Volume 196, Issue 5 , Pages e9-e10, May 2007
The uterine sandwich for persistent uterine atony: combining the B-Lynch compression suture and an intrauterine Bakri balloon
Article Outline
Objective
The objective of the study was to evaluate the effectiveness of a combination of surgical interventions for control of postpartum hemorrhage.
Study Design
At cesarean delivery, patients with persistent bleeding from uterine atony after the administration of oxytonics were treated with the placement of a B-Lynch suture. When the B-Lynch failed, subsequent placement of an intrauterine Bakri balloon followed. This combination is termed the uterine sandwich.
Results
The uterine sandwich was successful for all 5 patients undergoing this approach. The median nadir hematocrit was 21.1% (range 20.1% to 28%). The balloon was in place for a median duration of 11hours (range10-24 hours). The median volume infused into the balloon was 100 mL (range 60- 250 mL). No complications were observed.
Conclusion
Placing an intrauterine Bakri balloon in conjunction with the B-Lynch uterine compression suture was successful in treating uterine atony.
Key words: compression suture, intrauterine balloon, postpartum hemorrhage
Options for the management of postpartum hemorrhage resulting from uterine atony include uterotonics, selective devascularization by either suture ligation or angiographic embolization, uterine compression sutures, intrauterine packing, and hysterectomy. The goals of these therapies are to minimize blood loss, preserve fertility, and avoid life-threatening complications such as shock, hypoxic encephalopathy, renal failure, and coagulopathy.1, 2
Whereas many of the therapies for uterine atony are utilized in combination, no prior reports have described the techniques of external compression sutures with intrauterine tamponade. The present report discusses several cases utilizing this methodology.
Materials and Methods
The present study is a retrospective review of cases at Central Baptist Hospital (Lexington, KY) between May 2003 and May 2006. This study was approved by the Institutional Review Board of Central Baptist Hospital.
Patients who experienced persistent bleeding from uterine atony at the time of cesarean delivery after the administration of oxytonics were candidates for the uterine sandwich technique. Patients were identified because of the necessity for intraoperative consultation with a maternal-fetal medicine specialist. Intraoperative assessment through the hysterotomy site noted persistent fundal uterine bleeding and a hypotonic uterus not responsive to uterotonics in all cases. After a B-Lynch compression suture using 1-Vicryl was performed, the extent of uterine bleeding was reassessed and an intrauterine Bakri balloon (Cook Women’s Health, Spencer, IN) was placed if persistent bleeding was encountered.
The hysterotomy site was finally closed and the balloon filled with normal saline while visualizing the uterine response to increasing tamponade. Insufflation was stopped with evidence of uterine distension attempting to avoid undue blanching at the compression suture sites by visual inspection. Prophylactic antibiotics were administered while the balloon was in place.
Results
Five patients underwent this approach for management of uterine atony. The combined technique was successful in all cases evaluated to avoid the need for hysterectomy and reduce further blood loss. The median estimated blood loss was 2500 mL (range 2000-3000 mL). The median nadir hematocrit was 21.1% (range 20.1% to 28%) down from a preoperative median hematocrit of 33.1% (range 30.8% to 38.8%). The median transfused volume of PRBCs was 2 U (range 2-5 U). No platelet transfusions were required. All patients received high-dose oxytocin (more than 40 U per liter) and prostaglandin F2α as initial therapies unless contraindicated, as in 1 patient. Three patients received Methergine. Uterine artery ligation had been attempted in 1 patient prior to the interventions and no hypogastric artery ligations were performed. The median operative time was 2.3 hours (range 1.75 to 4.25 hours).
The balloon was in place for a median of 11 hours (range 10-24 hours). The median volume infused into the balloon was 100 mL (range 60-250 mL). One patient developed endomyometritis after an antepartum diagnosis of chorioamnionitis. Another patient developed postpartum oliguria, which improved with transfusion and fluid resuscitation. No case was complicated by acute tubular necrosis, hypoxic encephalopathy, or death. No uterine erosions were documented. Two patients have had a subsequent pregnancy, and 1 was complicated by uterine atony at the time of elective repeat cesarean delivery.
Comment
The management of uterine atony is dictated by several considerations including hemodynamic status and the desire to preserve fertility. Although hysterectomy is a definitive treatment, there is a desire to avoid the intervention particularly in younger women. We developed a technique to provide a less aggressive approach to uterine atony when other preliminary treatments have been attempted and persistent bleeding is evident. Two surgical treatment strategies have been combined to provide a successful alternative to hysterectomy and its associated surgical risks. We termed this combination of external compression and internal tamponade the uterine sandwich technique because it applies forces to both surfaces of the myometrium. We theorized that the B-Lynch suture would limit the migration of the fundus from the outward pressure of an intrauterine balloon and thereby enhance tamponade against blood flow through the spiral arteries.
We have undertaken this combination in women who had undergone cesarean delivery complicated by uterine atony. These women had failed more conservative therapies including the use of oxytocin, prostaglandins, and ergotamines. If these therapies do not control hemorrhage, further considerations have included uterine devascularization, compression suture, and uterine tamponade, each of which may be effective in isolation. Anecdotally, we noted most patients will not require further intervention after placement of a compression suture; however, when further bleeding is encountered, the question of what to do next is a clinical dilemma. No randomized trials have been published discussing the utility of various surgical interventions in a group of women who fail initial therapies.
Although we could not identify a prior report discussing this combined treatment strategy for women with uterine atony, Price et al3 have reported a woman with abnormal placentation successfully undergoing the combination. The relative ease of these interventions merits their consideration when faced with a circumstance of unremitting hemorrhage.
The potential complications of combined interventions include uterine necrosis resulting from poor perfusion, uterine lacerations, and endomyometritis. B-Lynch et al4 reported a uterine laceration associated with the use of a compression suture. To minimize the risk of uterine injury, we inflate the intrauterine balloon after placement of the compression suture while visualizing the myometrial response. We are mindful that excessive intrauterine pressure may result in a uterine laceration or subsequent necrosis. We also attempt to minimize the duration of the balloon insufflation to avoid these concerns, and we have decreased the duration of balloon time from 24 hours in our first patient (maximum recommended by the manufacturer) to 10 hours in our final 2 patients.
The surgical therapies for persistent hemorrhage from uterine atony, after medical therapies have been insufficient may involve multiple interventions. After a vaginal delivery and prior to laparotomy, a tamponade test with an intrauterine balloon may prove useful to avoid unnecessary intervention.5 If cesarean delivery has been performed or if laparotomy after vaginal birth is necessary, a sequential application of surgical interventions based on ease and rapidity of therapy includes compression suture and/or selective devascularization, followed by the uterine sandwich, and finally hysterectomy. Such a step-wise approach may minimize confusion in the operating room and the need for hysterectomy. Although the preliminary results with the uterine sandwich are encouraging, further evaluation is necessary.
References
- . Improving hospital systems for the care of women with major obstetric hemorrhage. Obstet Gynecol. 2006;107:977–983
- . Successful reduction of massive postpartum haemorrhage by use of guidelines and staff education. BJOG. 2004;111:495–498
- . Application of the B-Lynch brace suture with associated intrauterine balloon catheter for massive haemorrhage due to placenta accreta following second-trimester miscarriage. J Obstet Gynaecol. 2006;26:267–268
- . Partial ischemic necrosis of the uterus following a uterine brace compression suture. BJOG. 2005;112:126–127
- . The “tamponade test” in the management of massive postpartum hemorrhage. Obstet Gynecol. 2003;101:767–772
PII: S0002-9378(06)02202-2
doi:10.1016/j.ajog.2006.10.887
© 2007 Mosby, Inc. All rights reserved.
Volume 196, Issue 5 , Pages e9-e10, May 2007
