Title: Assisted vaginal delivery: an overview of its use in low and middle income countries

Shortened running title: AVD in low and middle income countries

Authors:

Patricia E. Bailey
RMNCH Unit, Global Health Programs
FHI 360
359 Blackwell Street
Durham, NC 27701
USA
Tel: 919-544-7040, extension 11523

Jos van Roosmaleen

Glen Mola
Head of Obstetrics and Gynecology, School of Medicine and Health Sciences,
Port Moresby General Hospital, Box 1156 Boroko, NCD
Papua New Guinea

Cherrie Evans
Helping Mothers Survive, Technical Leadership Office
Jhpiego
1615 Thames Street, Brown’s Wharf
Baltimore, MD 21231
USA

Luc de Bernis

Blami Dao
Jhpiego
1615 Thames Street, Brown’s Wharf
Baltimore, MD 21231
USA
Abstract

Objective

To assess the use of assisted vaginal delivery (AVD) in low and middle income countries, highlighting at
what level of care procedures were performed and systemic barriers to its use.

Design

Cross-sectional health facility assessments.

Setting

Up to 40 countries in Latin America and the Caribbean, sub-Saharan Africa, South and Southeast Asia
that conducted health facility assessments in the last 11 years.

Population or Sample

Surveys tended to be national in scope that included all hospitals and samples of midlevel facilities in
public and private sectors.

Methods

Descriptive secondary data analysis.

Main outcome measures

Percentage of facilities where health workers performed AVD in the 3 months prior to the survey,
instrument preference, which health workers performed the procedure, and why AVD was not
practiced.

Results

Fewer than 20% of facilities in Latin America and the Caribbean reported performing AVD in the last 3
months. In sub-Saharan African, 53% of 1,728 sub-Saharan hospitals had performed AVD but only 6% of
nearly 10,000 health centers had done so. It was not uncommon to find less than 1% of institutional
births delivered by AVD. Vacuum extraction appears preferred over forceps. Lack of equipment and
trained health workers were the two most frequent reasons for non-performance of AVD.

Conclusions

The low use of AVD in low and middle income countries is in stark contrast with many high income
countries, where high ising cesarean delivery rates are also associated with significant rates of AVD. On
the other hand in many developing countries, rising CS rates have not been associated with
maintenance of skills and practice of AVDmay create. a favorable climate for safe effective alternatives.
AVD is underutilized precisely in countries where pregnant women continue to face hardships accessing
emergency obstetric care.
Overview of the use of assisted vaginal delivery in low and middle income countries

Introduction (400 words, currently 441 words)

Assisted vaginal delivery (AVD) is most commonly performed under three conditions: suspicion of
immediate or potential fetal compromise, to shorten the second stage of labor for maternal benefit
(fatigue or when prolonged expulsive efforts are inadvisableon is not advised), or inadequate
progress/prolonged second stage of labor (1, 2). Over the past two decades a number of authorities
have declared the vacuum extractor (VE) the method of choice in modern obstetrical practice because
of safety for the fetus and less likelihood of maternal morbidity (3-6).

AVD reduces the rate of cesarean delivery and has pediatric and maternal benefits of a vaginal birth that
cesarean deliveries do not. In resource poor settings access to AVD has the potential to improve
maternal and newborn outcomes (PPH, postpartum sepsis, and fetal birth asphyxia are directly
relatedproportional to the duration of the second stage of labor).

In 2015 WHO issued a consensus statement regarding the optimum population-based level of cesarean
delivery, stating that rates above 10% are not associated with reductions in maternal or neonatal
mortality (7, 8). Cesarean deliveries carry an intrinsic risk, increase the costs of care and are associated
with sexual violence and psychological distress (9, 10). They also expose women to a lifetime risk in
subsequent pregnancies if access to quality antenatal and birthing care is unreliable.

The practice of AVD is more prevalent in high income countries than in low and middle income
countries. AVD rates in Australia, Canada, Ireland, Scotland and England range between 10 and 16%,
while in the Netherlands and Sweden rates have actually increased from 10 to 15% over the past 10
years (4, 11, 12). Although reliable global statistics on AVD are not available, there is unequivocal
evidence that cesarean deliveries are on the rise, reaching epidemic proportions in some countries (11,
13, 14).

A limited review of the emergency obstetric care (EmOC) signal functions in 2006 indicated that AVD
was the signal function least likely to have been performed across several global regions (15). A small
number of papers mostly describing single facility experiences with AVD has emerged from Latin
America, sub-Saharan Africa and Oceania, the majority of them calling for greater use of these
technologies (16-22).

The purpose of this paper is to compile an overview of the use of AVD (and to a lesser extent
craniotomy) from as many as 40 low and middle income countries that have carried out an assessment
on emergency obstetric and newborn care (EmONC) or health facility assessment in the last 11 years.
We describe the extent to which AVD was used, by whom and at what level of the health system, while
identifying some of the systemic barriers to its use.

Methods

EmONC assessments are cross-sectional health facility surveys that tend to be national in scope. They
rely on interviews with facility managers and health workers, extraction from registers and patient-level
charts, and limited observations. Content-wise they focus on intrapartum care for both routine
deliveries and complicated births.

Sampling
Ideally the scope of EmONC assessments is a census of all health facilities where deliveries take place, as
they are designed for planning purposes even at the facility level. In practice, the size and scope are
influenced by the funding available to conduct the field work. Assessments range from a census of
health facilities where deliveries were known or thought to take place (e.g. Angola 2007, Ethiopia 2008-
9, Guyana 2010, Mozambique 2012) to smaller purposively selected samples (e.g. Ecuador 2006,
Panama 2007, Lao PDR 2010-11). Most assessments, however, target all hospitals in the country and
select all lower level facilities that attend more than a fixed number of births per month – a “restricted
census” (e.g. Ghana 2010, Togo 2012 , Benin 2010), or they target all hospitals and a random sample of
lower level facilities (e.g. Malawi 2014, Zambia 2015).

Data collection forms

The Averting Maternal Death & Disability (AMDD) program developed an evolving and adaptable core
set of modular questionnaires (23). Among those relevant to this paper, modules covered basic
infrastructure; human resources (number of employees by cadre, what cadres provided the signal
functions); availability of drugs, equipment and supplies; service statistics of the last 12 months
including the number of deliveries and mode of delivery; performance of EmOC signal functions and
other key services, and if not performed, why not. These modules were adapted to each country’s
health system context, creating some variation in how items were reported (e.g. use of forceps or VE or
instruments combined, performance of the signal function in the last 3 or 12 months) or if certain items
were captured or reported (e.g. status of craniotomy).

Assessment organization

Assessments were led by ministries of health and supported by a range of partners. UNFPA and UNICEF
figured predominantly as in-country funders, but foundations and bilaterals also played a role. AMDD
technical advisors usually provided assistance during questionnaire adaptation, training of data
collectors, quality assurance, data management and analysis, and report writing. The intensity of
technical support ranged from country to country, and several countries included in this analysis
received no direct support from AMDD (e.g. Ecuador, Panama, Cote d’Ivoire, Eritrea, Zimbabwe, Nepal)
while Senegal, Afghanistan and Cambodia received little. Local public or private research institutions,
universities or government Central Statistical Offices coordinated the assessment process with oversight
by a local steering committee or technical working group.

Data collection and management

Data collectors were often health workers, public health school graduates, or retired midwives. Data
collectors were trained generally over the course of a one-week period that included a one-day field
activity of administering the forms in a nearby health facility. Teams of four were deployed to the field;
data collection often took 1-2 days in hospitals and half a day in health centers.

Initial analysis was performed using Excel, STATA, SPSS or CS-Pro. Statistical analyses were descriptive
and no statistical tests were conducted. When random samples of health centers were taken, those data
were weighted based on the facility’s probability of selection, and less frequently for non-response. How
the weighted data were reported varied from country to country.

Prior to training and deployment, local institutional review boards or ministries of health gave their
approval. The assessments never collected patient names or any identifiable information, nor were
health worker names documented in the data collection instruments. Teams requested consent from
facility directors on arrival as well as informed consent from health workers prior to their interview,
usually in oral format. For this paper, IRB approval was not sought given that most of the data were
extracted from final reports in the public domain that included no identifiable personal or facility level
information.

Analysis for this paper

The EmONC assessment process has never been highly centralized as country ownership and in-country
planning based on the data have been prioritized over the use of the data at global level for multi-
country analyses. However, we also recognize this missed opportunity. Without access to primary data,
we have relied on final reports. As many as 40 country reports were consulted, though none of the
tables includes all 40 countries.

Results

Frequency of AVD performance at hospital and non-hospital health facility levels

Table 1 summarizes 36 assessments conducted over the past 11 years, 27 of which are located in sub-
Saharan Africa; it shows the size of the population covered by the assessment, how many hospitals and
other non-hospital facilities were surveyed (a total of 15,720 health facilities), and the percentage of
hospitals and other facilities where AVD (defined as VE or forceps delivery) was performed in the last 3
months. Assessments ranged from 9 hospitals in one province in Ecuador to as many as 1,626 health
facilities in Burkina Faso.

To determine the EmOC status of each facility, the data collectors asked if each signal function had been
performed in the last three months. The percentage of hospitals where AVD was performed was low in
the four Latin American and Caribbean countries, ranging from 11% in Ecuador to 27% of hospitals in
Guyana and Panama. The percentage of hospitals in sub-Saharan Africa that had practiced AVD was
generally higher than the Latin American countries, ranging from 15% in Cote d’Ivoire to 94% in Eritrea.
Approximately 53% of the 1,728 surveyed sub-Saharan hospitals had conducted AVD in the last three
months (data not shown). Among the South and Southeast Asian hospitals, between a 31% and 98% of
the hospitals had practiced AVD in the last 3 months.

Many fewer health centers than hospitals reported practicing AVD. Countries where more than 10% of
health centers reported performing instrumental delivery included Niger and the Gambia (13% each) in
West Africa; Eritrea (34%), Ethiopia (27%), Malawi (18%) and Mozambique (11%) in East Africa; and
Chad (12%) and Angola (11%) in Central Africa. Cambodia (49%) and Afghanistan (89%) stand out with
half or more health centers providing AVD in the last three months, but facility sampling in these two
countries was unusual. The 2014 Cambodia assessment targeted only facilities that had been designated
for upgrading to basic or comprehensive EmOC status after the 2008 national assessment; and in
Afghanistan, the nine “other” facilities were comprehensive health centers in provinces without a
provincial or district hospital and served as first referral comprehensive EmOC centers.

A rapid overview of 40 assessments (those included in Table 1 plus Cameroon 2010, Mongolia 2009,
Nicaragua 2006, and Zanzibar 2012) showed that in 35 countries AVD was the most frequently missing
basic EmOC signal function. In Lesotho and Madagascar, AVD was tied as the least commonly reported
basic signal function with removal of retained products and neonatal resuscitation, respectively. In the
other five assessments, parenteral anticonvulsants were reported less frequently than AVD in Cambodia,
Lao PDR and Nepal, removal of retained products in Niger, and anticonvulsants and neonatal
resuscitation were both less frequent than AVD in Zanzibar.

Mode of delivery among institutional births

The percentage of facilities that recently performed AVD is an indicator best used in the context of the
other signal functions in order to identify the gaps in the provision of life-saving procedures, and thus,
programmatic, service delivery gaps. A more traditional indicator is to examine the distribution of
institutional deliveries by mode of delivery. For the 7 countries in Table 2 about 1% of all institutional
deliveries in the six sub-Saharan African countries were delivered by VE or forceps. Lao PDR, the only
non-African country, recorded as many as 3.4%, which might reflect its hospital-only sample. As
comparison, data from the WHO Global Survey and the African sampled countries are also shown, both
of which show about 3% of institutional deliveries (Shah 2009), possibly because the Global Survey
sampling strategy included only high volume hospitals.

The total assisted delivery rate (ADR) (the sum of cesareans and deliveries with VE or forceps divided by
all institutional deliveries) ranged from 5% in Senegal to 12.8% in Ghana. In theory, the total ADR
represents the percentage of pregnant women requiring an intervention for delivery, either abdominal
or vaginal, while taking into account the case mix of the populations who deliver in facilities (11). The
ratio of cesarean deliveries (CD) to instrumental deliveries highlights the differences in interventions for
delivery and allows for greater comparability across countries. The CD/AVD ratio ranges from a ratio of
2.1 (2.1 cesarean deliveries for every instrumental vaginal delivery) in Lao PDR to 22 in Ghana, indicating
a heavy reliance on surgery.

Reasons why AVD was not performed

Twenty-four countries reported systemic reasons why AVD had not been performed in the previous
three months (Table 3). Providers were allowed to give multiple responses. Including ties, the top reason
for non-performance in 10 countries was equipment-related; lack of staff training in seven countries;
policy-related issues (mostly related to the authorization of human resources) in four countries; and no
patient presented with the appropriate indication for AVD in six countries. In the case of “no indication,”
poor diagnostic skills, minimal training exposure to AVD and/or a lack of confidence or support for the
skills may be additional explanations. The availability of equipment and experienced human resources
were validated by specific questions targeted at these systemic problems.

Availability of equipment

The availability of functioning equipment was explored in an inventory of equipment, supplies and
drugs. The type of equipment found suggests clinical preferences (VE or forceps) in each country. In
some countries, the inventory interrogated about three types of delivery forceps (breech, mid-cavity,
and outlet forceps). Thus, a range in the percentage of facilities with functional forceps will indicate the
presence of more than one type (e.g. South Sudan 2013, Republic of Congo 2012). In Burkina Faso and
Malawi, vacuum extractors and delivery forceps were combined.

Table 4 confirms that hospitals were better equipped than non-hospitals regardless of the type of
equipment reported. It also indicates wider use of VE over forceps delivery, which is born out in Table 3,
where the percentage of births delivered by VE is higher than forceps in five countries, with the Republic
of Congo the exception (Mozambique did not distinguish between instruments). A clinical preference for
forceps over VE can be observed in Ecuador, Guyana, Mauritania, Republic of Congo, and Bangladesh.

Availability of human resources

Human resource availability, competency and confidence drive the practice of AVD. The EmONC
assessments varied widely in how they reported on human resources. Graphs from several countries
illustrate important patterns of AVD provision. Figure 1 shows the percentage of hospitals and non-
hospitals in Ghana staffed by different cadres who provided VE or forceps delivery. In hospitals, a higher
percentage reported health workers who provided VE than forceps: 46% of hospitals had a general
practitioner (GP), 37% an obstetrician/gynecologist (ob/gyn), and 30% a midwife who could perform VE,
compared with 17% of hospitals having a GP who provided forceps delivery, 23% an ob/gyn, and 2% a
midwife. In contrast, few health centers and clinics reported staff who reported using either instrument,
the exception being 15% that had a midwife who could perform VE. Note that in this graph and those
that follow, we do not know who among the different providers actually performed most of the
instrumental deliveries.

Ethiopia (Figure 2) is similar to Figure 1 only the type of equipment (vacuum extractor or forceps) is not
specified. A wide range of cadres reportedly conduct AVD at both hospitals and non-hospitals. More
than half of the hospitals reported a GP (71%), an ob/gyn (55%) or a diploma midwife (62%) capable of
performing AVD, and close to a third had a health officer (33%), a B.Sc. midwife (29%) or a diploma
nurse (29%). At non-hospitals mostly mid-level providers were performing AVD.

Hospitals in Mozambique present a different health worker pattern from the hospitals in Ghana and
Ethiopia where GPs and specialists were reported frequently (as were midwives) (Figure 3). About three-
quarters of Mozambican hospitals reported that medium and basic-level MCH nurses provided AVD,
compared to 35% of hospitals reporting a GP or surgical technician, and 25% an ob/gyn. Basic-level MCH
nurses providing AVD were also found in 31% of non-hospitals.

The final illustrative figure (Figure 4) shows two Southeast Asian countries -- Laos and Cambodia. In both
countries, facilities reported GPs as providers of AVD with greater frequency than other cadres
(although the instrument is not specified, evidence from Tables 2 & 4 suggests that VE predominates
over obstetric forceps in both countries). In Cambodia 42% of surveyed facilities also reported midwives
providing AVD.

Craniotomy

Most assessments included questions about the provision of craniotomy or other destructive deliveries
but few reported this information. Table 5 presents the percentage of surveyed facilities in 13 countries
that had practiced craniotomy in the last three months, and if they had not, eight countries reported
their reasons why. No facility in the Gambia had performed a craniotomy in the past three months while
as many as 21% of the facilities in Afghanistan had done so.

Discussion [<1200 words, currently 1335]

Main Findings

This overview of AVD in 40 countries shows that instrumental delivery is not widely practiced in the low
and middle income countries surveyed. AVD is more likely to be practiced in hospitals than in health
centers and clinics: in sub-Saharan Africa, the region best represented, 53% of hospitals reported recent
practice of AVD compared to only 6% of non-hospitals. As a proportion of institutional deliveries, usually
less than 1% were delivered with AVD.

Most surveyed countries appear to show a preference for VE over forceps, especially at non-hospitals.
This likely reflects the global transition in instrument choice, what is offered in preservice training and
residency programs and a consensus that forceps are more difficult to use and are less versatile (24, 25).

Staff at facilities reported that the primary obstacles to performing AVD were the lack of trained human
resources, lack of equipment, and national and institutional policies that fail to support AVD. Specialists
figured prominently as providers in hospitals as did midwives, while midwives were clearly the most
prevalent providers at non-hospitals. Mozambique exemplifies a country where facilities are staffed
with midwives who are trained to use VE, and reflects their successful task-sharing strategy for providing
obstetric care (26).

Few countries reported on the practice of craniotomy although standard questions about its practice
appeared in most questionnaires, and when reported, only single digit percentages of health facilities
responded affirmatively, with Afghanistan the exception. Again, training and equipment figured as
major reasons for not performing craniotomy.

Strengths and Limitations

A strength of our study is the presentation of national/regional AVD practice patterns in hospitals and
non-hospitals in many low and middle income countries, precisely where timely access to medical care is
not a given.

Few countries have had multiple assessments thus we have minimal trend data. If the purpose of an
EmONC assessment is to inform planning and implementation of routine delivery and EmONC services,
we might expect to see improvements in the years following an assessment, given how widespread
routine delivery and basic EmOC training has been. In Mozambique, which had its first national
assessment in 2007-8 and a second in 2012, the percentage of hospitals and health centers performing
AVD decreased. But the opposite was true for Malawi where three national assessments have been
conducted (2005, 2010, 2014), and where steady increases have been observed in both hospitals and
health centers with each assessment.

Other limitations to the data include the likelihood of under-reporting AVD procedures (and destructive
deliveries), which are often recorded simply as vaginal deliveries. The practitioner of a procedure may
not always be the person who completes the delivery or operating theatre register, primary data
sources in assessments. The performance of signal functions in recent months is determined through
interviews with the maternity in-charge, or her/his designee, and verified by examining the labor and
delivery ward registries. However, we know that staff may be unware of procedures performed by their
colleagues, data collectors do not always verify the performance results with service statistics, and
responses may be influenced by a courtesy bias. Furthermore, EmONC assessments historically have not
captured exactly who performs most of the procedures, something that has been corrected in current
questionnaires.

Had we had electronic access to the databases, we could have better standardized and refined analyses
to examine how practices varied by public private sector, urban rural differentials, high and low delivery
volume, quality or readiness indices, and other predictors of practice as well as mapped the geospatial
distribution of services.

Interpretation

The contrast between AVD’s relatively frequent use in high income countries and its low use in low and
middle income countries raises questions. The high rates of medical interventions during childbirth in
high income countries are well recognized; the total assisted delivery rate (cesarean deliveries + AVD) is
approaching or exceeds 40% in a number of English-speaking countries (11). National cesarean delivery
rates in many of the surveyed countries were below 5%. Although low, they mask some extremely high
rates in private hospitals and urban centers, revealing not just the inequitable distribution of emergency
services, but may be a harbinger of practice patterns in the future. AVD is an evidence-based obstetric
practice and when performed for appropriate indications and according to protocol can reduce maternal
and fetal morbidity and mortality, reduce short term and long term risks associated with cesarean
delivery, and reduce healthcare costs. Some low and middle income countries already face acute
financial crises associated with high cesarean rates and are proposing an increase in AVD as a path to
mitigating rising cesarean delivery rates (17, 19).

Two tiers of emergency obstetric care – basic and comprehensive – have been applied to countries
where governments maintain relatively few large hospitals but supplement those hospitals with highly
functioning health centers. Where these exist, they can and do treat a sizable proportion of the
pregnant women who develop complications, in addition to providing curative care for newborns and
young children. Theoretically, B-EmOC centers should decongest hospitals, be located closer to
communities, and be well linked to comprehensive EmOC facilities by the referral system.

The most controversial B-EmOC signal function is AVD where the debate centers on two issues: a
concern from some specialists who question its safety at a health center with no immediate access to
surgical backup; and from program managers and evaluators, who sometimes redefine basic EmOC by
dropping AVD as a signal function because it is so frequently missing. We do not have the rigorous
evidence for task-shifting AVD as we do for cesarean deliveries performed by surgical technicians, health
officers and other associate clinicians (26). The research community could provide the evidence that
midwives can perform VE as safely and effectively as doctors, as well as determine the costs and
benefits of performing AVD at the health center level. An advantage of midwives as AVD practitioners is
to reduce delays implicit in calling a specialist or other provider, especially at night. As long as specialists
are in short supply, midwives and associate clinicians should be targeted for preservice and in-service
skills-building around the use of VE. However, if specialists are not skilled or not in favor of the use of
VE, there will be fewer leaders for more supportive policy, training or use.

In response to the program managers who would like to show more progress from their B-EmOC
strengthening efforts, some countries like Panama and Nicaragua completely dropped AVD from their
assessments, other countries have devised a status of Basic -1, where the ‘-1’ stands for the missing
AVD. We believe the clinical importance of keeping a focus on AVD (and on craniotomy) is paramount,
and if it were dropped, would communicate an inappropriate obstetrical message of ‘unimportance’.

Conclusions
To turn around the underutilization of AVD, research on AVD’s safety and effectiveness in the hands of
midwives at health centers is encouraged. Global endorsement and a consensus to revitalize the use of
AVD are needed to activate policy makers and encourage stakeholder professional societies at stake
(obstetricians/ gynecologists, pediatricians/neonatologists, midwives, associate clinicians) to play a role
that could affect training policies and national guidelines, and increase clinical training opportunities,
even for ob/gyn specialists who have lost operative vaginal skills. Where AVD has been dropped from
preservice training or in-service B-EmOC training, we should understand why this has occurred. We need
to learn from successful pilots how to reintroduce AVD; there are examples in Tanzania (health centers
upgraded by the World Lung Foundation to provide cesarean delivery and AVD), in Uganda at the
Mulago National Referral Hospital and Ecuador at the Sotomayor Hospital in Guayaquil (17, 27, 28).
Finally we need to support new inexpensive device designs that do not require electricity and can be
used easily by midwives.
References

1. American College of Obstetricians and Gynecologists. Operative Vaginal Delivery. 2000.

2. Royal College of Obstetricians and Gynaecologists. Operative Vaginal Delivery. 2005.
3. Learman LA. Regional differences in operative obstetrics: a look to the South. Obstet Gynecol.
1998;92(4 Pt 1):514-9.
4. Hehir MP, Reidy FR, Wilkinson MN, Mahony R. Increasing rates of operative vaginal delivery
across two decades: accompanying outcomes and instrument preferences. Eur J Obstet Gynecol Reprod
Biol. 2013;171(1):40-3.
5. Ali UA, Norwitz ER. Vacuum-assisted vaginal delivery. Rev Obstet Gynecol. 2009;2(1):5-17.
6. Althabe F. Choice of instruments for assisted vaginal delivery: RHL commentary. The WHO
Reproductive Health Library. Geneva: World Health Organization; 2011.
7. World Health Organization Human Reproduction Programme A. WHO Statement on caesarean
section rates. Reprod Health Matters. 2015;23(45):149-50.
8. Betrán AP, Torloni MR, Zhang JJ, Gulmezoglu AM, Section WHOWGoC. WHO Statement on
Caesarean Section Rates. BJOG. 2016;123(5):667-70.
9. Souza JP, Gulmezoglu A, Lumbiganon P, Laopaiboon M, Carroli G, Fawole B, et al. Caesarean
section without medical indications is associated with an increased risk of adverse short-term maternal
outcomes: the 2004-2008 WHO Global Survey on Maternal and Perinatal Health. BMC Med. 2010;8:71.
10. Filippi V, Ganaba R, Calvert C, Murray SF, Storeng KT. After surgery: the effects of life-saving
caesarean sections in Burkina Faso. BMC Pregnancy Childbirth. 2015;15:348.
11. Gei AF. Prevention of the first cesarean delivery: the role of operative vaginal delivery. Semin
Perinatol. 2012;36(5):365-73.
12. Elvander C, Ekeus C, Gemzell-Danielsson K, Cnattingius S. Reason for the increasing use of
vacuum extraction in Sweden: a population-based study. Acta Obstet Gynecol Scand. 2013;92(10):1175-
82.
13. Betrán AP, Merialdi M, Lauer JA, Bing-Shun W, Thomas J, Van Look P, et al. Rates of caesarean
section: analysis of global, regional and national estimates. Paediatr Perinat Epidemiol. 2007;21(2):98-
113.
14. Hellerstein S, Feldman S, Duan T. China's 50% caesarean delivery rate: is it too high? BJOG.
2015;122(2):160-4.
15. Bailey P, Paxton A, Lobis S, Fry D. The availability of life-saving obstetric services in developing
countries: an in-depth look at the signal functions for emergency obstetric care. Int J Gynaecol Obstet.
2006;93(3):285-91.
16. Opoku BK. A Review of Vacuum Deliveries at Komfo Anokye Teaching Hospital, Kumasi. Ghana
Medical Journal. 2006;40(1):4.
17. Chang X, Chedraui P, Ross MG, Hidalgo L, Penafiel J. Vacuum assisted delivery in Ecuador for
prolonged second stage of labor: maternal-neonatal outcome. J Matern Fetal Neonatal Med.
2007;20(5):381-4.
18. Nkwabong E, Nana PN, Mbu R, Takang W, Ekono MR, Kouam L. Indications and maternofetal
outcome of instrumental deliveries at the University Teaching Hospital of Yaounde, Cameroon. Trop
Doct. 2011;41(1):5-7.
19. Ayala-Yanez R, Bayona-Soriano P, Hernandez-Jimenez A, Contreras-Rendon A, Chabat-
Manzanera P, Nevarez-Bernal R. Forceps, Actual Use, and Potential Cesarean Section Prevention: Study
in a Selected Mexican Population. J Pregnancy. 2015;2015:489267.
20. Okeke T, Ekwuazi K. Is there Still a Place for Vacuum Extraction (Ventouse) in Modern Obstetric
Practice in Nigeria. Ann Med Health Sci Res. 2013;3(4):471-4.
21. Mola GD, Amoa A, Edilyong J. Factors associated with success or failure in trials of vacuum
extraction. Australian and New Zealand Journal of Obstetrics and Gynaecology. 2002;42:41-5.
22. Dao B. Utilisation de la Ventouse dans 5 Pays de la Sago. FIGO Conference; 8 October 2015;
Vancouver, Canada2015.
23. AMDD. 2016. Available from: https://www.mailman.columbia.edu/research/averting-maternal-
death-and-disability-amdd/toolkit.
24. Goetzinger KR, Macones GA. Operative vaginal delivery: current trends in obstetrics. Womens
Health (Lond Engl). 2008;4(3):281-90.
25. Johanson RB, Menon V. Vacuum extraction versus forceps for assisted vaginal delivery. The
Cochrane Database of Systematic Reviews. 1999(2).
26. Schneeberger C, Mathai M. Emergency obstetric care: Making the impossible possible through
task shifting. Int J Gynaecol Obstet. 2015;131 Suppl 1:S6-9.
27. Dominico SA, Mwakatundu N, Mohamed H, Lobis S. Reviving vacuum extraction practice in
primary health care facilities in rural Tanzania. Tanzania Health Summit; 3 November 2015; Dar es
Salaam, Tanzania2015.
28. van Roosmalen J. FIGO; 8 October 2015; Vancouver, Canada2015.