Table 1.

Characteristics of included studies

Miller et al., 2006 [18]
Study design	Controlled before-after
Population	Women with obstetric haemorrhage and signs of shock (>750 mL of blood loss and either pulse of >100 beats per minute or systolic blood pressure of <100 mmHg)
Intervention	Standard care versus standard care plus NASG.
	Standard protocol: “administration of crystalloid intravenous fluids, use of uterotonic medications (IV oxytocin and rectal misoprostol), uterine massage, determining the source of bleeding, and providing blood transfusions and surgery as necessary. Vaginal procedures included manual removal of the placenta, suturing of lacerations, manual vacuum aspiration or suction curettage, and uterine evacuation using ring forceps (also known as ‘sponge sticks’ or ‘sponge holders’). Surgery was performed when the haemorrhage continued despite the other interventions. Practitioners used a variety of surgical interventions (step-wise ligation of uterine arteries, B-Lynch suture, or hysterectomy) according to their experience and the clinical situation. The only difference in care for the women in the NASG group was that the NASG was placed on the woman when she met study entry criteria. The NASG was left in vaginal procedures.”
Outcomes	Maternal mortality, maternal morbidity, overall blood loss
Setting	Four university teaching facilities in Egypt
Quality assessment	GRADE quality of evidence: Low, downgraded for study design, without serious limitations in risk of bias, inconsistency, and indirectness or imprecision.
	EPOC methods assessment: low risk of bias considering study design.
Miller et al., 2010 [19]
Study design	Controlled before-after
Population	All “pregnant (with a non-viable fetus), birthing, or postpartum women experiencing hypovolemic shock secondary to obstetric haemorrhage from any aetiology. Additional inclusion criteria were estimated blood loss of at least 1000 mL and/or at least 1 clinical sign of hypovolemic shock (systolic blood pressure 100 mm Hg or pulse 100 beats per minute)”.
Intervention	Standard care plus NASG versus standard care.
	Standard care consisted on “administration of crystalloid intravenous fluids (≥1500 mL in the first hour following study admission); administration of uterine massage and uterotonic medications for uterine atony (intravenous or intramuscular oxytocin, intramuscular ergometrine, and rectal misoprostol); vaginal procedures; provision of blood transfusions (standard in the 2 study sites for women who lost ≥1000 mL of blood); and surgery. The NASG was left in vaginal procedures”.
Outcomes	Extreme adverse outcomes combined with maternal mortality and morbidity; secondary outcomes (overall blood loss, urine output, emergency hysterectomy for uterine atony)
Setting	Two tertiary hospitals Egypt
Quality assessment	GRADE quality of evidence: Low (downgraded by study design, without serious limitations in risk of bias, inconsistency, indirectness or imprecision).
	EPOC methods assessment: low risk of bias considering study design.
Ojengbede et al., 2011 [20]
Study design	Controlled before-after
Population	Woman with post-partum haemorrhage (initial blood loss of ≥750 ml) due to uterine atony, retained placenta, ruptured uterus, vaginal and cervical lacerations or placenta accreta and one clinical sign of shock (systolic blood pressure <100 mmHg or pulse >100 beats/min).
Intervention	Standard care plus NASG versus standard care.
	Standard care: “administration of crystalloid intravenous fluids (≥1,500 ml in the first hour), uterotonic medications (oxytocin, ergometrine, syntometrine, misoprostol), uterine massage for patients with uterine atony, vaginal procedures (bimanual compression, manual removal of placenta or dilation, repair of lacerations and curettage for retained tissue) and abdominal surgeries (arterial ligation, B-Lynch compression sutures, hysterectomy) as necessary”.
Outcomes	Maternal mortality and overall blood loss
Setting:	Four tertiary facilities in Nigeria (2 teaching and 2 state facilities)
Quality assessment	GRADE quality of evidence: Low (downgraded by study design, without serious limitations in risk of bias, inconsistency, indirectness or imprecision).
	EPOC methods assessment: low risk of bias considering study design.
Maknikar et al., 2012 [22]
Study design	Non randomized clinical trial
Population	Woman with post-partum haemorrhage and signs of hypovolemic shock
Intervention	Standard care plus NASG versus Standard care
Outcomes	Maternal mortality, maternal morbidity, overall blood loss
Setting	Fifteen facilities India
Quality assessment	GRADE quality of evidence: Low (downgraded by study design, without serious limitations in risk of bias, inconsistency, indirectness or imprecision).
	EPOC methods assessment: high risk of bias as stated “lack of randomization”
	Only abstract available with complementary information provided by contact author
Magwali et al., 2012 [21]
Study design	Before-after
Population	Woman with post-partum haemorrhage and signs of hypovolemic shock
Intervention	Standard care versus standard care plus NASG
Outcomes	Maternal mortality and overall blood loss
Setting	Two hospitals Zimbabwe
Quality assessment	GRADE quality of evidence: Low (downgraded by study design, without serious limitations in risk of bias, inconsistency, indirectness or imprecision).
	EPOC methods assessment: low risk of bias considering study design.
	Only abstract available with complementary information provided by contact author.
Miller et al., 2013 [23]
Study design	Cluster randomized controlled trial
Population	Woman with obstetric haemorrhage from any aetiology and hypovolemic shock before removal from primary health care centres to higher level complexity of care facility, with at least two of the following eligibility criteria: visually estimated blood loss >500 mL, pulse > 100 BPM, systolic blood pressure <100 mm Hg.
Intervention	Standard care plus NASG application versus standard care
	“Standard shock/haemorrhage protocol: oxygen, IV fluids, uterotonics/uterine massage (for uterine atony), suturing of lacerations, manual removal of placenta or retained tissues, MVA, surgery, and blood transfusion, as necessary. The only differences in treatment received depended on haemorrhage aetiologies”.
Outcomes	Maternal mortality rates; survival with severe maternal morbidity; and extreme adverse outcome. As secondary outcomes included median blood loss measured by weighing the absorbent pad(s) upon admission; blood loss measured in the drape at arrival; blood loss during surgery; frequency of emergency hysterectomy for intractable uterine atony; and time to recovery from shock (defined as return to Shock Index (SI) 0.98 (SI = Heart Rate/Systolic Blood Pressure) as well as negative effects that might be attributable to the NASG application (decreased urine output, respiratory difficulties, nausea, vomiting and abdominal pain).
Setting	Primary health care services (38) in Zambia and Zimbabwe
Quality assessment	GRADE quality of evidence: Low (downgraded because imprecision, few events). No other limitation found.
	EPOC methods assessment: low risk of bias considering study design.