Evidence summary

Population and studies description

This interventional procedures overview is based on about 3,000 people included in studies on balloon disimpaction, from 1 systematic review (Cornthwaite 2024), 5 retrospective cohort studies (Sadler 2024, Sacre 2021, Hanley 2020, Chooi 2022, Safa 2016), 2 randomised controlled trials (Lassey 2020, Dutta 2019), 1 prospective non-randomised comparative study (Barman 2015) and 2 case reports (Jordan 2022, Mumtaz 2023). Both the randomised controlled trials and all the cohort studies except 1 (Sadler 2024) were also included in the systematic review. Of the 3,000 people included in the studies, 986 had the procedure. This is a rapid review of the literature, and a flow chart of the complete selection process is shown in figure 1. This overview presents 11 studies as the key evidence in table 2 and table 3, and lists 7 other relevant studies in appendix B, table 5.

Studies were from Australia, India, New Zealand, UK and US. The systematic review included 24 studies, 7 of which used balloon disimpaction (Cornthwaite 2024). Of the 7 studies, 6 have been included in the key evidence tables to give details of all the reported outcomes. The remaining study was a non-randomised comparative study that has been retracted since the systematic review was prepared. The systematic review assessed certainty of evidence across outcomes using GRADE and avoided pooling data for meta-analysis when there was high heterogeneity. It excluded a randomised controlled trial that had been retracted because of concerns about scientific integrity. It used a sensitivity analysis that excluded the non-randomised study that has also since been retracted. According to GRADE criteria, the certainty of evidence was very low.

A recently published large retrospective cohort study in New Zealand compared outcomes within and between 2 centres (Sadler 2024). In 1 centre, outcomes were compared before and after the balloon disimpaction device was introduced and outcomes were compared with and without its use. The outcomes in this centre were then compared with the outcomes in a centre that didn't have the device available. The primary outcome was uterine incision extension. There were statistically significant differences in demographic and clinical characteristics between the comparison groups, including parity, fetal head station and attempted assisted vaginal birth, suggesting that balloon disimpaction was used for more difficult deliveries. There were statistically significant differences in maternal ethnicity between the comparison groups. In the unit where the balloon disimpaction device was available, there was a higher proportion of people of Indian ethnicity (33%) and a lower proportion of people of Maori ethnicity (11%) in the group where the device was used compared to the group in which the device was not used (26% Indian and 16% Maori). When the 2 units were compared, the unit with the balloon device had a higher proportion of people with a specific ethnicity recorded (14% Maori, 25% Pacific, 29% Indian, 11% other Asian and 22% 'Other') compared with the unit without the balloon device (5% Maori, 6% Pacific, 16% Indian, 25% other Asian, and 48% 'Other').

The 2 randomised controlled trials were both small and 1 was published in a journal that was identified in the systematic review as being 'potentially predatory' (Dutta 2019). This trial was based in India and compared balloon disimpaction (n=25) with the modified Patwardhan technique (n=25). Although the paper describes the trial as a prospective comparative trial rather than a randomised controlled trial in the title and methods, it does describe a process of randomising treatment allocation. The primary outcome measure was not specified. The trial by Lassey et al. (2020) was based in the US and compared outcomes in nulliparous women having caesarean delivery in the second stage of labour using an inflated balloon disimpaction device (n=30) with those using a non-inflated device (n=30). The study assessed the use of the device in the second stage of labour and did not specifically address its use with impacted fetal head. The original trial description stated the estimated enrolment was 200 but the actual enrolment was 60. The primary outcome measure was time (in seconds) from hysterotomy to delivery of the neonate.

The remaining 4 retrospective cohort studies were based in the UK and Australia. The study by Sacre et al. (2021) included 391 caesarean births, 170 (44%) of which used balloon disimpaction based on the individual clinician's decision. There were no statistically significant differences between the groups in maternal age, body mass index, ethnicity, onset of labour, or fetal position. Most of the women (87%) were described as Caucasian. The proportion of nulliparous women was higher in the balloon disimpaction group as was the proportion of fetal head station reported below the ischial spines. There was a higher number of deliveries for fetal distress in the group without balloon disimpaction. The study by Hanley et al. (2020) included 174 caesarean births, 114 (66%) of which used balloon disimpaction. There were no statistically significant differences in baseline characteristics between the groups. The primary outcome of interest was operative complications, which included uterine incision extension, need for breech extraction, use of vertical incision and bladder, ureter or bowel injury. The study by Chooi et al. (2023) included all caesarean births where balloon disimpaction was used (n=53) and randomly selected a similar number of caesarean births without balloon disimpaction (n=48). Maternal and neonatal baseline characteristics were similar between the 2 groups. The study by Safa et al. (2016) compared all caesarean births done at full dilation using balloon disimpaction with those using the hand-push technique to disimpact the fetal head. Maternal baseline characteristics were similar between the groups.

The prospective non-randomised comparative study by Barman et al. (2015) was not included in the systematic review by Cornthwaite et al. (2024). It compared 75 caesarean deliveries using balloon disimpaction with 72 deliveries using the Patwardhan technique. Assignment to each group was done alternately. Maternal baseline characteristics were similar between the groups.

The case report by Jordan et al. (2022) describes uterine and bladder rupture when balloon disimpaction was used at a full dilation caesarean in a woman who had had a previous caesarean delivery at full dilation for obstructed labour. The case report by Mumtaz et al. (2023) describes 2 cases of posterior uterine wall rupture identified at caesarean immediately after balloon disimpaction was used.

Table 2 presents study details.

Figure 1 Flow chart of study selection

**Table 2 Study details**
Study no.	First author, date country	Characteristics of people in the study (as reported by the study)	Study design	Inclusion criteria	Intervention	Follow up
1	Cornthwaite K, 2024 Studies on balloon disimpaction were from Australia, India, UK and US	n=3,558 (24 studies) Of the 7 studies on balloon disimpaction (n=1,110), 6 included women with 10 cm cervical dilation and 1 study included women with cervical dilation 7 cm or more.	Systematic review, including 7 studies on balloon disimpaction. Of the 7 studies, 2 were randomised controlled trials and 5 were non-randomised comparative studies. The risk of bias was low in 1 of the randomised controlled trials and high in the other. All non-randomised studies were at serious or critical risk of bias. The certainty of evidence using GRADE was low or very low for all outcomes.	Studies comparing techniques or adjunctive measures to prevent or manage impacted fetal head at emergency caesarean birth (first or second stage). All randomised controlled trials of any size and non-randomised comparative prospective or retrospective cohort studies with 30 or more people per treatment arm were included. A trial that was retracted because of concerns about scientific integrity was excluded. Another trial with similar authors, data and methods was considered to be at risk of retraction and was excluded during sensitivity analysis.	The Fetal Pillow device was used in the intervention arm of all 7 studies on balloon disimpaction (n=533). The comparators were no pillow (n=453), non-inflated pillow (n=30), vaginal disimpaction (n=69) or Patwardhan technique (n=25).	Not reported
2	Sadler L, 2024 New Zealand	n=1,703 caesareans in second stage (375 with balloon disimpaction) When comparing use and non-use of balloon disimpaction, there was a higher proportion of use in nulliparous compared to parous women (82% versus 73%) and when the fetal head was at or below the ischial spines (station 0 to +3) (57% versus 44%), and after attempted operative vaginal birth (14% versus 7%).	Retrospective cohort (2 centres), comparing outcomes with or without balloon disimpaction. There were 3 comparisons: 2 were within a centre, before and after introduction of balloon device and with or without use of balloon device. The third compared the centre that used the device with a centre that did not use it. Study period: 2016 to 2021	Singleton, cephalic, term (37 weeks or more) pregnancies with caesarean in second stage. Cases were ascertained from maternity electronic records at 2 tertiary maternity units.	Second stage caesarean deliveries with (n=375) or without (n=1,328) balloon disimpaction using the Fetal Pillow device.	6 weeks
3	Lassey S, 2020 US	n=60 (30 with balloon disimpaction) nulliparous women who had caesarean delivery in the second stage of labour Median maternal age (years): Inflated group=33.0 Non-inflated group=30.5	Double blind randomised controlled trial Analysis was by intention to treat. Recruitment period: 2018 to 2019	At least 18 years old with a full term (37 weeks of gestation or greater) singleton fetus in cephalic presentation. Women were excluded if there was a contraindication to a vaginal delivery, prior caesarean delivery, or presence of congenital fetal anomaly. Non-English speaking women were also excluded.	The Fetal Pillow device (Safe Obstetric Systems UK) was used in all deliveries. It was inflated in the treatment group (n=30) and not inflated in the control group (n=30). Most women had a low transverse caesarean delivery (93%).	To hospital discharge
4	Dutta S, 2019 India	n=50 (25 with balloon disimpaction) Full dilation Mean age=25 years 60% of whole cohort were nulliparous. 46% were at fetal station 2 and 30% were at fetal station 1.	Randomised controlled trial Study period is unclear: most of the text refers to a period between 2016 and 2017 but some text refers to 2014. The abstract refers to the centre as West Bengal but the rest of the paper states the study was done in North Bengal.	Women with singleton pregnancies at least 36 weeks gestation having caesarean at full dilation, with a deeply engaged fetal head. Patients with active genital infections, fetal malpresentation, non-cephalic presentation were excluded.	The Fetal Pillow device (Safe Obstetrics Systems UK) was used in 25 (50%) deliveries. The comparator was the modified Patwardhan technique (n=25).	To hospital discharge
5	Sacre H, 2021 UK	n=391 (170 with balloon disimpaction) Full dilation Median maternal age=30 years 87% Caucasian 87% of those in the balloon disimpaction group were nulliparous compared with 73% in the control group (p=0.0009). There were statistically significant differences in station with fewer people in the balloon disimpaction group reported as above the ischial spines (RR 0.37, 95% CI 0.25 to 0.57, p<0.0001) and more below the ischial spines (RR 2.34, 95% CI 1.68 to 3.26, p<0.0001).	Retrospective cohort study Study period: 2014 to 2018	Full dilation caesareans. Twin pregnancies and breech presentations were excluded. All outcomes were obtained from the hospital electronic patient record systems Meditech and neonatal outcomes were obtained from Badger.net.	The Fetal Pillow device (Safe Obstetrics Systems UK) was used in 170 (44%) deliveries. The decision to use the device or not was left to the individual clinician.	To hospital discharge
6	Hanley I, 2020 Australia	n=174 (114 with balloon disimpaction) Full dilation Mean maternal age=28.0 years (balloon disimpaction), 26.6 years (control), p=0.1 Mean parity=0.41 (balloon disimpaction), 0.47 (control), p=0.35 Most were at fetal station 0 or 1.	Retrospective cohort study Study period: 2014 to 2018	All patients who had a caesarean delivery at full dilation or with the use of a balloon disimpaction device pillow were reviewed for this study. Any cases that used balloon disimpaction at less than full dilation were excluded. The study data were extracted from the study institution's electronic record system.	The Fetal Pillow device (Safe Obstetrics Systems UK) was used in 114 (66%) deliveries. Use of the device depended on obstetrician preference or situation‐specific factors (such as urgency of delivery).	To hospital discharge
7	Chooi K, 2023 Australia	n=101 (53 with balloon disimpaction) Cervical dilation 7 cm or above Mean maternal age=30.4 years 76% of whole cohort were nulliparous.	Retrospective cohort study Study period: 2018 to 2019 All caesarean births where balloon disimpaction was used and a similar number of births where caesarean birth occurred at full dilatation and balloon disimpaction was not used were included. These were randomly selected from the hospital database as they were coded as 'emergency caesarean' and 'full dilatation' across the same time period.	Women who had a singleton pregnancy requiring caesarean birth at term gestation in the second stage of labour. Women were excluded if they were identified as fully dilated on coding but their medical records documented that they were less than 7 cm dilated. Women who were 7 cm dilated and above, but not fully dilated, remained in the study.	The Fetal Pillow device (Safe Obstetrics Systems UK) was used in 53 (53%) deliveries.	To hospital discharge
8	Safa H, 2016 Australia	n=160 (91 with balloon disimpaction) Full dilation Mean maternal age (years): Balloon disimpaction=29.9 Hand-push method=31, p=0.39 75% of whole cohort were nulliparous. Fetal station was not reported.	Retrospective cohort study Study period: 2013 to 2015	Women who had singleton pregnancies and had caesarean delivery at full dilation at 37 week of pregnancy or later during the study period. Only deliveries where balloon disimpaction or the hand-push method were used were included in the analyses. Multiple pregnancies, pregnancies that resulted in fetal death in utero, and any major fetal congenital anomalies were excluded.	The Fetal Pillow device (Safe Obstetrics Systems UK) was used in 91 (57%) deliveries. The comparator was the hand-push technique (n=69).	To hospital discharge
9	Barman S, 2015 India	n=147 (75 with balloon disimpaction) Mean maternal age (years): Balloon disimpaction=21.9 Patwardhan=22.4 Station of head +1 or below: Balloon disimpaction=45% Patwardhan=54%	Non-randomised comparative study Women were alternately allocated to study or control group. Study period: 2012 to 2013	Women with single fetus with cephalic presentation at late stage of labour when head was deep in pelvis and needing emergency caesarean.	Fetal pillow (Safe Obstetric Systems UK), n=75 Patwardhan technique, n=72	Not reported
10	Jordan A, 2022 Australia	n=1 Full dilation 32-year old (parity 1) Fetal head was at station 1	Case report	Case report: uterine and bladder rupture	The Fetal Pillow device (Safe Obstetrics Systems UK) was used.	Not reported
11	Mumtaz H, 2023 UK	n=2 40-year old (parity 1) and 36-year old (parity 2) women	Case report	Case reports of posterior uterine rupture	The Fetal Pillow (Cooper Surgical) was used.	Not reported

**Table 3 Study outcomes**
First author, date	Maternal outcomes	Neonatal outcomes
Cornthwaite 2024	Pooled analysis that included balloon disimpaction device compared with no device, non-inflated device or vaginal disimpaction demonstrated no or equivocal differences in maternal outcomes based on very low or low certainty evidence. Uterine incision extension (balloon disimpaction compared with no balloon) RR=0.79 (95% CI 0.52 to 1.19); 4 studies, I²=31% After excluding a study at risk of retraction in sensitivity analysis, the RR was 0.90 (95% CI 0.67 to 1.23, I²=0%). GRADE assessment showed very low certainty for this outcome, with and without the study at risk of retraction. Postpartum haemorrhage (balloon disimpaction compared with no balloon) RR=1.10 (95% CI 0.81 to 1.49); 4 studies, I²=0% After excluding a study at risk of retraction in sensitivity analysis, the RR was 1.13 (95% CI 0.83 to 1.54, I²=0%). Operative time Although 1 non-randomised study reported a statistically significantly shorter operative time when using balloon disimpaction compared with no balloon (mean difference 20.81 [95% CI 22.31 to 19.31] minutes), it was at serious risk of bias with risk of retraction and very low certainty in the effect estimates.	Infant birth trauma (balloon disimpaction compared with no balloon) OR=1.02 (95% CI 0.46 to 2.24); 2 studies, I²=68% After excluding the study at risk of retraction in sensitivity analysis, the OR was 1.44 (95% CI 0.60 to 3.48) in the remaining study (Chooi 2022). GRADE assessment showed very low certainty for this outcome, with and without the study at risk of retraction. Apgar score less than 7 at 5 minutes (balloon disimpaction compared with no balloon) RR=1.01 (95% CI 0.84 to 1.21); 3 studies, I²=0%
Sadler, 2024	Comparison 1: after versus before introduction of balloon disimpaction device to centre Any uterine incision extension (primary outcome) After introduction of device=24.8% (219/883) Before introduction of device=26.8% (84/314) Adjusted OR=0.88 (95% CI 0.65 to 1.19) For nulliparous women, adjusted OR=1.13 (95% CI 0.78 to 1.62) For parous women, adjusted OR=0.47 (95% CI 0.26 to 0.82) Major uterine incision extension After introduction of device=7.7% (68/883) Before introduction of device=6.4% (20/314) Adjusted OR=1.20 (95% CI 0.71 to 2.04) Comparison 2: use versus non-use of balloon disimpaction device after introduction Any uterine incision extension (primary outcome) Device used=26.1% (98/375) Device not used=23.8% (121/508) Adjusted OR=1.14 (95% CI 0.83 to 1.57) Major uterine incision extension Device used=8.5% (32/375) Device not used=7.1% (36/508) Adjusted OR=1.35 (95% CI 0.80 to 2.27) Balloon disimpaction was used more frequently with nulliparous than parous women (82.4% versus 73.0%) and when the fetal head was at or below the ischial spines (station 0 to +3) (56.7% versus 43.9%), and after attempted operative vaginal birth (13.6% versus 6.9%). Comparison 3: unit with versus unit without availability of balloon disimpaction device Any uterine incision extension (primary outcome) Unit with device available=24.2% (139/574) Unit without device available=29.2% (148/506) Adjusted OR=0.73 (95% CI 0.54 to 0.99) Major uterine incision extension Unit with device available=7.1% (41/574) Unit without device available=6.5% (33/506) Adjusted OR=1.35 (95% CI 0.80 to 2.27) The following potential confounders, and predictors, of maternal and neonatal morbidity were included in multivariable models using logistic regression: maternal body mass index, ethnicity, parity, previous caesarean status, length of second stage, attempted assisted vaginal delivery, station of head at last vaginal assessment, urgency of caesarean, neonatal birthweight and gestational age.	Comparison 1: after versus before introduction of balloon disimpaction device to centre Composite neonatal outcome (including perinatal death, hypoxic ischaemic encephalopathy (Sarnat moderate or severe), any neonatal seizure, birth trauma, acidosis (cord pH less than 7.10) or neonatal intensive care unit admission for at least 48 hours) After introduction of device=4.3% (38/883) Before introduction of device=4.1% (13/314) Adjusted OR=1.07 (95% CI 0.55 to 2.06) Comparison 2: use versus non-use of balloon disimpaction device after introduction Composite neonatal outcome Device used=4.8% (18/375) Device not used=3.9% (20/508) Adjusted OR=1.01 (95% CI 0.51 to 1.99) Comparison 3: unit with versus unit without availability of balloon disimpaction device Composite neonatal outcome Unit with device available=4.7% (27/574) Unit without device available=6.5% (33/506) Adjusted OR=0.73 (95% CI 0.39 to 1.36)
Lassey, 2020	Median time from hysterotomy to delivery (primary outcome) Inflated device=31 seconds Non-inflated device (control)=54 seconds, p<0.01 Mean difference=-38.2 seconds (95% CI -56.1 to ‑20.3) Uterine incision extension Inflated device=20% (6/30) Non-inflated device (control)=43% (13/30), p=0.05 Of the 13 extensions in the non-inflated group, 4 were into the cervix, vagina or bladder compared with none in the inflated group. Median blood loss (ml) Inflated device=800 Non-inflated device (control)=900 Mean difference=-191.7, p=0.09 There were 3 blood transfusions in the non-inflated group and none in the inflated group. Maternal fever, intensive care unit admission, prolonged length of hospital stay and readmission were rare events and not different between groups. Ease of delivery Healthcare providers described 80% (24/30) of deliveries in the inflated group as 'easy' or 'very easy' compared with 43% (13/30) in the not-inflated group. None of the deliveries in the inflated group were described as 'difficult' or 'very difficult' compared with 37% (11/30) in the not-inflated group (p<0.01). 97% (29/30) of healthcare providers in the inflated group would use the balloon elevation device again or recommend it to others compared with 77% (22/30) in the not-inflated group (p=0.03).	There were no differences in neonatal outcomes between the groups, including birth weight, Apgar scores, neonatal intensive care, unit admission, intubation and other fetal morbidity. The median Apgar score at 1 minute was 8 in both groups. At 5 minutes, the median score was 9 in both groups.
Dutta 2019	Extension of uterine incision Balloon disimpaction=8% Modified Patwardhan technique=24%, p=0.001 Operating time (skin to skin), p=0.02 30 to 40 minutes Balloon disimpaction=80% Modified Patwardhan technique=64% 40 to 50 minutes Balloon disimpaction=12% Modified Patwardhan technique=28% 50 to 60 minutes Balloon disimpaction=8% Modified Patwardhan technique=4% More than 1 hour Balloon disimpaction=0% Modified Patwardhan technique=4% Uterine incision to delivery time, p=0.04 0 to 2 minutes Balloon disimpaction=52% Modified Patwardhan technique=8% 2 to 4 minutes Balloon disimpaction=48% Modified Patwardhan technique=44% 4 to 6 minutes Balloon disimpaction=0% Modified Patwardhan technique=48% Number of mops soaked in blood, p=0.02 1:balloon disimpaction=12%, modified Patwardhan technique=0% 2: balloon disimpaction=72%, modified Patwardhan technique=80% More than 2: balloon disimpaction=16%, modified Patwardhan technique=20% Need for blood transfusion Balloon disimpaction=0% Modified Patwardhan technique=16%, p=0.002	Admission to special newborn care unit Balloon disimpaction=12% Modified Patwardhan technique=60%, p=0.04 Out of the 15 babies admitted to the special newborn care unit in the modified Patwardhan group, 1 had bruises over the back and abdomen and died on day 11 (early neonatal death). Occurrence of neonatal seizure Balloon disimpaction=0% Modified Patwardhan technique=4%, p=0.03
Sacre, 2021	Median estimated blood loss, ml (interquartile range) Balloon disimpaction=600 (500 to 900) Control=600 (500 to 800) Estimated blood loss more than 1,000 ml Balloon disimpaction=22.9% (39/170) Control=18.6% (41/221) RR=1.24 (95% CI 0.84 to 1.83), p=0.29 Estimated blood loss more than 1,500 ml Balloon disimpaction=8.8% (15/170) Control=6.3% (14/221) RR=1.39 (95% CI 0.69 to 2.81), p=0.35 Blood transfusion Balloon disimpaction=5.4% (8/170) Control=4.1% (9/221) RR=1.16 (95% CI 0.46 to 2.9), p=0.76 Uterine incision extension Balloon disimpaction=21.8% (37/170) Control=21.3% (43/221) RR=1.02 (95% CI 0.70 to 1.50), p=0.91 Subgroup analysis: deliveries at the ischial spines or lower Median estimated blood loss, ml (interquartile range) Balloon disimpaction=600 (500 to 887.5) Control=600 (500 to 800) Estimated blood loss more than 1,000 ml Balloon disimpaction=23.0% (32/139) Control=14.2% (18/126) RR=1.61 (95% CI 0.95 to 2.72), p=0.07 Estimated blood loss more than 1,500 ml Balloon disimpaction=8.6% (12/139) Control=3.2% (4/126) RR=2.72 (95% CI 0.90 to 8.22), p=0.08 Blood transfusion Balloon disimpaction=5.6% (8/139) Control=3.2% (4/126) RR=1.81 (95% CI 0.56 to 5.88), p=0.32 Uterine incision extension Balloon disimpaction=21.6% (30/139) Control=24.6% (31/126) RR=0.87 (95% CI 0.56 to 1.36), p=0.56	Median birthweight, g (interquartile range) Balloon disimpaction=3,658 (3,350 to 3,958) Control=3,650 (3,328 to 4,030) Apgar score less than 7 at 5 minutes Balloon disimpaction=7.1% (12/170) Control=5.4% (12/221) RR 1.30 (95% CI 0.60 to 2.82), p=0.51 Arterial pH less than 7.1 Balloon disimpaction=7.1% (12/170) Control=13.1% (29/221) RR 0.54 (95% CI 0.28 to 1.02), p=0.06 Admission to neonatal intensive care unit Balloon disimpaction=8.8% (15/170) Control=12.2% (27/221) RR 0.72 (95% CI 0.40 to 1.31), p=0.29 Subgroup analysis: deliveries at the ischial spines or lower Apgar score less than 7 at 5 minutes Balloon disimpaction=7.2% (10/139) Control=5.6% (7/126) RR 1.30 (95% CI 0.51 to 3.30), p=0.59 Arterial pH less than 7.1 Balloon disimpaction=7.2% (10/139) Control=18.3% (23/126) RR 0.39 (95% CI 0.20 to 0.80), p=0.001 Admission to neonatal intensive care unit Balloon disimpaction=7.9% (11/139) Control=11.9% (15/126) RR 0.67 (95% CI 0.31 to 1.39), p=0.28
Hanley, 2020	Need to use 'hand push' technique Balloon disimpaction=0.9% (1/114) Control=10.0% (6/60), p=0.007 Operative complications (including uterine incision extension, need for breech extraction, use of vertical incision, and bladder, ureter or bowel injury) – primary outcome Balloon disimpaction=16.7% (19/114) Control=25.0% (15/60) OR=0.56 (95% CI 0.26 to 1.22), p=0.146 Mean estimated blood loss (ml) Balloon disimpaction=730 Control=725, p=0.98 Estimated blood loss more than 1 litre Balloon disimpaction=13.2% (15/114) Control=11.7% (7/60) OR=1.15 (95% CI 0.44 to 3.00), p=0.77 Need for blood transfusion Balloon disimpaction=0.9% (1/114) Control=5.2% (3/60) OR=0.22 (95% CI 0.03 to 1.64), p=0.14 Mean length of hospital stay Balloon disimpaction=65.8 hours Control=72.6 hours, p=0.71	Mean Apgar score at 5 minutes Balloon disimpaction=8.65 Control=8.63, p=0.87 Apgar score more than 7 at 5 minutes Balloon disimpaction=92.9% (105/113) Control=93.3% (56/60) OR=1.07 (95% CI 0.31 to 3.70), p=0.91 Neonatal intensive care unit stay Balloon disimpaction=33.0% (37/112) Control=40.0% (24/60) OR=0.76 (95% CI 0.39 to 1.47), p=0.41 Mean duration of stay in neonatal intensive care unit Balloon disimpaction=1 hour Control=1.5 hours, p=0.24 Mean umbilical cord arterial pH Balloon disimpaction=7.25 Control=7.19, p=0.0001
Chooi 2023	Uterine incision extension Balloon disimpaction=20.8% (11/53) Control=25.0% (12/48) RR=0.83 (95% CI 0.40 to 1.72), p=0.61 Blood transfusion Balloon disimpaction=3.8% (2/53) Control=2.1% (1/48) RR=1.81 (95% CI 0.18 to 38.3), p=0.62 Admission to high dependency unit Balloon disimpaction=11.3% (6/53) Control=10.4% (5/48) RR=1.09 (95% CI 0.35 to 3.56), p=0.88 Mean blood loss, ml Balloon disimpaction=699.1 (SD 393.9) Control=797.9 (SD 488.3) Mean difference=-98.9 (95% CI -273.3 to 75.6), p=0.26 Blood loss more than 1,000 ml Balloon disimpaction=26.4% (14/53) Control=29.2% (14/48) RR=0.91 (95% CI 0.48 to 1.72), p=0.76 Mean length of hospital stay, days Balloon disimpaction=3.22 (SD 1.4) Control=3.45 (SD 1.2) Mean difference=-0.23 (95% CI -0.75 to 0.29), p=0.39	Need for resuscitation Balloon disimpaction=28.3% (15/53) Control=25.5% (12/47) RR=1.11 (95% CI 0.58 to 2.18), p=0.76 Apgar score below 7 at 1 minute Balloon disimpaction=15.1% (8/53) Control=18.8% (9/48) RR=0.81 (95% CI 0.33 to 1.94), p=0.62 Apgar score below 7 at 5 minutes Balloon disimpaction=3.8% (2/53) Control=6.2% (3/48) RR=0.60 (95% CI 0.08 to 3.50), p=0.57 Mean cord arterial lactate Balloon disimpaction=4.19 (SD 1.76) Control=4.41 (SD 1.74) Mean difference=-0.22 (95% CI -0.98 to 0.54), p=0.56 Mean cord arterial pH Balloon disimpaction=7.25 (SD 0.06) Control=7.24 (SD 0.08) Mean difference=-0.00 (95% CI -0.03 to 0.03), p=0.92 Admission to special care baby unit or neonatal intensive care Balloon disimpaction=30.2% (16/53) Control=46.8% (22/47) RR=0.64 (95% CI 0.38 to 1.07), p=0.09 Jaundice Balloon disimpaction=28.3% (15/53) Control=42.6% (20/47) RR=0.67 (95% CI 0.38 to 1.14), p=0.14 Seizures or jitters Balloon disimpaction=0% (0/53) Control=8.5% (4/47), p=0.045 Neonatal trauma (soft tissue injury, bony trauma or intracranial bleed) Balloon disimpaction=30.2% (16/53) Control=23.4% (11/47) RR=1.29 (95% CI 0.67 to 2.59), p=0.45
Safa 2016	Mean estimated blood loss, ml Balloon disimpaction=273 (SD 145) Hand-push technique=403 (SD 199), p=0.026 Blood transfusion Balloon disimpaction=3% (3/91) Hand-push technique=3% (2/69), p=0.75 Uterine angle extension Balloon disimpaction=20% (18/91) Hand-push technique=35% (24/69), p=0.061 Mean length of hospital stay, hours Balloon disimpaction=77.9 (SD 19.6) Hand-push technique=97.8 (SD 27.6), p=0.002	Apgar score below 7 at 5 minutes Balloon disimpaction=3% (3/91) Hand-push technique=6% (4/69), p=0.44 Need for endotracheal intubation Balloon disimpaction=0% (0/91) Hand-push technique=3% (2/69), p=0.10 Admission to neonatal intensive care unit Balloon disimpaction=15% (14/91) Hand-push technique=25% (17/69), p=0.14 Mean cord arterial pH Balloon disimpaction=7.24 (SD 0.06) Hand-push technique=7.19 (SD 0.09), p=0.003
Barman 2015	'Easy' delivery of fetal head Balloon disimpaction=76.0% (57/75) Patwardhan technique=31.9% (23/72) OR=6.75 (95% CI 3.08 to 14.95) Mean baby delivery time (seconds) Balloon disimpaction=180 (SD 29) Patwardhan technique=270 (SD 50), p<0.0001 Mean total operative time (skin to skin; minutes) Balloon disimpaction=36.8 (SD 7) Patwardhan technique=40.2 (SD 7), p=0.004 Extension of uterine angle Balloon disimpaction=25.3% (19/75) Patwardhan technique=18.1% (13/72), p=0.285 Uterine artery tear Balloon disimpaction=20.0% (15/75) Patwardhan technique=9.7% (7/72), p=0.081 Need for blood transfusion Balloon disimpaction=5.3% (4/75) Patwardhan technique=6.9% (5/72)	Admission to neonatal intensive care unit Balloon disimpaction=17.3% (13/75) Patwardhan technique=33.3% (24/72), p=0.025 Mean Apgar score at 1 minute Balloon disimpaction=6.23 (SD 2.15) Patwardhan technique=3.93 (SD 2.13), p<0.0001 Mean Apgar score at 5 minutes Balloon disimpaction=7.79 (SD 2.03) Patwardhan technique=5.96 (SD 2.29), p<0.0001 2 babies in the Patwardhan technique group were stillborn.
Jordan 2022	Caesarean at full dilation, for obstructed labour with the fetal head at station 1. The woman had a previous caesarean delivery at full dilation for obstructed labour. As soon as the balloon disimpaction device was inserted frank blood was seen in the catheter and the urinary bag. Intraoperatively, there were extensive adhesions and a Bandl's ring. The lower segment of the uterus had ruptured into the bladder, with fetal parts present within the bladder cavity and abdomen. Surgical repair involved the urology team. Total maternal blood loss was 3.1 litres. After resuscitation, the immediate and long term recovery were uneventful. The authors noted that there was a weakened zone because of a previous caesarean section scar. This ruptured when elevation of the fetal head caused a shift in pressure to the weakest point.	A live male infant weighing 3,850 g was delivered in good condition (Apgar score 9).
Mumtaz 2023	Case 1: 40-year old woman of Asian origin with a history of caesarean section 7 years previously because of failure to progress. Labour progressed smoothly until full cervical dilation was reached. There were concerns about fetal distress during active pushing and it was decided to proceed with an emergency caesarean (station at 0 and left occiput transverse position). A balloon disimpaction device was inserted vaginally. Intraoperatively haemoperitoneum was noted. The anterior uterine wall was intact. The baby was delivered through a lower-segment transverse incision and was in good condition. A vertical full thickness uterine posterior wall rupture was noted, which was repaired with sutures. Total estimated blood loss was 1,000 ml. The immediate and long-term recovery were uneventful. Case 2: 36-year old woman of African origin with a history of 2 previous vaginal deliveries. After 2 hours of active pushing at full cervical dilation, the fetal head was in direct occiput anterior position, station at ischial spines and caput+2. A decision was made to proceed with an emergency caesarean section. A balloon disimpaction device was inserted vaginally. Intraoperatively blood-stained peritoneal fluid was observed on entering the abdominal cavity. The anterior uterine wall was intact. The baby was delivered through a lower-segment transverse incision, in good condition. A vertical partial thickness posterior wall dehiscence (intact serosal layer) was noted on the uterus, which was repaired with sutures. Total estimated blood loss was 1,300 ml. The immediate and long-term recovery were uneventful. The authors noted that it is uncertain whether the posterior uterine wall ruptures were caused by an increase in intra-uterine pressure secondary to the use of the balloon disimpaction device or because of the fragility of the uterus during labour.	Both babies were delivered in good condition (Apgar scores 8 and 10, and 9 and 10, respectively).

Procedure technique

All studies used the Fetal Pillow (Safe Obstetrics Systems UK, now CooperSurgical) for balloon disimpaction at caesarean during the second stage of labour or at full dilation (1 study included some women who were not fully dilated; Chooi 2023).

Maternal outcomes

Uterine incision extension

Uterine incision extension or uterine angle extension was reported as an outcome in 7 studies. In the systematic review by Cornthwaite et al. (2024), the RR for uterine incision extension when balloon disimpaction was used compared with no balloon was 0.90 (95% CI 0.67 to 1.23, I²=0%) based on 3 studies (Chooi 2023, Hanley 2020 and Sacre 2021).

The study by Hanley et al. (2020) was a retrospective cohort of 173 caesarean deliveries with or without balloon disimpaction and the primary outcome was a composite measure of 'operative complications', including uterine incision extension, need for breech extraction, use of vertical incision and bladder, ureter or bowel injury. The rates for this outcome were 17% (19/114) in the balloon disimpaction group and 25% (15/60) in the control group (OR=0.56, 95% CI 0.26 to 1.22, p=0.146; Hanley 2020).

In the cohort study of 1,703 caesareans, the rate of uterine incision extension was 25% (219/883) after the balloon disimpaction device was introduced to the maternity unit compared with 27% (84/314) before (adjusted OR=0.88, 95% CI 0.65 to 1.19; Sadler 2024). For nulliparous women, the adjusted OR was 1.13 (95% CI 0.78 to 1.62) and for nulliparous women it was 0.47 (95% CI 0.26 to 0.82). After the device had been introduced to the maternity unit, the rate of uterine incision extension was 26% (98/375) when the device was used compared with 24% (121/508) when it wasn't used (adjusted OR 1.14, 95% CI 0.83 to 1.57). The rate of uterine incision extension in the unit with availability of the balloon disimpaction device was 24% (139/574) compared with 29% (148/506) in a unit without availability of the device (adjusted OR 0.73, 95% CI 0.54 to 0.99).

In the randomised controlled trial of 60 caesarean deliveries, the rate of uterine incision extension was 20% (6/30) when a balloon disimpaction device was used and inflated compared with 43% (13/30) when the device was not inflated (p=0.05). Of the 13 extensions in the non-inflated group, 4 were into the cervix, vagina or bladder compared with none in the inflated group (Lassey 2020). In the randomised controlled trial of 50 caesareans at full dilation with a deeply engaged fetal head, the rate of uterine incision extension was 8% in the balloon disimpaction group and 24% in the modified Patwardhan technique group (p=0.001; Dutta 2019).

In the retrospective cohort study of 160 caesarean deliveries, the rate of uterine angle extension was 20% (18/91) when balloon disimpaction was used and 35% (24/69) when a hand-push technique was used (p=0.061; Safa 2016).

In the prospective non-randomised comparative study of 147 caesarean deliveries, the rate of uterine angle extension was 25% (19/75) when balloon disimpaction was used and 18% (13/72) when the Patwardhan technique was used (p=0.285; Barman 2015).

Uterine wall rupture

Three cases of uterine wall rupture were described in 2 papers (Jordan 2022, Mumtaz 2023). In 1 woman, who had had a previous caesarean, the lower segment of the uterus ruptured into the bladder before the baby was delivered. The rupture was surgically repaired and the total maternal blood loss was 3.1 litres (Jordan 2022). In the 2 other women, 1 of whom had had a previous caesarean, there was a partial or full thickness uterine posterior wall rupture identified after the baby was delivered. These were repaired surgically and estimated blood loss was 1,000 ml and 1,300 ml respectively (Mumtaz 2023). In all 3 cases, the baby was delivered in good condition (Apgar scores 8 and above) and immediate and long-term recovery were uneventful.

Uterine artery tear

In the prospective non-randomised comparative study of 147 caesarean deliveries, the rate of uterine artery tear was 20% (15/75) when balloon disimpaction was used and 10% (7/72) when the Patwardhan technique was used (p=0.081; Barman 2015).

Bleeding

In the systematic review by Cornthwaite et al. (2024), the RR for postpartum haemorrhage (blood loss more than 1,000 ml) when balloon disimpaction was used compared with no balloon was 1.13 (95% CI 0.83 to 1.54, I²=0%) based on 3 studies (Chooi 2023, Hanley 2020 and Sacre 2021).

In the randomised controlled trial of 60 caesarean deliveries, the median blood loss was 800 ml when an inflated balloon disimpaction device was used compared with 900 ml when a non-inflated device was used (mean difference=‑191.7, p=0.09; Lassey 2020). In the randomised controlled trial of 50 caesareans at full dilation with a deeply engaged fetal head, the rate of blood transfusion was 0% in the balloon disimpaction group and 16% in the modified Patwardhan technique group (p=0.002; Dutta 2019).

In the retrospective cohort study of 160 caesarean deliveries, the mean estimated blood loss was 273 ml when balloon disimpaction was used and 403 ml when a hand-push technique was used (p=0.026). The rate of blood transfusion was 3% in both groups (Safa 2016).

In the prospective non-randomised comparative study of 147 caesarean deliveries, the rate of blood transfusion was 5% (4/75) when balloon disimpaction was used and 7% (5/72) when the Patwardhan technique was used (Barman 2015).

Operative time

In the randomised controlled trial of 60 caesarean deliveries, the median time from hysterotomy to delivery was 31 seconds when an inflated balloon disimpaction device was used compared with 54 seconds when a non-inflated device was used (mean difference=‑38.2 seconds, 95% CI -56.1 to -20.3, p<0.01; Lassey 2020). In the randomised controlled trial of 50 caesarean deliveries, the uterine incision to delivery time was 0 to 2 minutes in 52% of those who had balloon disimpaction and 8% of those who had modified Patwardhan technique, 2 to 4 minutes in 48% of those who had balloon disimpaction and 44% of those who had modified Patwardhan technique and 4 to 6 minutes in none of those who had balloon disimpaction and 48% of those who had modified Patwardhan technique (p=0.04; Dutta 2019).

In the prospective non-randomised comparative study of 147 caesarean deliveries, the mean baby delivery time was 180 seconds when balloon disimpaction was used and 270 seconds when the Patwardhan technique was used (p<0.0001). The mean total operative time (skin to skin) was 36.8 minutes in the balloon disimpaction group and 40.2 minutes in the Patwardhan technique group (p=0.004; Barman 2015).

Admission to high dependency unit

In the cohort study of 101 caesarean deliveries, the rate of admission to a high dependency unit was similar with or without balloon disimpaction (11% in the group with balloon disimpaction and 10% in the group without balloon disimpaction (p=0.88; Chooi 2023).

In the randomised controlled trial of 60 caesarean deliveries comparing the use of an inflated balloon disimpaction device with a non-inflated device, the authors stated that intensive care unit admission was a rare event and not different between the groups (Lassey 2020).

Length of hospital stay

In the retrospective cohort of 173 caesarean deliveries with or without balloon disimpaction, the mean length of stay was 65.8 hours in the balloon disimpaction group and 72.6 hours in the control group (p=0.71; Hanley 2020). In the cohort study of 101 caesarean deliveries, the mean length of stay was 3.22 days in the balloon disimpaction group and 3.45 in the control group (mean difference=-0.23, 95% CI -0.75 to 0.29, p=0.39; Chooi 2023). In the retrospective cohort study of 160 caesarean deliveries, the mean length of stay was 77.9 hours when balloon disimpaction was used and 97.8 hours when a hand-push technique was used (p=0.002; Safa 2016).

Ease of delivery

In the randomised controlled trial of 60 caesarean deliveries comparing the use of an inflated balloon disimpaction device with a non-inflated device, healthcare providers described 80% (24/30) of deliveries in the inflated group as 'easy' or 'very easy' compared with 43% (13/30) in the not-inflated group. None of the deliveries in the inflated group were described as 'difficult' or 'very difficult' compared with 37% (11/30) in the not-inflated group (p<0.01). 97% (29/30) of healthcare providers in the inflated group would use the balloon elevation device again or recommend it to others compared with 77% (22/30) in the not-inflated group (p=0.03; Lassey 2020).

In the prospective non-randomised comparative study of 147 caesareans, delivery of the fetal head was described as easy in 76% (57/75) of those in the balloon disimpaction group and 32% (23/72) of those in the Patwardhan technique group (OR=6.75, 95% CI 3.08 to 14.95; Barman 2015).

Neonatal outcomes

Neonatal trauma

Neonatal trauma (soft tissue injury, bony trauma or intracranial bleed) was reported in 30% (16/53) of babies in the balloon disimpaction group and 23% (11/47) of babies in the control group, in the cohort study of 101 caesarean deliveries (RR=1.29, 95% CI 0.67 to 2.59, p=0.45; Chooi 2023).

Apgar score at 5 minutes

In the systematic review by Cornthwaite et al. (2024), the RR for an Apgar score of less than 7 at 5 minutes when balloon disimpaction was used compared with no balloon was 1.01 (95% CI 0.84 to 1.21, I²=0%) based on 3 studies (Chooi 2023, Hanley 2020 and Sacre 2021).

In the randomised controlled trial of 60 caesarean deliveries comparing the use of an inflated balloon disimpaction device with a non-inflated device, the median Apgar score at 5 minutes was 9 in both groups (Lassey 2020).

In the retrospective cohort study of 160 caesarean deliveries, the proportion of babies with an Apgar score below 7 at 5 minutes was 3% (3/91) when balloon disimpaction was used and 6% (4/69) hours when a hand-push technique was used (p=0.44; Safa 2016).

In the prospective non-randomised comparative study of 147 caesarean deliveries, the mean Apgar score at 5 minutes was 7.79 in the balloon disimpaction group and 5.96 in the Patwardhan technique group (p<0.0001). Two babies in the Patwardhan technique group were stillborn (Barman 2015).

Admission to special care baby unit or neonatal intensive care unit

In the randomised controlled trial of 50 caesarean deliveries, 12% of babies in the balloon disimpaction group and 60% of those in the modified Patwardhantechnique group were admitted to a special newborn care unit (p=0.04; Dutta 2019). One of the babies in the modified Patwardhan group died on day 11.

None of the retrospective non-randomised comparative studies reported a statistically significant difference in the rate of admission to a special care baby unit or neonatal intensive care. In the cohort study of 391 caesarean deliveries, the rate of admission was 9% (15/170) in the balloon disimpaction group and 12% (27/221) in the control group (RR 0.72, 95% CI 0.40 to 1.31, p=0.29; Sacre 2021). In the cohort study of 173 caesarean deliveries with or without balloon disimpaction, 33% (37/112) of babies in the balloon disimpaction group had a stay in neonatal intensive care compared with 40% (24/60) in the control group (OR=0.76, 95% CI 0.39 to 1.47, p=0.41; Hanley 2020). The mean duration of stay was 1 hour in the balloon disimpaction group and 1.5 hours in the control group (p=0.24). In the cohort study of 101 caesarean deliveries, the rate of admission was 30% (16/53) in the balloon disimpaction group and 47% (22/47) in the control group (RR=0.64, 95% CI 0.38 to 1.07, p=0.09; Chooi 2023). In the retrospective cohort study of 160 caesarean deliveries, the rate of admission was 15% (14/91) when balloon disimpaction was used and 25% (17/69) when a hand-push technique was used (p=0.14; Safa 2016).

In the prospective non-randomised comparative study of 147 caesarean deliveries, the rate of admission to a neonatal intensive care unit was 17% (13/75) in the balloon disimpaction group and 33% (24/72) in the Patwardhan technique group (p<0.0001). Two babies in the Patwardhan technique group were stillborn (p=0.025).

Umbilical cord arterial pH

In the retrospective cohort study of 391 caesarean deliveries, umbilical cord arterial pH was less than 7.1 in 7% (12/170) of babies in the balloon disimpaction group and 13% (29/221) in the control group (RR 0.54, 95% CI 0.28 to 1.02, p=0.06; Sacre 2021). In the retrospective cohort of 173 caesarean deliveries with or without balloon disimpaction, the mean umbilical cord arterial pH was statistically significantly higher in the balloon disimpaction group (7.25) than in the control group (7.19; p=0.0001, Hanley 2020). In the cohort study of 101 caesarean deliveries, the mean cord arterial PH was similar in the 2 groups (7.25 in the balloon disimpaction group and 7.24 in the control group (p=0.92; Chooi 2023). In the retrospective cohort study of 160 caesarean deliveries, the mean cord arterial pH was statistically significant higher when balloon disimpaction was used (7.24) compared with a hand-push technique (7.19, p=0.003; Safa 2016).

Composite outcome

The cohort study of 1,703 caesareans used a composite neonatal outcome, which included perinatal death, hypoxic ischaemic encephalopathy, seizure, birth trauma, acidosis (cord pH less than 7.10) or neonatal intensive care unit admission for at least 48 hours. The rate for this outcome was 4% both before and after the balloon disimpaction device was introduced to the maternity unit. After the device was introduced to the maternity unit, the rate was 5% (18/375) when the device was used compared with 4% (20/508) when it wasn't used (adjusted OR 1.01, 95% CI 0.51 to 1.99). When comparing 2 different maternity units, the composite neonatal outcome was 5% (27/574) in the unit with availability of the balloon disimpaction device and 7% (33/506) in the unit without availability of the device (adjusted OR 0.73, 95% CI 0.39 to 1.36; Sadler 2024).

Seizures or jitters

In the randomised controlled trial of 50 caesarean deliveries, neonatal seizure was reported in none of the babies in the balloon disimpaction group and 4% of those in the modified Patwardhantechnique group (p=0.03; Dutta 2019). Seizure or jitters was reported in none of the babies in the balloon disimpaction group and 9% (4/47) of babies in the control group, in the cohort study of 101 caesarean deliveries (p=0.045; Chooi 2023).

Anecdotal and theoretical adverse events

Expert advice was sought from consultants who have been nominated or ratified by their professional society or royal college. They were asked if they knew of any other adverse events for this procedure that they had heard about (anecdotal), which were not reported in the literature. They were also asked if they thought there were other adverse events that might possibly occur, even if they had never happened (theoretical).

They listed the following anecdotal adverse events:

Neonatal sepsis
Time to insert
Failure

Three professional expert questionnaires for this procedure were submitted. Find full details of what the professional experts said about the procedure in the specialist advice questionnaires for this procedure.

Validity and generalisability

The published evidence includes data from the UK.
One study included some women whose cervix was less than fully dilated (7 cm or above; Chooi 2023). The other studies stated that the caesarean deliveries were done when the cervix was fully dilated or during the second stage of labour.
Of the 2 randomised controlled trials included in the systematic review by Cornthwaite et al. (2024), 1 was assessed as low risk of bias (Lassey 2020) and 1 was at high risk of bias (Dutta 2019). All 5 non-randomised studies included in the review were at serious or critical risk of bias. The certainty of evidence was low or very low, using GRADE criteria. One of the studies was identified as being at risk of retraction and has since been retracted. Another study was published in a journal that was described in the review as 'potentially predatory' (Dutta 2019).
The authors of the systematic review noted that there was a lack of standardisation in the measuring and reporting of outcomes associated with caesarean birth.
One small randomised controlled trial, included in the systematic review, compared an inflated device with an uninflated device. The primary outcome for this trial was the time between hysterotomy and delivery, and it was underpowered to detect significant differences in other outcomes (Lassey 2020).
In the cohort study by Sadler et al. (2023), there were statistically significant differences in demographic and clinical characteristics between comparison groups.
The primary outcome differed between studies.
The fetal head station was not reported in all studies and this is likely to be associated with the degree of difficulty encountered in the delivery.
None of the studies reported the time between the decision to do an emergency caesarean and delivery, or the time taken to insert the balloon disimpaction device.
Most of the studies were underpowered to detect clinically relevant differences in rarer outcomes.
Most of the non-randomised comparative studies were retrospective, so there is a risk of bias in patient selection and definitions of maternal and neonatal outcomes.
Ease of delivery is a subjective outcome and prone to bias.
In the studies by Lassey et al. (2020) and Dutta et al. (2019), balloon disimpaction devices were donated by Safe Obstetric Systems, UK. None of the other studies declared any potential conflicts of interest.

How are you taking part in this consultation?

Interventional procedure overview of balloon disimpaction of the baby's head at emergency caesarean during the second stage of labour