr S. Ghourab, Obstetrics and Gynaecology Department (36), King Khalid University Hospital, P.O. Box
7805, Riyadh 11472, Kingdom of Saudi Arabia.
0144± 3615/99/020155± 04 $9.50 € Institute of Obstetrics and Gynaecology Trust, 1999
pelvic packing are presented, literature concerning
managem ent of abnormally adherent placenta praevia
and packing practice is brie¯ y reviewed.
Material and methods
Detailed clinical data of ® ve patients who were treated
by abdomino-pelvic packing were collected pros-
pectively at King Khalid University Hospital, Riyadh,
Saudi Arabia, over the years 1993 to 1995 inclusive.
Fo llow -up c lin ic al in fo r m ation w as ob taine d by
rev iew in g the s ubs eq uen t en tries in th e m ed ic al
record.
Antepartum diagnosis of placenta accreta or percreta
was based on ultrasonographic ® ndings of absence of
nor mal translucent decidual interface between the
placenta and uterine wall (Cox et al., 1988). Localiza-
tion of the lower edge of the placenta and its adhesive-
ness to the uterine wall was checked at the time of
surgery (caesarean section).
All wom en were managed by a multidisciplinary
team , including m ore than one senior obstetrician,
anaesthetists, haematologist, vascular surgeon, nurses,
blood bank and intensive care back-up. All of the ® ve
women had a hysterectomy; in two, the hysterectomy
was pe rfor m e d after failure of standard surgical
techniques, such as suturing placental site, uterine
packing and internal iliac artery ligation.
Decision to pack the pelvis was taken when non-
surgical haem orrhage associated with clinical and
laboratory evidence of coagulopathy had developed
and blood loss could not be replaced adequately in the
face of continuous haemorrhage. Abdominal-pelvic
packing was carried out at the end of the initial
laparotomy in patients Nos 1 and 3 (Table I) while the
other three patients required packing at the second
laparotomy which was perform ed 6± 28 hours after the
initial surgery because of severe and persistent post-
operative intra-abdominal haem orrhage. Coagulation
pro® le wa s m onitored carefully, arterial blood pH
values were also measured. The lowest recorded reading
of pH , tem perature, prothrom bin time (PT), activated
p ar ti a l th ro m b o p la s tin tim e (A P T T ), p l at e l e ts,
D-D imer test and ® brinogen level were taken before
packing either in the operating theatre or in the
surgical intensive care unit (SIC U).
 156 S. Ghourab et al.

Table I. Summary of pre-operative clinical features of the ® ve patients with surgically obstetrics uncontrollable
haemorrhage

Patient No. Treatment year Age (years) Parity Pre-operative diagnosis Presenting symptoms

1 1993 36 7+3 Previous 1 LSCS 32 weeks’ Massive APH

gestation Anterior Pl. Pr.
Placenta Accreta
2 1994 38 7+2 Previous 2 LSCS 35 weeks Emergency admission
gestation Anterior Pl. Pr. Massive APH Shock
totalis Hb=4´49 g/dl
3 1995 31 3+1 Previous 3 LSCS 30 weeks’ Severe APH
gestation Anterior Pl. Pr.
totalis suspected accreta
4 1995 40 8+1 Previous LSCS 37 weeks Elective LSCS
gestation Pl. P. totalis
Suspected accreta
5 1996 32 3+0 Previous 3 LSCS 34 weeks’ Moderate APH Fetal distress
gestation Pl. Pr. totalis

LSCS = lower segment caesarean section; Pl. Pr. = placenta praevia; APH = antepartum haemorrhage; Hb =
haemoglobin.
Downloaded by [University of Sheffield] at 20:16 05 November 2015

Pelvi-abdominal packing was performed by applying
dry laparotomy pads ® rm ly over the bleeding raw
surfaces in the pelvis and abdomen. Around 10± 12
separate laparatomy pads were required for eac h
patient. The abdom en was closed after observation of
cessation of bleeding for 5± 10 minutes. All patients
had positive D-Dim er assay (evidence of ® brinolysis)
and were given 2± 2´5 g of anti® brinolytic agents (tran-
examic acid) intravenously. Intraoperative ¯ uid replace-
ment such as crystralloids, colloids, packed red blood
cells and blood component were infused through rapid
transfusion devices with a warmer. An average of 3
units of fresh frozen plasm a and 6± 10 units of platelets
were given for each 10 units of packed red blood cells
transfused. When serum ® brinogen concentration was
low, 10± 12 units of cryoprecipitate were adm inistered.
Intravenous antibiotics in the form of 2 g of cefoxition
and 0´5 g of metronidazole were given three times daily
for 5± 10 days. All patients were transferred to the
surgical intensive care unit (SICU) for post-operative
care. Each patient had an indwelling Foley’s catheter in
situ and closed suction drains are placed into perito-
neal cavity for continuous drainage and estimation of
blood loss. A central venous line was sited to monitor
¯ uid replacement and balance. The haematological
indices, ¯ uid balance, cardiopulm onary and renal func-
tions were constantly m onitored according to the
protocol of the intensive care unit. Laparotomy for
pack removal was performed 34± 48 hours after the
initial surgery when coagulation and haemodymanic
disorders were corrected.
Results
Five multiparous patients with symptom atic anterior
placenta praevia and previous lower segment caesarean
section scars were identi® ed to be at high risk of peri-
operative haem orrhage. The clinical feature of the ® ve
patients are sum marized in Table I. These women
consented to possible hysterectomy. Three exhibited
sonographic placental pattern suggestive of placental
accreta. All ® ve patients were found to have total
place nta praevia at lap arotomy, and one wom an
(patient No. 5) had extensive intra-abdominal adhe-
sions which involved small bowel and urinary bladder.
The placenta in patient N o. 3 was very adherent to the
anterior lower uterine segm ent, and an attem pt to
separate it resulted in a 3 × 2 µm rent in the posterior
wall of the bladder. This patient was m anaged with
early total caesarean hysterectomy with bladder repair,
she required abdomino-pelvic packing at the conclu-
sion of the total hysterectomy. Total caesarean hyster-
ectomy was also perform ed in patients N os 2 and 4,
when severe haemorrhage developed during removal
of very adherent placenta. Bilateral internal iliac artery
ligation was also performed in patient No. 2 to achieve
haem ostasis in conjunction with adequate replace-
ment of blood products. As both patients continued to
have post-operative intra-peritoneal bleeding, rela-
parotomy was undertaken after 28 and 6 hours in
patients Nos 2 and 4 respectively. At laparotomy diffuse
bleeding from raw surfaces and suture lines was found
and complete haemostasis was achieved with abdomino-
pelvic packing.
Following the com plete but difficult rem oval of
adherent placenta, conservative surgical technique was
initially attem pted in patients Nos 1 and 5 in order to
preserve the uterus. M ultiple suture ligatures and
packing of the lower uterine segment failed to control
uter in e ble e din g. H ys te re c to m y w as un de r ta ke n
immediately in patient No. 1 and after 8 hours in patient
No. 5. Both cases continued to have uncontrollable
post-hysterectomy bleeding. T herefore, ab dom ino-
pelvic packing was resorted to. Complete haemostasis
was achieved in all cases after packing.
Table II shows the lowest recorded readings of pH,
tem perature, platelet prothrom bin (PLT), (PT),
Table II. Pre-packing lowest recorded pH, temperature
(Temp), platelet (PLT), prothrombin (PT) and activated
partial thromboplastin time (APTT)
Patient No. pH Temp PLT PT APTT
1 7´8 35´6 56 59 >2 min
2 7´77 34.5 28 23 >2 min
3 7´19 35´8 41 20 >2 min
4 7´18 38´8 63 19´3 72 s
5 7´2 36´1 89 72 53 s
 Control of haemorrhage followin g caesarean hysterectomy with abdomino-pelvi c packing
activated partial thromboplastin (APTT), which were
mainly recorded in the operative room before packing,
D-Dim er assays were positive in all of them. These
res ult re¯ e ct the developm en t of intra -ope rative
acidosis, hypothermia and coagulation disorders. The
above abnormalities were corrected within 24 hours
after packing and replacement of blood and blood
components. Table III summarises fetal and maternal
outcome, com plications and total blood and blood
products transfused. In all patients, less than 20% of
blood and blood com ponents were replaced after
packing.
The packs were rem oved within 34± 48 hours. They
were stained with serosanguinous ¯ uid but none were
adherent to the raw surfaces and no bleeding resulted
at pack rem oval. All of the ® ve patients have rem ained
free of any long term gynaecological complications.
H istopathological diagnosis, con® rmed the clinical
and ultrasound ® ndings of placenta percreta in patient
No. 3 and placenta accreta in the other four cases.
Downloaded by [University of Sheffield] at 20:16 05 November 2015
Discussion
Obstetric haemorrhage is the leading cause of maternal
death in Saudi Arabia and ranks third in the U nited
Kingdom (M aternal M ortality in Saudi Arabia, 1989 ±
1992 ; Cham berain and Patel, 1992 ; Report on Con® -
dential Enquiry into M aternal D eath in the U nited
K ingd om , 19 94 ). M assive peri-operative obstetric
haem orrhage due to a m orbidly ad herent-anterior
placenta praevia in association with previous lower
segment caesarean section scar is a life-threatening
condition (Stanco et al., 1993). Fox (1972) reviewed
622 reported cases of placenta accreta between 1945 ±
1969 and found that the maternal mortality in his series
was 9´8% . Breen et al., (1977) reported only one death
Table III. Peri-operative characteristics of the ® ve patients treated with pelvi-abdominal packing
Blood components transfused (units)
Patient SICU
No. admission PBC FFP CP
1 38 hours 12 6 10
2 3 days 24 7 12
3 2 days 14 4 6
4 24 hours 18 4 12
5 2 admissions 12 4 ±
24 hours each
after ® rst and
third lap
SICU = surgical intensive care unit; PBC = packed blood cell; FFP = fresh frozen plasma; CP = cryoprecipitates; OP
= operative; RPR = relaparotomy for pack removal; HS = hospital stay.
157
(2´5%) in 40 cases. M orson (1978) however, reported two
mortalities (3%) in 67 patients whereas Read et al. (1980)
had one mortality (4´5%) in 22 cases of placenta accreta.
In most reported series of placenta accreta, there is a
high incidence of implantation of the placenta in the
lower uterine segment. Delivery by caesarean section has
increased steadily over the past two decades (National
Institute of Health Consensus Development Task Force,
1981) therefore awareness of the long term obstetric
com plications associated w ith caesarean births is
becoming increasingly important. In our community
there is no upper limit to the number of caesarean
sections that may be performed and it is suggested that
the incidence of morbidly adherent placenta increases
exponentially with the number of caesarean deliveries
(unpublished data). Clark et al. (1985) reported that the
combination of placenta praevia and one previous
caesarean section has 24% risk of placenta accreta. This
risk increases to 67% in the presence of placenta praevia
in combination with four or more previous caesarean
sections. All ® ve cases we reported had a previous lower
segment caesarean section scar, the number ranging
between one and three. The above association is usually
complicated by massive uncontrollable haemorrhage,
which can be exacerbated by intravascular coagulopathy.
When efforts to detach the placenta accreta are unsuc-
cessful and bleeding continues despite uterine massage,
administration of pressor drugs, surgical repair of
traumatized tissue and ligation of uterine and internal
iliac arteries, the safest treatment is to perform prompt
hysterectomy (Cunningham et al., 1989).
The control of uterine haemorrhage by the use of
intracavity packing has been described more than
c en tu ry ago (R am s b oth am , 1 85 6), bu t fe ar s o f
concealing further bleeding, uterine traum a and infec-
tion together with increasingly effective drugs to treat
Post-op
complication and
PLT Total units Infant status procedures
6 34 1´8 kg/Apgar 2´7 at Fever 38± 39Ê C RPR
1´5 min 34 hours
HS: 10 days
14 71 Fresh stillbirth Fever 38± 40Ê C RPR
2´95 kg after 48 hours;
Small pelvic
haematoma
HS: 15 days
14 38 1´65 kg/Apgar 8,10 Low grade fever,
at 1,5 min RPR 36 hours
HS: 13 days
6 30 2.3 kg/Apgar 6,8 at Fever 38± 39Ê C for 2
1,5 min days; RPR 48 hours
HS: 10 days
6 22 1´84 kg/Apgar 6,10 Fever 37´5± 39Ê C for
at 1,5 min 7 days; RPR 36
hours; Third
laparoreus at 5th
post-op day for
small bowel
obstruction
HS: 14 days
 158 S. Ghourab et al.
uterine atony have reduced this practice. H owever, the
risks have probably been overestim ated. Drucker and
Wallach (1979) recommended its reassessment. Others
have attempted to control uterine bleeding by using
uterine tamponade (Lester et al., 1965 , Goldrath, 1993).
Packing of the lower uterine segment is recom mended
when control of bleeding is unsuccessful and hyster-
ectomy is contraindicated in order to preserve reproduc-
tive function (M aier, 1993).
The contributing factors for development of persistent
coagulopathy, and metabolic acidosis in ® ve cases we
reported, prior to packing, include one or more of the
following factors: massive obstetric haemorrhage, delay
in de® nitive treatment (hysterectomy) and inadequate
replacement of blood and blood products in face of
continuous bleeding. Abdom ino-pelvic packing has
proved to be an effective method in all ® ve women and
had achieved haemostasis that might not otherwise
possible. It must be stressed that packing should not be
used before securing and ligating major vessels. Ideally
Downloaded by [University of Sheffield] at 20:16 05 November 2015
pack removal should not be attempted before correction
of haemodynam ic and coagulation disorders which
frequently takes 48± 72 hours (Sharp and Locicero, 1992).
Relaparotomy for pack removal is a major disadvantage
of abdomino-pelvic packing technique. As the packs were
not adherent and bleeding was not renewed when the
packs were removed, relaparotomy might have been
avoided by using continuous wide gauze pack an d
bringing the end out through the abdominal wall inci-
sion or through the vaginal vault.
A bdom inal packing is not a new method to control
bleeding. It was used in World War II, but it was
discouraged becau se of com plications (Sharp an d
Locicero, 1992). Abdominal packing was repopular-
ised and im proved by Feliciano et al. (1981), Svoboda
et al. (1982), and Carmona et al. (1984). It was used as
an adjunct to surgical techniques in obtaining haemos-
tasis in coagulopathy and has regained popularity over
the past decade (Sharp and Locicero, 1992).
Pelvi-abdominal packing proved to be safe and effec-
tive in controlling an otherwise uncontrollable obstetric
haemorrhage. Its role has to be seriously considered
whenever there is refractory bleeding from raw surfaces
or suture lines after complicated obstetrical or gynae-
cological operations as in the case of pelvic infection,
endometriosis or m alignancy.
References
Breen J. L., Neubecker R., Gregori C. A. and Franklin J. E.
Jr (1977) Placenta acrreta, increta and percreta. A survey
of 40 cases. Obstetrics and Gynecology, 49, 43± 47.
Carmona R. H., Peck D. Z. and Lim R. C. (1984) The role of
packing and planned re-operation in severe hepatic trauma.
Journal of Trauma, 24, 779± 784.
Chamberlain G. V. P. and Patel N. (1992) Maternal Mortality.
The Way Forward, pp. 54± 62. London, RCOG Press.
Clark S. L., Koonings P. P. and Phelan J. P. (1985) Placenta
praevia/accreta and prior cesarean section. Obstetrics and
Gynecology, 66, 89± 92.
Cox S. M., Carpanter R. J. and Cotton D. B. (1988) Placenta
percreta: ultrasound diagnosis and conservative surgical
management. Obstetrics and Gynecology, 71, 454± 456.
Cunningham F. G., MacDonald P. C. and Gant N. E. (1989)
In: W illiam s O bstetrics, pp. 695± 725. N orwaik, C T
Appleton Lan ge.
Drucker M. and Wallach R. C. (1979) Uterine packing: a
reappraisal. Mount Sinai Journal of Medicine, 46, 191± 194.
Feliciano D. V., Mattox K. L. and Jordan G. L. (1981) Intra
abdominal packing for control of hepatic haemorrhage: a
reappraisal. Journal of Trauma, 21, 285± 290.
Fox H. (1972) Placenta accreta, 1945± 1969. Obstetrics and
Gynecology Survey, 475± 490.
Goldrath M. H. (1983) Uterine tamponade for the control of
acute uterine bleeding. American Journal of Obstetrics
Gynecology, 147, 869± 872.
Lester W. M., Bartholomew R. A., Colvin E. D., Grimes W.
H., Fish S. H. and Galloway W. H. (1965) Reconsideration
of the uterine packing in post partum haemorrhage.
A m erican Journal of O bstetrics and G ynecology, 98,
321± 329.
Maier R. C. (1993) Control of post-partum haemorrhage
with uterine packing. American Journal of Obstetrics and
Gynecology, 169, 317± 323.
Maternal Mortality in Saudi Arabia 1989± 1992. Kingdom
of Saudi Arabia.
Morison J. E. (1978) Placenta accreta: a clinicopathologic
review of 67 cases. Obstetric and Gynecology Annual, 7,
107± 123.
National Institute of Health Consensus Development Task
Force (1981) Statement on cesarean childbirth. American
Journal of Obstetrics and Gynecology, 139, 902± 909.
Ramsbotham F. H. (1856) Principles and Practice of Obstetric
M edicine and S urgery, Fourth edition. London, John
Churchill.
Read J. A., Cotton D. B. and Miller F. C. (1980) Placenta
accreta: changing clinical aspects and outcome. Obstetrics
and Gynecology, 56, 31± 34.
Report on Con® dential Enquiry into Maternal Death in the
United Kingdom 1988± 1990 (1994) pp. 43± 44. London
HMSO, Department of Health.
Sharp K. W. and Locicero R. J. (1992) Abdominal packing
for surgically uncontrollable haemorrhage. Annals of
Surgery, 215, 467± 475.
Stanco L. M., Schrimmer D. B., Paul R. H. and Mishell D.
R. Jr (1993) Emergency peri-partum hysterectomy and
associated risk factors. Obstetrics and Gynecology, 168,
879± 883.
Svoboda J. A., Peter E. T., Dang C. V., Parks S. N. and
Ellyson J. H (1982) Severe liver trauma in the face of
coagulopathy. American Journal of Surgery, 144, 717± 721.
Taylor V. M., Kramer M. D., Vaughan T. L. and Peacock S.
(1994) Placenta previa and prior caesarean delivery. How
strong the association? Obstetrics and Gynecology, 84,
55± 57.
Zelop C. M ., Harlow B. L., Frigoletto F. D., Jr Safon L. D.
and Saltzm an D. H . (1993) E m ergency peripartum
hysterectomy. A merican Journal of Obstetrics and G yne-
cology, 168, 1443± 1448.