Postpartum hemorrhage: Use of an intrauterine hemorrhage-control device

INTRODUCTION — Postpartum hemorrhage (PPH) is an obstetric emergency that can be managed by a variety of potentially effective medical (eg, uterotonics, tranexamic acid), nonsurgical (eg, uterine massage; intrauterine tamponade with an intrauterine balloon, vacuum device, or packing), and surgical (eg, repair of a deep laceration or tear in the uterus, cervix or vagina; uterine artery ligation; curettage; uterine compression sutures; hysterectomy) interventions (table 1). Management is often a multistage, sequential process beginning with less invasive interventions and proceeding to more invasive interventions, as appropriate for the cause and severity of hemorrhage. In clinical practice, brisk hemorrhage or patient instability may necessitate deploying such measures in parallel (simultaneously) to ensure hemostatic control.

This topic will discuss the role of intrauterine hemorrhage-control devices, such as an intrauterine balloon, intrauterine vacuum device, and/or packing, in the management of PPH. Detailed information on causes of PPH, patient assessment and planning, and medical and surgical management is found separately.

●(See "Overview of postpartum hemorrhage".)

●(See "Postpartum hemorrhage: Medical and minimally invasive management".)

●(See "Postpartum hemorrhage: Management approaches requiring laparotomy".)

●(See "Secondary (late) postpartum hemorrhage".)

PATIENT SELECTION

Candidates — Intrauterine hemorrhage-control devices (balloon, vacuum, packing) are most commonly used for patients with bleeding due to uterine atony not controlled by uterotonic medications and uterine massage. One of these methods is often attempted before resorting to more invasive or permanent surgical approaches (table 1). (See "Postpartum hemorrhage: Medical and minimally invasive management", section on 'Manage atony'.)

An intrauterine hemorrhage-control device may also be used for patients with lower uterine segment bleeding following delivery complicated by placenta previa, low-lying placenta, or placenta accreta spectrum, or those with delayed (secondary) PPH. Control of bleeding in such patients is variable and other hemostatic measures (eg, compression sutures) are also often used. This is discussed in detail separately. (See "Placenta accreta spectrum: Management" and "Secondary (late) postpartum hemorrhage", section on 'Definition/diagnosis'.)

Contraindications — Contraindications to an intrauterine hemorrhage-control device include:

●Allergy to any component of the device or packing material.

●Retained products of conception; a hemorrhage-control device should not replace evacuation of pregnancy tissue.

Chorioamnionitis/endometritis are not contraindications to an intrauterine hemorrhage-control device, but appropriate antibiotic coverage should be provided. (See "Postpartum endometritis", section on 'Treatment'.)

Alternative methods of management should also be employed in clinical settings where an intrauterine hemorrhage-control device is unlikely to be effective (eg, uterine abnormalities that prevent effective tamponade, suspected uterine rupture or other genital tract laceration), or when immediate hysterectomy is likely to be needed to save the patient's life. (See "Postpartum hemorrhage: Medical and minimally invasive management" and "Postpartum hemorrhage: Management approaches requiring laparotomy".)

TYPES AND EFFICACY

Intrauterine balloon tamponade

●Types – Multiple types of balloon catheters have been used for intrauterine tamponade, ranging from devices specifically designed and marketed for intrauterine use (eg, Bakri, BT-cath, ebb complete) (table 2) to those designed for other indications (table 3). All apply inward to outward hydrostatic pressure against the uterine wall; although the pressure may not be in excess of systemic arterial pressure, compression of proximal endometrial and myometrial blood vessels appears to reduce blood flow and facilitate clotting [1-3]. Mechanical stimulation of myometrial contraction may contribute to a reduction in blood flow [4].

Devices designed specifically for intrauterine tamponade feature at least one balloon that remains collapsed until it is placed within the uterine cavity and a double lumen catheter to allow balloon filling through one portion of the catheter and drainage of blood from the uterus (figure 1). Most balloons marketed specifically for intrauterine use come preassembled and offer an attachment that allows rapid filling using a bag of crystalloid rather than having to manually fill the balloon using a 60 mL syringe. Each device is made from slightly different materials, has a unique size and volume, and has unique properties. Differences in the distensibility and maximum possible fill volume have been demonstrated in one in vitro trial [5]; however, rarely will such large volumes be required to achieve effective tamponade.

●Efficacy – The efficacy of intrauterine balloon tamponade in controlling bleeding is unclear. As such interventions alone have the potential to increase blood loss (eg, cervical trauma or uterine perforation resulting from balloon placement) (see 'Complications' below), introducing uterine balloon tamponade in resource-limited settings on its own without access to other lifesaving interventions (eg, surgical interventions, blood products) may not reduce PPH-related deaths or morbidity [6]. Local policies, resources, and outcomes should be monitored as part of ongoing quality management practice.

Randomized trial data are limited:

•In a meta-analysis of nine randomized trials from countries in the Middle East and Africa evaluating the efficacy of mechanical and surgical interventions for controlling PPH, condom catheter compared with normal care (ie, misoprostol) resulted in more patients with blood loss of ≥1000 mL (relative risk [RR] 1.52, 95% CI 1.15-2; 1 study, 113 patients); however, the quality of evidence ranged from very low to low and the analysis included studies with an uncertain risk of bias [7]. Other outcomes (ie, mortality due to bleeding, hysterectomy, blood transfusion) were similar between groups.

•Two other randomized trials also demonstrated worse outcomes (eg, increased PPH-related surgery and death) with condom catheter use in low-resource settings [8,9]. However, these studies also had several limitations, including being underpowered, underutilization of tamponade in the tamponade group, suboptimal management of PPH in both the tamponade and no-tamponade group, and delayed recognition of uterine atony [8].

•In the meta-analysis of randomized trials described above, latex balloon intrauterine tamponade compared with normal care (ie, uterine massage, uterotonics) resulted in similar rates of hysterectomy (one study, 240 patients), but the certainty of evidence was very low [7].

•In a randomized trial including 66 patients with PPH following vaginal birth, treatment with the Bakri balloon compared with condom catheter tamponade resulted in similar success and complication rates [10]. The Bakri balloon required a shorter duration to stop the bleeding (9.1 versus 11.8 minutes), but this may not be clinically significant, and the certainty of evidence was very low.

Data from nonrandomized studies (with or without inclusion of randomized trials) are also limited but describe more favorable results:

•In a meta-analysis including over 4700 patients with PPH (7 randomized or cluster randomized trials, 15 nonrandomized studies, 69 case series), the overall pooled uterine balloon tamponade success rate was 85.9 percent (95% CI 83.9-87.9), with the highest success in uterine atony (87.1 percent) and placenta previa (86.8 percent) and the lowest in placenta accreta spectrum (66.7 percent) and retained products of conception (76.8 percent) [11]. Hemorrhage stopped more frequently when balloons were placed after vaginal compared with cesarean birth (87 versus 82 percent, respectively). In this meta-analysis, success was defined as arrest of bleeding without maternal death or additional surgical or radiologic interventions after balloon placement. However, this is a narrow definition as the balloon can be useful as a temporizing measure to reduce blood flow to the uterus and thus provide time to initiate and continue other measures to stabilize the patient.

•In a subsequent retrospective study including 1761 patients with PPH treated with tamponade, efficacy was also high (88.9 percent); patients in whom tamponade failed had higher rates of preeclampsia (adjusted odds ratio [aOR] 2.3, 95% CI 1.3-4), cesarean birth (aOR 4.2, 95% CI 2.9-6), and uterine rupture (aOR 14.1, 95% CI 2.4-83) [12].

Vacuum-induced tamponade — Vacuum-induced tamponade is a novel hemorrhage-control device that applies low-level intrauterine vacuum (70 to 90 mmHg) to rapidly evacuate blood and clot and facilitate physiologic uterine contractions (figure 2) [13-17].

The efficacy of vacuum-induced tamponade appears promising [18]. In a prospective multicenter single-arm treatment study including 106 patients with PPH, the vacuum device controlled bleeding in 94 percent of patients; control of bleeding typically occurred within two to five minutes of beginning the vacuum [16]. There were no serious procedure-related adverse events. Forty participants received at least one unit of blood transfusion; five of the six cases in which the device was unsuccessful required additional treatment to control atony-related hemorrhage, and the remaining case was successfully treated for a previously unrecognized cervical laceration. Based on this study, the US Food and Drug Administration granted premarket approval for use in the United States in August 2020 [19].

Subsequently, in a retrospective multicenter postmarketing registry including 800 patients treated with vacuum-induced tamponade from 16 academic centers, treatment success (defined as bleeding control after insertion with no treatment escalation or bleeding recurrence) occurred in 84 and 93 percent of patients following cesarean and vaginal birth, respectively [20]. Success rates were higher for those with isolated uterine atony (88 and 96 percent, respectively; 527 patients) and when the device was inserted prior to 3000 mL of cumulative blood loss. Success rates were similar in patients who delivered at <34 and ≥34 weeks. Time to bleeding control was available for 392 patients, of whom approximately two thirds experienced control of bleeding within five minutes; median indwelling time after cesarean and vaginal birth was 4.6 and 3.1 hours, respectively. Rebleeding after device removal occurred in 4.1 and 2.8 percent of cesarean and vaginal births. Serious device-related adverse events (eg, endometritis, hemorrhagic shock) occurred in three patients. While this study validated the findings in the original premarketing cohort, additional studies are needed to further evaluate the efficacy, effectiveness, and safety of vacuum-induced tamponade.

Vacuum-induced tamponade using an intrauterine balloon attached to suction has also been described [21,22].

Intrauterine packing — Various types of packing can be used for intrauterine tamponade.

●Plain gauze or laparotomy sponges – Traditionally, sterile gauze or laparotomy sponges have been used to pack the uterine cavity. Originally, these types of gauze packing were developed for the battlefield to allow quick temporization of traumatic bleeding and improve survival to surgery. The primary mechanism by which uterine packing works is similar to that of balloon tamponade: increased intraluminal pressure resulting in a reduction in uterine artery perfusion pressure.

Despite studies from the early to mid-1900s describing success rates close to 100 percent, uterine packing fell out of favor without a clear reason other than a belief that the procedure was unphysiologic and therefore unacceptable [23], but continues to be used in settings where other hemorrhage-control devices are not available.

●Hemostatic gauze – Specialized hemostatic or trauma gauze is preferable to plain gauze, especially in the setting of heavy bleeding or if there is concern for coagulopathy (table 4) [24-26]. Hemostatic gauze confers superior benefit due to the additional hemostatic mechanisms of the substances with which they are impregnated: kaolin-laden gauze activates the body's own coagulation cascade, and chitin-laden gauze forms a gel to which blood cells adhere outside the clotting cascade.

Contemporary small studies of the use of hemostatic gauze for tamponade report success rates of >90 percent [27-29].

●Compressed mini-sponges – A novel packing device contains compressed mini-sponges encased in a mesh pouch; a long, tubular applicator facilitates placement [30]. The mini-sponges rapidly absorb blood, expand and conform to the shape of the uterus, and exert intraluminal pressure for up to 24 hours.

In a prospective study including nine patients experiencing blood loss of >500 mL after vaginal birth, all nine patients experienced cessation of bleeding within one minute of placement of a mini-sponge tamponade device; no adverse events were noted related to the device, and no patients experienced a second bleeding episode [30]. Further studies with larger numbers of participants are planned.

CHOICE OF METHOD — The choice of method (ie, intrauterine balloon, vacuum, packing) is largely driven by local availability, provider preference, and cost rather than patient factors (eg, estimated blood loss, degree of cervical dilation) [31]. In the setting of ongoing uterine bleeding, rapid use of any method is often more important than the specific use of one type over another when a choice is available.

While intrauterine balloon tamponade has historically been the preferred method of tamponade in national guidelines for management of PPH [32], due to the ease of use and average short indwelling time, vacuum-induced tamponade has become popular in centers where this equipment is available. Packing remains an important option, especially in settings where other methods are not available. Regardless of the method, tamponade should be initiated before coagulopathy develops as it is less likely to be successful if clotting is impaired.

Few comparative studies have been performed and do not provide clear guidance as to the superiority of any one method [7,24,29,33,34]. In one cohort study of 378 patients with PPH, those managed with vacuum (52 percent) compared with balloon tamponade had similar quantitative blood loss after placement, rates of blood transfusion, and discharge hematocrit [33]. Similarly, in an earlier cohort study of 114 patients with PPH, those treated with vacuum (32 percent) compared with balloon tamponade had similar rates of intensive care unit admission, uterine artery embolization, and hysterectomy [34]. However, median estimated blood loss was lower in the vacuum group (1500 versus 1875 mL) and fewer patients in the vacuum group required transfusion with ≥2 and ≥4 units packed red blood cells. Clinical trials are needed to establish if any one device is superior.

Advantages and disadvantages of all three methods are detailed in the table (table 5).

DEVICE INSERTION

General principles

●Be prepared. Identify patients at high risk of PPH and have appropriate uterotonics, blood products, and equipment readily available. Because PPH can also occur in low-risk patients, the labor and delivery unit should always have the necessary supplies and equipment should be easily accessible.

●Perform tamponade before coagulopathy develops, as it is less likely to be successful once coagulopathy has occurred.

●To facilitate placement, the procedure should be performed with adequate lighting, appropriate patient positioning, and assistance from one or more people.

●Inspect the packaging and the device before use; avoid use if damaged.

●Empty the bladder to facilitate palpation and contraction of the uterus, and placement of the intrauterine hemorrhage-control device. This may be performed using an indwelling (ie, Foley) or intermittent catheter. A Foley catheter confers the additional benefits of facilitating voiding with the tamponade or packing in place, and close monitoring of urine output.

●Assess cervical dilatation. Some degree of dilatation (usually 1 to 2 cm) is necessary for intrauterine balloon placement (ring forceps may be used to gently dilate the cervix but should be abandoned if gentle use does not open the cervix to avoid creating a false track). For vacuum-induced tamponade, the manufacturer recommends dilation of at least 3 cm prior to placement. (See 'Device insertion' above.)

●Continue concurrent medical management of hemorrhage, including use of uterotonic medications and, when appropriate, tranexamic acid. (See "Postpartum hemorrhage: Medical and minimally invasive management".)

●If unable to successfully place the device, or if bleeding continues despite intrauterine tamponade, initiate additional hemostatic procedures (eg, compression sutures, uterine artery ligation, hysterectomy). (See "Postpartum hemorrhage: Management approaches requiring laparotomy".)

Intrauterine balloon

Equipment — The following equipment should be assembled:

●Balloon catheter of choice.

●Sterile fluid for irrigation (500 to 1000 mL of crystalloid to allow adequate fluid for instillation plus additional to spare if needed).

●At least two large Luer lock syringes (≥60 mL).

●Basin to facilitate quick syringe filling.

●500 to 1000 mL bag of crystalloid (sterile saline, water, or Lactated Ringer) for kits with a bag-spike adapter for rapid filling. Balloons should never be insufflated with air or carbon dioxide due to the risk of air embolism.

●Atraumatic forceps (eg, ring forceps).

●Portable ultrasound machine to look for retained products of conception (remove if present) and guide placement (performed by an assistant during transvaginal placement).

●Antiseptic solution (eg, povidone-iodine).

●Gauze or vaginal packing (if vaginal packing or balloon stabilization needed).

●Straight and/or indwelling bladder catheter.

●Urine collection bag with urometer (for monitoring and accurate measurement of ongoing bleeding through the efflux port).

Technique — Each balloon catheter is unique; therefore, insertion technique varies slightly depending on the device used. Provider preferences can also result in slight variations in technique.

●To place the balloon after a vaginal birth or after closure of the abdomen following cesarean birth, place the patient in the lithotomy position in stirrups and insert the balloon transvaginally. Whenever possible, use ultrasound guidance during placement.

•Cleanse the cervix and vagina with antiseptic solution, such as povidone-iodine.

•Grasp the anterior cervical lip with ring forceps.

•Apply gentle traction so the cervical canal and uterine cavity are aligned.

•Use forceps (smooth long dressing, ring) to insert the balloon as high in the cavity as possible without using excessive force (which can lead to uterine perforation). The nondominant hand can be used to exert transabdominal pressure to maintain alignment of the cervical canal and uterine cavity.

The balloon can also be placed into the uterine cavity manually.

•If you encounter resistance before the entire balloon is above the internal cervical os, readjust the angle of insertion or abandon the procedure.

•Placement of hemostatic gauze packing plus intrauterine balloon tamponade has been described [35]. The gauze is first placed within the intrauterine cavity to promote a local hemostatic effect; then, the balloon is placed to provide direct intraluminal pressure. With this approach, removal of both the balloon and gauze is essential.

●To place the balloon at cesarean birth before closure of the hysterotomy, bend the patient's legs at the knees with the hips externally rotated ("butterfly" position). Alternatively, to facilitate placement of the balloon and monitor for ongoing bleeding, the patient's legs can be supported in raised stirrups.

•Insert the balloon through the hysterotomy and into the uterine cavity.

•Pass the stem of the catheter through the cervix and into the vagina. Some catheters may permit removal of a stopcock to facilitate placement through the cervix; the stopcock must be reattached to the tubing prior to fluid instillation. Ask an assistant to pull the end of the catheter out through the introitus.

Use of a traction stitch to keep the balloon within the uterine cavity has been described [36].

•Carefully close the uterine incision to avoid puncturing the balloon prior to fluid instillation. Uterine compression sutures can be used with balloon tamponade if the balloon alone does not control bleeding. (See 'Surgical intervention' below.)

●To place the balloon after the hysterotomy is closed but the abdomen is still open following cesarean birth, an assistant can insert the balloon transvaginally (as described above for vaginal birth), and the surgeon can monitor insertion from above to ensure appropriate placement and that the balloon fills appropriately and does not disrupt the hysterotomy repair. Any intra-abdominal bleeding following balloon inflation must also be addressed.

Uterine compression sutures can be used with balloon tamponade if the balloon alone does not control bleeding. Care must be taken to avoid puncturing the balloon during placement of compression sutures. (See 'Surgical intervention' below.)

Filling the balloon

●Verify the position of the balloon with transabdominal ultrasound (or perform real-time, continuous ultrasound guidance if an assistant is available).

●Fill the balloon(s) with warm sterile fluid (warmth enhances the coagulation cascade) until slight resistance to further instillation is encountered and bleeding slows or stops.

●Ideally, one person holds the balloon in place (vaginally) while a second person instills fluid into the balloon to help prevent expulsion from the uterine cavity.

Perform a tamponade test. The tamponade test is positive if bleeding stops when the balloon is partially or completely filled and negative if bleeding fails to stop with maximum balloon filling. Cessation of bleeding from the lower segment often occurs between 250 and 300 mL of instillation, but more fluid (>500 mL) may be required to effectively stop bleeding in cases of uterine atony [37].

●Adding an additional 50 to 100 mL of fluid helps keep the balloon in place after bleeding stops. The maximum manufacturer-recommended fill volumes vary by type of balloon catheter (table 2 and table 3); actual filling capacities (as demonstrated by in vitro trials prior to balloon rupture) may be higher and are important to note if instillation of additional fluid is required to stop bleeding [5].

●If necessary, a vaginal pack may be placed to keep the balloon in position; if a double-balloon device is used, the vaginal balloon is filled to aid retention.

●Attach a collection bag, ideally with a flowmeter (such as for an indwelling urinary catheter), to the drainage port.

●Mark the level of the fundus to facilitate posttamponade monitoring. If the fundus rises (with or without visible bleeding from the drainage port), the patient should be evaluated for occult intrauterine bleeding.

Intrauterine vacuum

Equipment — The following equipment should be assembled:

●Vacuum device

●Vacuum canister and sterile standard vacuum tubing

●Regulated vacuum source

●Luer lock syringes (10 mL, 60 mL)

●Sterile solution to fill cervical seal

●Portable ultrasound machine

●Antiseptic solution (such as povidone-iodine)

●Tape

Technique — The intrauterine vacuum is placed using a vaginal approach, regardless of route of delivery. If used after cesarean birth, the hysterotomy should first be closed and the legs repositioned, if necessary. Closure of the abdomen after vacuum placement allows the surgeon to monitor insertion from above and ensure appropriate placement and that the vacuum does not disrupt the hysterotomy repair.

●Connect a vacuum canister and sterile standard vacuum tubing to a regulated vacuum source.

●Attach a sterile Luer tapered syringe to remove any air that is in the cervical seal.

●Fill sterile Luer lock syringe with 60 mL of sterile fluid.

●Confirm the cervix is dilated ≥3 cm to allow for placement of the device.

●Using a hand, compress the intrauterine loop near the distal tip for support and insert the vacuum transvaginally, leading with the intrauterine Loop. Avoid excessive force.

●Use gentle traction on the anterior cervical lip to stabilize the cervical opening, if needed. An instrument can be placed on the anterior cervical lip, but do not grasp the device with an instrument to facilitate intrauterine insertion.

●Insert the intrauterine loop such that it is located in the uterus and oriented in the frontal plane of the body by assuring the seal valve on the device is oriented at either 3 or 9 o'clock. Ultrasound may be used to confirm proper placement of the Intrauterine Loop within the uterus.

●After insertion, the intrauterine loop should be within the uterus while the cervical seal should be located within the vagina at the external cervical os.

●While holding the cervical seal steady, attach a sterile Luer-lock syringe with 60 mL to fill the cervical seal. If needed add an additional 60 mL of fluid.

●Set the vacuum source to 80 mmHg (±10 mmHg; maximum pressure 90 mmHg) while occluding the end of the tubing, or tissue trauma may occur.

●Confirm vacuum pressure and connect the device to the sterile vacuum tubing. Blood flow and improvement in uterine tone should be noted.

●Confirm the cervical seal remains at external cervical os and reposition if needed.

●Tape the device to the inner thigh or secure to the Foley leg pad, without tension.

Intrauterine packing

Equipment — The following equipment should be assembled:

●Gauze packing

●Atraumatic forceps (such as ring forceps)

●Long smooth forceps (without teeth)

●Portable ultrasound machine

●Straight and/or indwelling bladder catheter

●Antiseptic solution (such as povidone-iodine)

Technique — Gauze is usually placed vaginally, regardless of the route of delivery. At cesarean birth, it is placed after closure of the hysterotomy incision to prevent inadvertent suturing of the packing material to the uterine cavity.

●Long, single gauze ribbons are preferred to small pieces to limit the total number of fabric pieces placed. Occasionally, laparotomy sponges may be used. If more than one packing is required, we suggest tying them together and recording exactly how many gauze pieces are placed to ensure complete removal. Radio-opaque materials should be used in case evaluation for a suspected retained piece of gauze is required.

●The initial steps are similar to balloon placement, but instead of inserting a balloon, gauze is gently placed through the cervix to the fundus and layered within the uterine cavity either manually (requires a fully dilated cervix) or using packing forceps.

●The gauze can be packed all the way down the vagina, leaving a small "tail" of gauze to remain outside the introitus to allow visualization of any blood-tinged staining or ongoing bleeding and to facilitate removal. We secure this tail to the patient's thigh to facilitate identification and ensure that one does not inadvertently leave the intrauterine packing as a retained foreign body.

●Document in the medical record how many pieces of gauze were used so that it is clear (to a different provider who may remove the packing) how many pieces of gauze must be removed.

●Monitor placement with transabdominal ultrasound guidance, if available.

●Mark the level of the fundus to facilitate posttamponade monitoring. If the fundus rises (with or without visible bleeding from the drainage port), the patient should be evaluated for intrauterine bleeding.

PATIENTS WITH PERSISTENT BLEEDING

Surgical intervention — Patients with persistent bleeding despite an intrauterine hemorrhage-control device should proceed immediately to surgical hemostatic measures. (See "Postpartum hemorrhage: Management approaches requiring laparotomy".)

●External compression sutures – External compression sutures can be used when balloon (or vacuum) tamponade fails [38-40]. Generally, the decision to use a combination of a tamponade balloon and compression sutures should be made quickly, prior to leaving the operating room rather than as serial measures over time, to avoid unnecessary delay that can lead to clinical deterioration and consumptive coagulopathy.

If the balloon is already in place, the balloon should be removed or deflated prior to placing the compression suture to avoid inadvertent puncture. The balloon can then be replaced and refilled with fluid, which should still be possible, provided that the compression sutures used were placed longitudinally (such as a B-Lynch (figure 3) or Hayman stitch (figure 4)) and do not rely upon suturing the anterior and posterior uterine walls together (as with Cho or multiple box sutures). Stop the fluid instillation if blanching at the compression suture sites begins to occur as excessive reduction of myometrial blood flow could lead to uterine rupture or necrosis. (See "Postpartum hemorrhage: Management approaches requiring laparotomy", section on 'Uterine compression sutures'.)

Compression sutures may also be used with the intrauterine vacuum; however, care must be taken to avoid puncturing the cervical seal. As with balloon tamponade, care should also be taken to avoid suturing the device into the intrauterine cavity.

●Other – Other surgical measures (eg, manual compression or balloon occlusion of the aorta, uterine tourniquet, ligation of the uterine and utero-ovarian arteries) are discussed in detail separately. (See "Postpartum hemorrhage: Management approaches requiring laparotomy".)

Role of arterial embolization — Arterial embolization by an interventional radiologist can be considered in hemodynamically stable patients where personnel and facilities are readily available. As transfer to a radiology suite in necessary for the procedure, and a significant delay is likely before the uterine vessels can be occluded, embolization should not be considered an emergency procedure for managing uncontrolled PPH. (See "Postpartum hemorrhage: Medical and minimally invasive management", section on 'Consider uterine or hypogastric artery embolization'.)

POSTPLACEMENT CARE AND MONITORING

Postplacement care — In addition to routine postpartum care:

●Antibiotic prophylaxis – We suggest antibiotic prophylaxis with broad spectrum antibiotics while the balloon/vacuum/pack is in place to reduce the risk of endometritis. In our practice, we use cefazolin 2 g intravenously (IV) every six hours.

Alternative regimens may include:

•Gentamicin 1.5 mg/kg IV every eight hours plus metronidazole 500 mg orally or IV every eight hours or

•Gentamicin 1.5 mg/kg IV every eight hours plus clindamycin 300 mg orally or IV every six hours.

Evidence of efficacy of antibiotic prophylaxis for such patients is low quality. In a retrospective study of 113 patients undergoing intrauterine balloon tamponade, patients who did not receive prophylactic antibiotics compared with those who did had higher rates of endometritis (26 versus 5 percent, odds ratio [OR] 6.5, 95% CI 1.8-24.3); these results persisted after adjusting for mode of delivery and group B Streptococcus status [41].

●Uterotonic medications – Oxytocin is administered to achieve uterine tone; other uterotonic medications (eg, misoprostol, methylergonovine) may also be used. We also continue an oxytocin infusion for 6 to 12 hours to maintain uterine tone. (See "Postpartum hemorrhage: Medical and minimally invasive management", section on 'Administer additional uterotonic medications'.)

In a review of cohort studies and case reports including patients with PPH managed with intrauterine tamponade, the majority (11 studies) used oxytocin after placement of the balloon with a range of 2 to 82 hours [42]. Prolonged administration of oxytocin can result in fluid retention and hyponatremia; this is discussed in detail separately. (See "Induction of labor with oxytocin", section on 'Side effects'.)

●Blood products – Blood transfusion(s) and/or blood products should be administered, as needed, to correct prior deficits.

●Other – In patients with a balloon tamponade, periodically flushing the drainage port (if present) with sterile saline may be required to prevent occlusion of the port by clots.

Monitoring — The patient should be closely monitored for signs and symptoms of ongoing blood loss, such as pallor, dizziness, hypotension, tachycardia, confusion, uterine enlargement, abdominal pain, abdominal distension, and oliguria. This is especially important when a gauze pack is used because a large amount of blood can collect behind the pack and conceal ongoing blood loss.

Obstetric providers should be prepared to intervene surgically if the patient's hemodynamic status worsens or does not immediately improve. (See 'Surgical intervention' above.)

DURATION OF DEVICE USE AND DEVICE REMOVAL — Prior to removing the balloon or packing, we recommend ensuring that emergency surgical services and experienced staff members are readily available in the event that bleeding resumes.

Intrauterine balloon

●Duration of use – The optimal duration of balloon inflation has not been established. After bleeding has abated, the balloon is generally left in place for at least 2 hours; longer durations (12 hours) are also reasonable. Manufacturers of devices specifically designed for intrauterine use recommend balloon removal within 24 hours [43,44].

In a retrospective cohort study of 274 patients who underwent balloon tamponade (excluding those with expulsion or need for other surgical intervention within the first 2 hours), 75 percent had tamponade for >12 hours and 25 percent had tamponade between 2 and 12 hours [45]. Patients whose balloon remained in place for >12 hours had a higher rate of postpartum fever, but other outcomes (eg, estimated mean blood loss, rate of transfusion of blood products, transfusion of ≥4 units of red blood cells, uterine artery embolization, hysterectomy, intensive care unit admission) were similar between groups.

●Removal – Balloon catheters may be deflated and removed either all at once or slowly; there is no evidence to favor one approach over the other. If the balloon is deflated slowly, it can be quickly reinflated to temporize bleeding and allow preparation for surgical intervention; however, we recommend not reinflating a balloon if bleeding resumes after removal. We consider this an indication for additional interventions. (See "Postpartum hemorrhage: Medical and minimally invasive management".)

For very rapid removal, the catheter tubing may be cut, which permits free and quick drainage of the fluid from the balloon. Once the balloon is completely emptied, it may be removed by pulling gently on the tubing.

Intauterine vacuum

●Duration of use – The intrauterine vacuum device may be removed after one hour of hemostatic control and should be removed within 24 hours of insertion [46]. In an early feasibility study including 10 patients with PPH following vaginal delivery, the average time from placement to removal was 152 minutes (range 60 to 390 minutes); bleeding was controlled within 2 minutes for all ten patients [46].

●Removal – Verify the bleeding has stopped. Disconnect the tubing while the vacuum is on (to collect any blood from the tubing into the canister) and keep the tubing secure in the event re-application of the vacuum is needed. Empty the fluid from the cervical seal with a 60 mL Luer lock syringe. Keep the device in place for at least 30 minutes while monitoring for any additional bleeding. If no further bleeding, remove the device, while maintaining transabdominal fundal pressure to reduce the risk of uterine inversion.

Caution should be taken to avoid removing the vacuum device while on active suction, to prevent inadvertent uterine inversion. All fluid should be removed from the cervical seal to avoid vaginal laceration with removal.

Intrauterine packing

●Duration of use – Similar to balloon tamponade, the optimal duration of intrauterine gauze packing has not been established. We remove gauze between 2 and 12 hours after insertion if bleeding has abated. However, if bleeding saturates completely through the gauze at any time, the gauze should be removed, and additional surgical treatment rendered.

●Removal – Removal of the packing may be more easily performed in the operating room with anesthesia for patient comfort. Removal under anesthesia in the operating room is advised if the clinician suspects that bleeding will resume and require further interventions.

The patient is placed in lithotomy position or with the legs separated to allow access to the perineum and vagina. The tail of vaginal gauze packing, if present, may be moistened with sterile saline (10 to 20 mL) to facilitate removal. Gently, but quickly, pull the packing from the vagina and assess for bleeding.

COMPLICATIONS — Potential complications from intrauterine hemorrhage-control devices are uncommon. In the meta-analysis including over 4700 patients with PPH treated with intrauterine balloon tamponade (see 'Intrauterine balloon tamponade' above), complications occurred in ≤6.5 percent of patients and included [11]:

●Uterine perforation.

●Cervical trauma due to balloon inflation at an incorrect location [37,47].

●Fever – In the meta-analysis of randomized trial discussed above (see 'Intrauterine balloon tamponade' above), condom catheter balloon tamponade compared with gauze packing resulted in reduced rates of fever (relative risk [RR] 0.47, 95% CI 0.38-0.59; 1 study, 212 patients), but the quality of evidence was very low [7].

●Infection – Infection has been reported, but causality is difficult to prove as these patients have had multiple risk factors for postpartum fever. A review of uterine packing found no cases of serious infection [23].

Hysterotomy repair dehiscence and uterine rupture have also been described, though it is likely that intrauterine tamponade was not the causal factor [48-50].

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Obstetric hemorrhage".)

SUMMARY AND RECOMMENDATIONS

●Role of intrauterine hemorrhage-control devices – Postpartum hemorrhage (PPH) is an obstetric emergency that can be managed by a variety of potentially effective medical and surgical interventions (table 1). Use of an intrauterine hemorrhage-control device is one method for controlling postpartum hemorrhage unrelated to uterine rupture or other genital tract lacerations. (See 'Introduction' above.)

●Candidates – Patients with bleeding due to uterine atony not controlled by uterotonic medications or uterine massage are candidates for an intrauterine hemorrhage-control device. An intrauterine hemorrhage-control device is often attempted before resorting to more invasive or permanent surgical approaches. (See 'Candidates' above.)

●Types and efficacy – Intrauterine hemorrhage-control devices include an intrauterine balloon (table 2 and table 3 and figure 1), vacuum (figure 2), or packing (table 4). The effectiveness of these devices in controlling bleeding is unclear as data from randomized trials are limited. As such interventions alone have the potential to increase blood loss (eg, cervical trauma or uterine perforation resulting from placement), introducing uterine hemorrhage-control devices in resource-limited settings on its own without access to other lifesaving interventions (eg, surgical interventions, blood products) may not reduce PPH-related deaths or morbidity. (See 'Types and efficacy' above and 'Complications' above.)

●Choice of method – The choice of method is largely driven by local availability, provider preference, and cost rather than patient factors (eg, estimated blood loss, degree of cervical dilation). In the setting of ongoing uterine bleeding, rapid use of any method is often more important than the specific use of one type over another when a choice is available. Regardless of the approach, the device should be initiated before coagulopathy develops as it is less likely to be successful if clotting is impaired. Advantages and disadvantages of all three methods are detailed in the table (table 5). (See 'Choice of method' above.)

●Postplacement care

•We suggest antibiotic prophylaxis while the balloon/vacuum/pack is in place to reduce the risk of endometritis (Grade 2C). In our practice, we use cefazolin 2 g intravenously every six hours. Alternative regimens may be used. (See 'Postplacement care' above.)

•We also suggest administering an oxytocin infusion after placement of the balloon, vacuum, or packing to maintain uterine tone (Grade 2C). In our practice we typically administer oxytocin for 6 to 12 hours. Additional uterotonics (eg, misoprostol, methylergonovine) are also often needed to achieve uterine tone. (See 'Postplacement care and monitoring' above and "Postpartum hemorrhage: Medical and minimally invasive management", section on 'Administer additional uterotonic medications'.)

●Duration and removal (see 'Duration of device use and device removal' above)

•The balloon, vacuum, or packing is removed between 1 and 24 hours after insertion if bleeding has abated. Retention of a device longer than 12 hours has been associated with an increased risk for fever.

•If bleeding does not stop, an intrauterine hemorrhage-control device can still be an effective temporizing measure to reduce blood flow to the uterus and thus provide time to initiate and continue other measures to stabilize the patient.

•Prior to removing the balloon, vacuum, or packing we ensure that emergency surgical services and experienced staff members are readily available in the event that bleeding resumes. Removal of a gauze pack may be more easily performed in the operating room with anesthesia for patient comfort. Removal under anesthesia in the operating room is advised if the clinician suspects that bleeding will resume and require further interventions.

ACKNOWLEDGMENTS — The UpToDate editorial staff acknowledges Younes Bakri, MD; Jorge Tolosa, MD, MSCE; and Sabaratnam Arulkumaran, PhD, DSc, FRCS, FRCOG, who contributed to earlier versions of this topic review.