CASE REPORT

Sengstaken-Blakemore Tube to Control MassivePostpartum Haemorrhage

R P ]aparaj, MOG*, S Raman, FRCOG**

'Maternal Fetal Medicine Unit, Department of Obstetrics and Gynecology, Hospital Ipoh, Jalan Hospital, 30990, Ipoh,Perak, "Fetal Medicine and Gynaecology Centre, 1st Floor WIM, 7, Jalan Abang Haji Openg, Taman Tun Dr Ismail, 60000Kuala Lumpur

Introduction

Massive postpartum haemorrhage is a knowncomplication after Caesarian section for placentaprevia. The bleeding in these cases is due to thepersistent oozing from the placental site due to theinability of the lower segment of the uterus tocontract adequately. In many cases, when such acomplication occurs there is a predilection for amore radical approach of management whichincludes ligation of the internal iliac arteries andhysterectomy, when simple measures like suturingthe placental bed fails. These radical measures arerisky and require a skilled surgeon. Recently therehas been a reemergence of the role of uterinepacking in dealing with such cases especially in theyoung primiparous patients. We report a case of aprimigravida who had a lower segment Caesariansection for bleeding placenta previa who later

developed postpartum haemorrhage and wassuccessfully managed by using a Sengstaken­Blakemore tube as a tamponade for the loweruterine segment. This is also the first such casereport in Malaysia.

Case Report

A 27-year-old Chinese primigravida was admittedto the labour ward of Hospital Seri Manjung on 18September 1998 at 38 weeks gestation withpainless vaginal bleeding. Her antenatal coursewas uncomplicated. On admission, her vital signswere stable. Her uterus corresponded to a termsize uterus with a singleton fetus. The fetal heartwas heard. An ultrasound scan of the uterusconfirmed the examination findings and alsorevealed a posterior Type II placenta previa.

This article was accepted: 14 March 2003 .Corresponding Author: Japara; Robert Peter, 74, Jalan Rishah 6, Taman Rishah, 30700 Ipoh, Perak (email: ;apara;@hotmail.com)

604 Med J Malaysia Vol 58 No 4 October 2003

Sengstaken-Blakemore Tube to Control Massive Postpartum Haemorrhage

Sengstaken-Blakemoretube

After the operation, the Sengstaken-Blakemoretube was then pulled gently from below to ensurethat the gastric balloon fitted tightly in the lowersegment of the uterus.

The overall blood loss was estimated to be 2000ml.Transfusion in total, consisted of 6 units of wholeblood, 3 units of cryoprecipitate and 8 units offresh frozen plasma. Intravenous antibiotic coverwas given for 7 days. The patient was monitored inthe intensive care unit. Her condition was stableover the follOWing 6 hours. There was no vaginalbleeding noted and the uterus was well contractedwith the aid of the oxytocin infusion (80 IU inSOOml of Hartman's solution, running over 12hours). The gastric balloon was thendecompressed by removing 100ml of normalsaline from the balloon. This was to prevent bloodfrom accumulating above the gastric balloon in theuterine cavity. There was no bleeding noted. Thetube was then removed the following morning.The patient made a full and uncomplicatedrecovery and was discharged after 7 days.

Sengstaken-Blakemore tube withinflated gastric balloon placed at thelower uterine segment

Fig 1:

Blood was crossed matched and the patient wassent up to the operation theatre foran emergencylower segment Cesarean section. The operationwas done under spinal anesthesia. A healthy babyboy weighing 3.38kg was delivered. Intra­operatively the placenta previa was confirmed andestimated blood loss was 800ml. An oxytocininfusion (40 IU in SOOml of Hartman's solution)was commenced. After confirming that theplacenta was completely removed haemostasissecured, the uterine incision on the lower uterinesegment was sutured in 2 layers in the usualmanner.

One hour after the Cesarean section, while thepatient was still in the operating theatre recoveryroom, the patient was noted to have fresh vaginalbleeding again. The estimated blood loss wasaround SOOml. The oxytocin infusion was doubledto 80 IU in SOOml of Hartman's solution and ablood transfusion was also started. Anintramuscular injection of Carboprost· 2S0mcg wasgiven. The uterus appeared to be well contractedbut the vaginal haemorrhage continued. She lostanother SOO ml of blood over the following twohours of observation. Her blood pressure wasaround 80/S0-mmHg and pulse rate around 110beats per minute. Her clotting studies werenormal. At this stage, it was decided that sherequired an exploratory laparotomy.

During the relaparotomy, bleeding from theplacental site was noted. A few haemostaticsutures were placed at the placental site. Bilateraluterine artery ligation was also done. Thebleeding from the placental site was still persistent.In view of the age and parity of the patient, wedecided to manage the haemorrhageconservatively. Uterine packing was considered asa control measure to stop the haemorrhage. Toachieve adequate uterine tamponade, aSengstaken-Blakemore esophageal varices tubewas used. The gastric end was left in the uterinecavity and inflated with 200ml of normal saline.The proximal end of the tube was pushed throughthe cervical os and out into the vagina (Figure l).The uterus was then sutured in the usual way.

Med J Malaysia Vol 58 No 4 October 2003 605

CASE REPORT

Discussion

The control of uterine fu:a~monrhage by the use ofintracavitary packing nas L;een described for over acentury. Traditionally, the packing was performedusing several yards of gauze placed inside theuterine cavity. This technique fell out of favour, asthere were fears of continued bleeding, uterinetrauma, and infection associated with thisprocedure. The increasingly effective medicalmeans to treat uterine atony also led to its fall fromfavour.

Lately, uterine packing has found its place again inthe management of massive postpartumhaemorrhage. The procedure can be consideredfor haemorrhage secondary to uterine atony,placenta accreta and placenta previa. Theprinciple behind uterine packing is the tamponadeeffect it has on the bleeding surfaces of the uterinewall especially from the placental site at the lowersegment of the uterus. Haemorrhage from thisarea is quite resistant to medical management asthe muscular component in this area of the uterusis deficient.

Apart from gauze, other modalities used for thesetamponade effects are the Foley's balloon' and theSengstaken-Blakemore tube>. Balloon tamponadeis quite commonly used in controlling bleedingafter a variety of gynaecological procedures suchas laser ablation and evacuation of retained

product of conception. In these cases, a Foley'scatheter is usually used.

The volume of the immediate postpartum uterusmay be too large for effective tamponade to beachieved with a Foley's catheter. A Sengstaken­Blakemore tube was used for our patient because,the gastric balloon is much larger than the Foley'scatheter and this would provide a greatertamponade on the lower uterine segment. Thistechnique should only be considered after theremoval of the placenta (except in placentaaccreta) and when uterine rupture and lowergenital tract lacerations have been ruled out. Wealso emphasize here, the attention that needs to be

paid to meticulous haemostasis at surgery. The. useof oxytocics and prostaglandins cannot be avoidedprior to attempting this method. The tube shouldbe left in situ for 12 to 24 hours. Some authorshave advocated packing of the upper vagina toavoid expulsion of the tube', We did not pack thevagina as the tube was well placed in the uterus.Some of the concerns with this procedure are,whether the gastric balloon should be inflatedbefore closure of the lower uterine segment. Wefeel that inflation of the gastric balloon before theclosure is better, compared to after the closure, asthis avoids the risk of an over inflated gastricballoon bursting through the sutures at the loweruterine segment after the lower uterine segment isclosed. However, care must be taken to avoidpuncture of the gastric balloon during the uterineclosure. The needle must not be passed throughthe decidua and must be confined to themyometrium. Another concern is whether theplacement of the inflated gastric balloon at thelower uterine segment would impair the woundhealing. This should not be a problem, as theinflated gastric balloon is only left in situ for a day.

In order to maintain uterine contractions over theballoon, oxytocin infusion (40 ID in 500ml ofnormal saline) is run over a period of 12 to 24hours. Prophylactic antibiotic therapy should alsobe administered to minimize the risk of sepsis.The patient is closely monitored including the vitalsigns, fluid input-output, fundal height and vaginalloss. If the patient's condition worsens, thereshouldn't be any hesitation to resort to moreradical management i.e. bilateral ligation of theinternal iliac arteries or hysterectomy. Thefollowing day, when the patient's condition isstable, the gastric balloon is deflated, leaving it insitu, and the patient is observed for approximately30 minutes. If there is no further bleeding, theballoon is withdrawn.

In conclusion, Sengstaken-Blakemore tube used asuterine tamponade is an effective technique for thecontrol of massive postpartum haemorrhagesecondary to bleeding from lower uterinesegment, as demonstrated in this case.

mill iUUU'"'' E 1.17 FBI • 1LJ712 111117 7711/l1T11 171717.1.

606 Med JMalaysia Vol 58 No 4 October 2003

Sengstaken-Blakemore Tube toControl Massive Postpartum Haemorrhage

Marcovici I, Scocia B. Postpartum hemorrhage andintrauterine balloon tamponade: a report of threecases. J Reprod Med 1999; 44: 122-26.

Katesmark M, Brown R, Raju K S. Successful use ofa Sengstaken-Blakemore tube to control massive

Malaysia Vol 58 No 4 October 2003

postpartum haemorrhage. Br J Obstet Gynaecol1994; 101: 259-60.

3. Mousa H A, Walkinshaw S. Major postpartumhaemorrhage. Curr Opin Obstet Gynecol 2001; 13:595-603.

607