Tamponade test
A common challenge for the OB when a woman has a massive postpartum hemorrhage is to determine—quickly—if she requires exploratory laparotomy or can be treated with conservative measures. In one case series, investigators concluded that, in women who have severe postpartum hemorrhage that is unresponsive to uterotonics, immediate placement of a uterine tamponade balloon is an effective test for rapidly distinguishing patients who need exploratory laparotomy from those whose condition can be managed conservatively.3
In that series, women who developed postpartum hemorrhage were treated first with uterotonics, including oxytocin, ergometrine, and carboprost. Next, they underwent initial exploration to look for lacerations of the uterus, cervix, and vagina and to ensure that no placental tissue had been retained.
Sixteen subjects (most of whom had uterine atony as a provisional diagnosis) continued to have massive bleeding after those first two steps were employed. They next had a Sengstaken-Blakemore esophageal catheter placed in the uterine cavity. The catheter was then filled with 70 to 300 mL of warm saline. Gentle downward traction was applied to the catheter to ensure contact of the catheter with the lining of the uterus.
Uterine bleeding stopped quickly in 14 women after the intrauterine balloon was filled to produce the tamponade effect. Bleeding continued in two women; they underwent exploratory laparotomy. Route of delivery for these 16 women included 10 vaginal and six cesarean deliveries; two of the 10 vaginal deliveries involved prostaglandin-induced second trimester abortions. This indicates that the tamponade test can be used after either vaginal or cesarean delivery and in either the second or third trimester.
The authors of a more recent case series also reported that balloon tamponade is successful for postpartum hemorrhage in most cases.4
The Bakri balloon can also be used in conjunction with additional interventions, including uterine compression sutures and uterine artery embolization.
The uterine sandwich
Combined use of a uterine compression stitch (B-Lynch compression suture using a 1-Vicryl suture) along with the placement of the Bakri balloon—the so-called uterine sandwich—may be indicated when postpartum hemorrhage occurs after C-section in select patients. In a report of a case series of five women in whom this combined technique was first described, uterotonics had failed to control the hemorrhage and no retained placental products were detected.5 Management proceeded as follows:
- A B-Lynch compression suture was placed with a 1-Vicryl suture
- A Bakri balloon was also placed, if determined to be necessary, with the stem brought out through the vagina
- The Bakri balloon was filled with, on average, 100 mL of fluid (range, 60 to 250 mL).
This approach controlled bleeding in all five cases.
Balloon combined with uterine artery embolization
Treatment of severe postpartum hemorrhage can also be accomplished by combining the Bakri balloon and uterine artery embolization. Initial management steps include, as mentioned, administration of uterotonics; uterine massage; fluid resuscitation; examination to detect and repair cervical and vaginal lacerations; and evaluation for retained placental tissue.
If those steps do not stop hemorrhage, a Bakri balloon can be placed and IR consultation requested to perform uterine artery embolization. Gathering a team to perform uterine artery embolization often takes 1 to 4 hours; in this setting, a Bakri balloon is a useful temporizing step that reduces bleeding while the IR team is assembled.
Potentially, a lifesaver
Postpartum hemorrhage occurs after approximately 5% of deliveries. A clinician who performs, say, 160 deliveries a year will confront this complication about eight times in that span. For an event that occurs only every 6 weeks, you need to develop, and rehearse, a systematic plan of response6 (TABLE). One component of a good plan, I believe, should be the Bakri balloon, which can be a life-saving device for women whose severe postpartum hemorrhage hasn’t responded to first-line measures.