Vacuum‐assisted hemorrhage control device versus uterine balloon tamponade for refractory atonic postpartum hemorrhage in a protocol‐driven care pathway—A retrospective cohort study

Abstract

Introduction

Globally, 1 in 5 maternal deaths are attributable to major postpartum hemorrhage (PPH), despite widespread uterotonic use. Intrauterine tamponade is recommended as second‐line therapy for refractory atonic PPH. Uterine balloon tamponade (UBT) achieves hemostasis through intrauterine pressure. Conversely, the vacuum‐assisted hemorrhage‐control device (VHD) promotes uterine contraction through negative pressure. Limited data compare these modalities. We compared the clinical effectiveness of UBT and VHD in refractory atonic PPH.

Material and Methods

This retrospective cohort study was conducted at the National University Hospital in Singapore and included all women treated for primary PPH with either the UBT or VHD between January 2020 and September 2025, during which ROTEM‐guided transfusion and E‐MOTIVE‐aligned practices were implemented. The composite morbidity endpoint, comprising transfusion of ≥4 units of packed red blood cells (PRBC), plasma or cryoprecipitate administration, examination under anesthesia, high dependency or intensive care unit admission, or device failure was compared between groups with logistic regression adjusted for placenta previa major and cesarean delivery.

Results

Fifty‐three women met inclusion criteria (33 UBT, 20 VHD). The composite morbidity outcome occurred less frequently with VHD than UBT (45.0% versus 75.8%; unadjusted OR 0.26, 95% CI 0.08–0.86, p = 0.03); after adjusting for placenta previa major and cesarean delivery, this difference was no longer statistically significant (adjusted OR 0.38, 95% CI 0.10–1.41, p = 0.15). Baseline characteristics were similar except for higher rates of placenta previa (45.5% versus 5.0%) and cesarean delivery (78.8% versus 50.0%) in the UBT group. Median estimated blood loss at device insertion was comparable (1500 mL). Among secondary outcomes, VHD was associated with lower median PRBC transfusion (1 vs. 2 units, p = 0.009), fewer women receiving ≥4 PRBC units (5.0% versus 36.4%, p = 0.02), fewer plasma transfusions (5.0% versus 36.4%, p = 0.02), shorter device indwelling duration (10.1 versus 24.3 h, p < 0.001), shorter hospital stay (54 versus 72 h, p = 0.007), and fewer high‐dependency admissions (45.0% versus 72.7%, p = 0.04).

Conclusions

The VHD demonstrated outcomes comparable to balloon tamponade in managing refractory atonic PPH, with lower transfusion requirements and shorter hospitalization. VHD may be a useful component of standardized PPH protocols incorporating the E‐MOTIVE bundle and ROTEM‐guided transfusion, warranting further evaluation in broader clinical settings.

In this retrospective cohort study, the vacuum‐assisted hemorrhage‐control device for refractory postpartum hemorrhage, implemented within an E‐MOTIVE and ROTEM‐guided protocol, was associated with reduced transfusion requirements, decreased device indwelling time, fewer high‐dependency admissions, and a shorter duration of hospital stay compared to uterine balloon tamponade.

Abbreviations

EUA

Examination under anesthesia

FFP

Fresh frozen plasma

FIGO

International Federation of Gynecology and Obstetrics

HD

High dependency

ICM

International Confederation of Midwives

ICU

Intensive care unit

LMIC

Low‐ and middle‐income countries

PPH

Postpartum hemorrhage

PRBC

Packed red blood cells

ROTEM

Rotational Thromboelastometry

UBT

Uterine balloon tamponade device

VHD

Vacuum‐assisted hemorrhage‐control device

WHO

World Health Organization

Key message.

Every minute counts when managing postpartum bleeding. Using vacuum‐assisted hemorrhage‐control devices together with E‐MOTIVE and ROTEM protocols can help resolve bleeding quickly, decrease the need for transfusions, and support a faster recovery.

1. INTRODUCTION

Postpartum hemorrhage (PPH) remains the leading cause of maternal mortality globally, with one mother dying of PPH approximately every six minutes. Although most births and maternal deaths occur in low‐ and middle‐income countries (LMICs), ¹ , ² data on PPH characteristics, management, and outcomes in Asian populations remain sparse, particularly among the region's highly heterogeneous communities spanning East, South, and Southeast Asia. ³ , ⁴ Population‐specific differences in baseline hemostatic profiles may also influence the evolution of hemorrhage‐specific associated coagulopathy, further underscoring the importance of region‐specific data. ⁵ For every woman who dies from PPH, an estimated 200 experience severe morbidity. ¹ Structured bundle approaches, such as the World Health Organization's E‐MOTIVE protocol, have demonstrated that timely, coordinated, and physiologically guided responses to hemorrhage can significantly reduce morbidity and mortality. ⁶ Nonetheless, in up to 20% of PPH, bleeding remains refractory despite prompt use of medical management. ⁷ , ⁸

When pharmacologic and supportive measures fail, mechanical interventions—uterine balloon tamponade (UBT) and vacuum‐assisted hemorrhage‐control devices (VHD)—are often employed. ⁹ , ¹⁰ However, only a few previous studies have directly compared their effectiveness, all of which have reported the results without adjusting for mode of delivery, and none of them explicitly focused on cases of refractory PPH following E‐MOTIVE workflows and ROTEM‐guided transfusion algorithms, tools that now enable early recognition of coagulopathy and individualized, fibrinogen‐targeted correction. ¹¹

Given these gaps in the available evidence, the true comparative effectiveness of these two mechanical interventions in massive or refractory PPH remains unclear, particularly in the context of modern physiologically guided care frameworks. In this study, we sought to examine whether women with refractory PPH unresponsive to standardized E‐MOTIVE‐based medical therapy experienced comparable outcomes when managed with UBT or VHD.

2. MATERIAL AND METHODS

2.1. Study design and data collection

We conducted a retrospective cohort study between January 2020 and September 2025 at the National University Hospital, Singapore, which sees about 5000 deliveries annually. The inclusion criteria were cases of primary PPH, defined as visual estimated blood loss >500 mL from the genital tract during the third stage of labor or within 24 h of delivery, in which the attending clinician determined that intrauterine tamponade device placement was necessary for hemostatic control. The exclusion criteria were cases of planned cesarean hysterectomy for placenta accreta spectrum and secondary PPH. Clinical data, including maternal demographics, obstetric characteristics, and PPH‐related outcomes, were collected from patients' electronic medical records.

2.2. Mechanical devices

The UBT device (Bakri balloon; Cook Medical, Bloomington, IN, USA) has been used at our institution since 2013 for PPH management. It is inserted transvaginally under ultrasound guidance and inflated with 300–400 mL of sterile water, with the drainage port connected to a graduated collection bag to monitor output. The balloon is left in situ for up to 24 h, after which it is deflated in a stepwise manner over 6 h. Following deflation, the patient is monitored for a further 6 h before the device is removed completely, provided there is no recurrent bleeding.

In 2020, the US Food and Drug Administration cleared an intrauterine VHD (JADA® System; Laborie, Portsmouth, NH, USA) to control abnormal postpartum uterine bleeding or hemorrhage when conservative management is warranted. Vacuum‐induced hemorrhage‐control devices were introduced in Singapore in October 2023, following regulatory approval. Multiple simulation training sessions were conducted to familiarize healthcare providers with the device before practical implementation. The device is inserted transvaginally into the uterine cavity, and the cervical seal, positioned at the external cervical os, is filled with up to 60 mL of sterile water to establish a seal. The device is then connected via suction tubing to a vacuum canister delivering a negative pressure of 80 mmHg. Active suction is maintained for 1 h. After clinical review confirms no active bleeding, suction is discontinued, and the device is left in situ for 30 minutes. The JADA System is completely removed once hemostasis is sustained, within a maximum of 24 h from insertion, by gently withdrawing the device.

Choice of device during PPH was based on the individual healthcare provider's clinical judgment and preferences, with our PPH management protocol listing both devices as equally viable options for mechanical management of refractory PPH. Figure 1 shows the distribution of use of UBT and VHD throughout the study period.

FIGURE 1.

Number of intrauterine tamponade device used for PPH between 2020 and 2025, along with rates of Postpartum Hemorrhage more than 500 mL (%).

2.3. PPH management protocol

The PPH management protocol at our institute was reviewed and introduced in September 2023. The revision adheres to the WHO's E‐MOTIVE bundle and aims for early detection, quantitative blood loss assessment, administration of oxytocin and tranexamic acid as first‐line therapy, uterine massage, and early escalation. At the same time, we adopted ROTEM‐guided transfusion to enable early, individualized, goal‐directed correction of coagulopathy (Figure 2).

FIGURE 2.

Standard Operating Protocol (SOP) for Management of Postpartum Hemorrhage at the National University Hospital, Singapore.

2.4. Statistical analysis

Continuous variables were summarized as medians with interquartile ranges (IQR), and categorical variables were presented as frequencies with percentages. Group comparisons (UBT vs. VHD) were performed using the Mann‐Whitney U test for continuous variables and the chi‐square test or Fisher's exact test for categorical variables. A composite adverse outcome was defined as the occurrence of any one of the following: transfusion of >4 units of packed red blood cells (PRBC), transfusion of fresh frozen plasma (FFP), transfusion of cryoprecipitate, need for examination under anesthesia (EUA), need for higher acuity care such as surgical high dependency (HD) or intensive care unit (ICU) admission, device expulsion or device failure which was defined as continued or recurrent bleeding after device placement requiring additional interventions such as examination under anesthesia, uterine artery embolization, or hysterectomy. Logistic regression was used to estimate the odds of the composite outcome with UBT vs. VHD use, adjusting for placenta previa major and cesarean delivery. Analyses were conducted using IBM SPSS Statistics for Windows, Version 29.0 (IBM Corp., Armonk, NY, USA). A p‐value of 0.05 was considered statistically significant.

3. RESULTS

3.1. Study population and baseline characteristics

A total of 59 women with refractory atonic primary PPH that were managed either with UBT or VHD were identified during the study period. Six cases were excluded from the analysis for the following reasons: two cases due to the inability to insert UBT and cessation of bleeding with conservative management; one case involved the use of UBT in first trimester incomplete miscarriage; one case involved the use of UBT in secondary PPH; and the last case involved a device used in an unrecognized uterine rupture.

In total, 53 cases were analyzed—33 managed with the UBT and 20 with VHD. Figure 1 shows the utilization of intrauterine tamponade devices during the study period.

Baseline maternal characteristics were similar between groups with respect to median age, pre‐pregnancy weight, and multiparity. (Table 1) However, placenta previa major was significantly more frequent in the UBT group compared with the VHD group (45.5% vs. 5.0%), and cesarean delivery was more common among women managed with the UBT balloon (78.8% vs. 50.0%). The severity of PPH was comparable with median estimated blood loss of 1500 mL at the time of device placement of in both groups.

TABLE 1.

Comparison of demographics, obstetric characteristics, and delivery outcomes between women who received the UBT and VHD for PPH.

Variables	UBT (n = 33)	VHD (n = 20)	p‐value
Maternal Characteristics
Age, years, median (range)	33 (30–37)	31 (29–37)	0.44 (Z‐score = 0.78)
Weight, kg, median (range)	65.8 (56.2–74.3)	75.8 (62.9–89.3)	0.12 (Z‐score = −1.61)
Parity, n (%)
Primiparous	12 (36.4)	8 (40)	0.79
Multiparous	21 (63.6)	12 (60)
Obstetric Characteristics
Multiple pregnancy, n (%)	2 (6.1)	2 (10.0)	0.63
Placenta previa, n (%)	15 (45.5)	1 (5.0)	0.002*
Placental abruption, n (%)	2 (6.1)	2 (10.0)	0.63
Mode of delivery, n (%)
Vaginal	7 (21.2)	10 (50.0)	0.03
Cesarean	26 (78.8)	10 (50.0)
Delivery Outcomes
Primary Cause of PPH, n (%)
Atony	12 (36.4)	19 (95.0)	0.0001* (Atony vs. non‐atony)
Retained placenta	3 (9.1)	0
Placenta previa	15 (45.5)	1 (5.0)
Focal placenta Accreta	2 (6.1)	0
Placenta inversion	1 (3.0)	0
Estimated blood loss, milliliters (range)	1500 (1000–2200)	1500 (1000–1813)	0.41 (z‐score 0.83)
Blood loss ≥1500 ml, n (%)	20 (60.6)	11 (55.0)	0.69

Abbreviations: PPH, postpartum hemorrhage; UBT, uterine balloon tamponade; VHD, vacuum‐induced hemorrhage‐control device.

^* p‐value <0.05 is significant.

3.2. Clinical outcomes

The incidence of the composite outcome was significantly lower in the VHD group compared with the UBT group (45.0% (9/20) versus 75.8% (25/33), unadjusted OR 0.26, 95% CI 0.08–0.86, p = 0.03). In a multiple logistic regression model adjusting for placenta previa major and mode of delivery, this difference was attenuated and no longer statistically significant (adjusted OR 0.38; 95% CI, 0.10–1.41; p = 0.15) (Table 2). Considering the individual components and secondary outcomes contributing to this composite endpoint: The decrease in hemoglobin from admission to nadir post‐delivery did not differ significantly between groups (2.8 g/dL vs. 2.0 g/dL, p = 0.15), but women managed with VHD required significantly fewer transfusions. The proportion requiring ≥4 units of PRBCs was significantly lower (5.0% vs. 36.4%, p = 0.02) as was the need for fresh frozen plasma (5.0% vs. 36.4%, p = 0.02) (Table 2).

TABLE 2.

Comparison of clinical outcomes between women who received UBT and VHD.

Clinical Outcomes	UBT (n = 33)	VHD (n = 20)	Unadjusted OR (95% CI)	Adjusted OR (95% CI)	p‐value
Composite Outcomes, n (%)	25 (75.8)	9 (45.0)	0.26 (0.08–0.86)	0.38 (0.10–1.41)	0.15
Number of PRBC transfused, n	2 (2–4)	1 (0–2)	‐	‐	0.009* (Z‐score 2.63)
Need for PRBC >1, n (%)	27 (81.1)	13 (65.0)	‐	‐	0.17
Need for PRBC ≥4, n (%)	12 (36.4)	1 (5.0)	0.09 (0.01–0.78)	0.15 (0.01–0.63)	0.02*
Need for FFP, n (%)	12 (36.4)	1 (5.0)	0.09 (0.01–0.78)	0.12 (0.01–0.63)	0.02*
Need for Cryoprecipitate, n (%)	4 (12.1)	0 (0)	Not estimable #		0.28
Hb difference between pre‐delivery and lowest recorded post‐delivery in g/dL (range)	2.8 (1.4–4.6)	2.05 (1.78–2.88)	‐	‐	0.15 (Z‐score 1.45)
Need for EUA and laparotomy, n (%)	4 (12.1)	1 (5.0)	‐	‐	0.64
Need for hysterectomy	0	0	‐	‐
Means indwelling time of device, Hours (range)	24.3 (7–38)	10.1 (2.5–24.5)	‐	‐	<0.001*
Need for HD/ICU, n (%)	24 (72.7)	9 (45.0) All SHD no ICU*	‐	‐	0.04*
Hospital length of stay post‐delivery, hours (range)	72 (72–96)	54 (48–72)	‐	‐	0.007* (Z‐score 2.71)
Device expulsion or failure, n (%)	2 (6.1%)	0 (0)	‐	‐	0.28

Abbreviations: EUA, examination under anesthesia; FFP, fresh frozen plasma; Hb, hemoglobin; HD, high dependency; ICU, intensive care unit; PRBC, packed red blood cells; UBT, uterine balloon tamponade; VHD, vacuum‐induced hemorrhage‐control device.

^# 0 events in one cell‐ OR are not estimable by standard methods without a continuity correction.

^* p‐value <0.05 is significant.

[Correction added on 14 August 2026, after first online publication: The p‐value of Composite outcomes, n(%) in the table 2 has been corrected from 0.03 to 0.15 in this version]

Regarding inpatient care, fewer women managed with VHD required SHD or ICU admission (45.5% versus 72.7%, p = 0.04). The median hospital length of stay was shorter with VHD (54 versus 72 h, p = 0.007) and VHD had a shorter mean indwelling time (10.1 h, range 2.5–24.5) compared with the UBT (24.3 h, range 7–38; p < 0.001).

There were two cases (6.1%) of device expulsion or failure in the UBT group. In the first case, the balloon was incidentally found in the vagina within 24 h of insertion. Bleeding had ceased and no further intervention was required. The second case, a 40‐year‐old multiparous woman with atonic PPH following spontaneous vaginal delivery, experienced recurrent bleeding after Bakri balloon expulsion into the cervical canal 24‐h post‐insertion. Uterine artery embolization was subsequently performed with successful hemostasis and she was discharged on postnatal day six.

4. DISCUSSION

This retrospective cohort study compared the clinical outcomes of UBT and VHD for the management of refractory atonic PPH within a tertiary obstetric unit operating a structured, protocol‐driven care pathway. After adjusting for placenta previa major and cesarean delivery, the composite primary morbidity outcome did not differ significantly between devices (adjusted OR 0.38, 95% CI 0.10–1.41, p = 0.15). Nevertheless, in unadjusted and secondary analyses, VHD use was associated with lower transfusion requirements, shorter device indwelling time, reduced need for high‐dependency care, and shorter hospitalization. These associations should be interpreted as hypothesis‐generating rather than definitive, given the observational design and confounding from case‐mix differences. Our findings suggest that both devices are effective for the management of PPH, while VHD may offer advantages in selected secondary clinical outcomes.

A key strength of this study, and arguably its most clinically relevant contribution, is that both devices were evaluated within a standardized, physiology‐guided PPH pathway incorporating the WHO E‐MOTIVE bundle together with ROTEM‐guided transfusion, the first such comparison, to our knowledge, outside a North American setting. Within this framework, the VHD use was associated with a notable reduction in plasma product utilization, including FFP (5.0% versus 36.4%) and cryoprecipitate (0% versus 12.1%), consistent with the goal‐directed, fibrinogen‐targeted correction that ROTEM guidance enables. This suggests that with protocolized, physiology‐guided care and device selection strategy VHD could be integrated into the care pathways without compromising outcomes.

Our findings of reduced transfusion requirements with VHD are consistent with Gulersen et al., ¹² who reported a nearly 90% reduction in massive transfusion rates with VHD use, but contrast with Chan et al. ¹³ and Shields et al., ¹⁴ who reported comparable success between devices. Variations in study design, population characteristics, and timing of device placement, as well as differences in institutional PPH protocols—particularly the absence of E‐MOTIVE and ROTEM‐guided adjuncts—likely account for these discrepancies. A detailed comparison of study designs, inclusion criteria, and outcomes across these four published studies and the current cohort is provided in Table S1. Prior studies were also conducted in North American populations within limited Asian representation (6.8%–13.2%), whereas our cohort represents, to our knowledge, the first Pan‐Asian comparative dataset for these devices.

An important contextual factor in interpreting our findings relates to the higher rate of placenta previa major in the UBT group, likely reflecting pragmatic operator preference for the balloon's outward hydrostatic effect on the lower uterine segment in placental‐site bleeding, rather than any protocol‐driven preference. ¹⁵ This case‐mix imbalance likely accounts for at least part of the higher transfusion burden observed in the UBT group and underscores the importance of the adjusted analysis. Indeed, major trials such as the RUBY and PEARLE underrepresented cases of placenta previa (4.5% of cases), highlighting the need for future studies to specifically evaluate vacuum‐based devices within this subgroup.

Device expulsion is a recognized limitation of UBT, we observed two cases (6.1%), one of which required escalation to uterine artery embolization. This rate is comparable to that reported in other studies, such as Chan et al., who observed balloon expulsion in 8.5% (11/130) of cases, ¹³ and Cho et al., who reported a rate of 4.7% among cases used for placenta previa major. ¹⁶ Reviews similarly describe overall complication frequencies of UBT as generally low, but emphasize the risk of expulsion. ¹⁷ Vacuum‐based devices may be less prone to expulsion, as negative pressure promotes apposition of the uterine walls around the catheter. Prospective studies specifically comparing device stability (expulsion and migration over time) are limited and represent a practical research gap.

Consistent with data from the RUBY trial and Shields et al., which found tamponade devices less effective when placed after substantial blood loss, more than half of our cohort had already lost >1500 mL at device placement, suggesting an opportunity for earlier mechanical intervention within E‐MOTIVE‐guided workflows to reduce downstream transfusion and critical care needs.

Limitations of this study include a single‐center, retrospective design, which introduces potential for selection and temporal bias related to changes and updates in the institutional PPH management protocol which coincided with the introduction of VHD at our institution. While the protocol does not recommend one device over another, the VHD was newly introduced during the study period, and its adoption curve, coupled with growing operator familiarity, may have influenced which patients received it. In future, a comparison of outcomes in the earlier versus later halves of the study period with an extended dataset may help clarify whether temporal changes in practice, independent of device choice, contributed to observed differences. While we adjusted for major confounders such as placenta previa and mode of delivery, unmeasured factors, including clinician experience and evolving transfusion thresholds, may have influenced outcomes. An additional limitation pertains to visual estimation of blood loss, which is known to be less accurate than quantitative blood loss measurement methods and may underestimate true hemorrhage volume. However, visual estimation was consistently used across all providers as well as throughout the study period, which mitigates differential interobserver bias between study groups. Lastly, the modest sample size limits the power to detect subtle differences between groups. Despite these constraints, this study contributes meaningfully to the global evidence base by providing context‐specific real‐world data.

5. CONCLUSION

In conclusion, the VHD was associated with clinical outcomes comparable to the UBT and had more favorable transfusion requirements, shorter device indwelling time, and reduced hospital stay, when used for refractory primary PPH within a structured, protocol‐driven care pathway. Both devices remain effective options for hemorrhage unresponsive to first‐line medical therapy, and their use in practice should be guided by clinical context, availability, cost considerations, and clinician familiarity.

AUTHOR CONTRIBUTIONS

Kelvin ZX Lee and Calista LS Por—conducted a comprehensive literature review; performed critical analysis and interpretation of study findings in the context of existing evidence; drafted, structured, and formatted the manuscript. Sarah WL Li—conceived and designed the study; developed the methodology, protocol, and analytical framework; oversaw data interpretation; led the ethics approval process; provided overall oversight of study conduct; and critically reviewed, revised, and approved the final manuscript. Tak Yein Lim; Pradip Dashraath—provided clinical input into the design and optimization of the postpartum hemorrhage management protocol, including structured implementation of uterine balloon tamponade, vacuum‐induced hemorrhage‐control devices, and ROTEM‐guided hemostatic strategies, as well as reviewed the manuscript. Liang Shen—contributed to the statistical analysis of the collected data and interpretation of study findings. Harshaana Ramlal—coordinated the conduct of the study; supported the ethics submission process; facilitated data collection; ensured data accuracy and integrity; and prepared figures for the manuscript. Wei Wei Teo, Christopher King, and Bryan SW Ng—contributed expert input to the development of postpartum hemorrhage management algorithms, with particular focus on ROTEM‐guided protocols and their integration into clinical decision pathways, as well as reviewed the manuscript. Mahesh Choolani, Ganesh Acharya, Arijit Biswas, and Sabaratnam Arulkumaran—provided supervision and mentorship for the study; contributed to the conceptual framing and strategic direction of the research; and critically reviewed and revised the manuscript.

FUNDING STATEMENT

This work received no financial support.

CONFLICT OF INTEREST STATEMENT

Dr. Sarah W.L. LI and Dr. Tak Yein LIM have previously received honoraria from Organon for invited lectures related to postpartum hemorrhage management and the Jada System. Since then, The JADA System was acquired by Laborie Medical Technologies. The current study was conceived, conducted, and written independently of Organon or Laborie without external funding, sponsorship, or material support. All devices used in this study were procured through standard hospital supply channels or paid for by patients as part of routine clinical care. All other authors declare no other conflicts of interest.

ETHIC STATEMENT

The National Healthcare Group Domain‐Specific Review Board (NHG DSRB Ref: 2025/1029) approved this study on 29 August 2025, with a waiver of informed consent. Patients were not directly involved in the design or conduct of this study.

Supporting information

Table S1. Comparison of Studies on Uterine Balloon vs. Vacuum‐Induced Tamponade Devices.

DATA AVAILABILITY STATEMENT

The datasets generated or analyzed during this study are not publicly available due to ethical restrictions but are available from the corresponding author on reasonable request.