A 31-year-old nulliparous woman in established labor presented at the delivery suite at 40 + 4 weeks' gestation following spontaneous rupture of membranes with clear amniotic fluid. Her pregnancy had been uncomplicated. The first and second stages of labor progressed uneventfully and she delivered a healthy 3610-g male. The placenta was complete and intact. A second-degree perineal tear was repaired. Despite bimanual compression, bladder drainage, aggressive fluid resuscitation, oxytocin infusion and administration of additional uterotonics (syntometrine and ergometrine), there was persistent bleeding, so an examination under anesthesia was performed. The estimated blood loss at this point was 1000 mL and the patient was hypotensive and tachycardic, with a blood pressure of 90/56 mmHg and pulse rate of 110 bpm. Intraoperatively, 4 L crystalloid/colloids was administered. The integrity of the lower genital tract was visualized systematically and a thorough digital examination of the upper genital tract demonstrated no further lacerations. An additional 500 mL of clot was evacuated manually from the uterus, but this was without placental fragments. The uterus continued to be atonic intermittently, so intracervical prostaglandin F2α was administered (5 mg/mL in 20 mL normal saline, given in two 2.5-mL doses). The estimated blood loss at this point exceeded 2 L.
After excluding retained placental tissue, a Sengstaken-Blakemore balloon catheter (SBEC, Cook Medical, Brisbane, Queensland, Australia) was inserted and inflated with 320 mL normal saline. The bleeding stopped almost immediately and the uterus was well contracted. A transabdominal ultrasound examination was performed to confirm the positioning of the SBEC and, surprisingly, the inflated balloon was located not at the uterine fundus but in the lower segment (Figure 1). The endometrial thickness measured 4.1 mm and there was no intracavitary blood or retained placenta (Figure 2). A vaginal pack and an hourly urinary catheter were inserted and oxytocin infusion was continued. The intraoperative hemoglobin level was 6.6 g/dL (a drop from 13.8 g/dL preoperatively) and the coagulation profile was normal.
Postoperatively, the woman was transferred to the intensive care unit, where she received four units of packed cell transfusion. She remained clinically stable; her blood pressure increased, the tachycardia reduced and she was discharged on day 6 with a hemoglobin level of 9.3 g/dL.
To our knowledge, this is the first documented case of ultrasonographic visualization of an inserted SBEC used to create uterine balloon tamponade in a woman with massive postpartum hemorrhage. Balloon tamponade of the uterus is a well-recognized therapy in those with massive and intractable postpartum hemorrhage1, 2. An inflated balloon catheter can prevent the need for surgical intervention, rapidly identifying those women who require a laparotomy/hysterectomy; this is the basis for the ‘tamponade test’3. In this case, therefore, the SBEC was used as a diagnostic test as well as a therapy.
The ultrasonographic images captured after balloon placement demonstrated that the uterine cavity was empty, with a thin endometrium and without retained blood or placental tissue (Figure 2). The balloon was located in the lower segment of the uterus, near the isthmus (at the level of the uterine arteries). While this is, in fact, incorrect positioning of the balloon according to the manufacturers of a similar balloon catheter (the Bakri balloon tube), the intractable bleeding was controlled effectively. This incidental finding raises the possibility that the mechanism of action of balloon tamponade differs from that thought previously.
The uterus is composed of a unique interlacing network of muscle fibers, the ‘myometrium’. The blood vessels that supply the placental bed pass through this latticework of uterine muscle4. Myometrial contraction is the main driving force for both placental separation and hemostasis, through constriction of these blood vessels4. Therefore, the conventional wisdom is that balloon tamponade works primarily by applying pressure directly on the placental bed/myometrial vessels, thus mimicking normal physiology. The success of surgical compression techniques5–7 has added validity to this theory. Similarly, the generally accepted rationale for the mechanism of action of the balloon catheter is that of hemostasis through the direct pressure applied to the myometrium and placental bed8–10, provided the uterus is able to contract down on the balloon.
This case, however, highlights the possibility that the predominant mechanism of action of the SBEC may be not direct myometrial tamponade, but rather its hydrostatic pressure effect directly occluding the uterine arteries. It is well known that during pregnancy 90% of the blood supply to the uterus is via the uterine arteries11. In uncontrollable hemorrhage, the uterine arteries lose their ability to constrict because of an unidentified mechanism12. We believe that this case raises the possibility that uterine balloon tamponade causes direct compression of the uterine vessels rather than in the uterine cavity itself. Our theory is further substantiated by the current and past practices of surgical uterine artery ligation13 as well as uterine artery embolization performed during interventional radiology. These invasive techniques have varying success rates, comorbidities and potential lasting complications14. We propose that the insertion of the balloon catheter is akin to non-surgical uterine artery ligation or non-interventional uterine artery embolization.
One limitation of this case is that we did not verify objectively a decrease in uterine artery perfusion after insertion of the SBEC. This could be done by evaluating uterine artery Doppler measurements before and after insertion of the balloon catheter15. This could be a focus of future studies, although it would be difficult to achieve in an emergency situation in which the maternal condition is critical and time is of the essence.
The SBEC, or any other uterine balloon catheter, is a life-saving measure that takes minutes to insert in women with primary postpartum hemorrhage and allows for replacement of clotting factors whilst arresting the hemorrhage3. Its ultrasonographic visualization in this case raises the hypothesis of direct bilateral uterine artery tamponade as a possible mechanism of action. Future studies are required to corroborate this new theory. Both to elucidate its mechanism of action and to prevent perforation of the recently gravid uterus, we believe that sonographic confirmation of balloon catheter location should be an integral part of uterine balloon tamponade in women with massive primary postpartum hemorrhage.