Risk Factors for Hemorrhage in The Third Stage of Labor: A Systematic Review Protocol

Postpartum hemorrhage (PPH) is the most common form of obstetric hemorrhage. This is the main cause of maternal death around the world: the incidence varies among countries and accounts for 27% (in some countries, more than 50%) of direct obstetric maternal deaths, mainly in the postpartum period. Recognizing risk factors for PPH in prenatal care and during childbirth care is the rst stage to prevent maternal death from PPH. The objective this review is: To identify the risk factors for hemorrhage in the third stage of labor described in the literature from 2000 to 2020.


Abstract Background
Postpartum hemorrhage (PPH) is the most common form of obstetric hemorrhage. This is the main cause of maternal death around the world: the incidence varies among countries and accounts for 27% (in some countries, more than 50%) of direct obstetric maternal deaths, mainly in the postpartum period. Recognizing risk factors for PPH in prenatal care and during childbirth care is the rst stage to prevent maternal death from PPH. The objective this review is: To identify the risk factors for hemorrhage in the third stage of labor described in the literature from 2000 to 2020.

Discussion
Prevention and control of hemorrhage must be initiated in the prenatal period, requiring competent professionals to carry out the appropriate clinical evaluation to classify the degree of risk to which the woman is exposed. This systematic review will support the studies of professionals who working in Angola and Brazil. The reason for maternal mortality and its causes is an indicator of the living conditions and health care of a population; since almost all deaths are preventable and mostly take place in developing countries [1].
Postpartum hemorrhage (PPH) is one of the main causes of maternal morbidity and mortality worldwide. PPH is de ned as blood loss above 500 ml, measured up to 24 hours postpartum, while this amount of blood loss after 24 hours is de ned as secondary PPH. Blood loss up to 500ml among healthy women does not lead to negative consequences; however, uncontrolled blood loss over 500ml can be fatal [2,3].
Hemorrhage is the greatest preventable cause of maternal death around the world. In Brazil, in the period from 1997 to 2009, maternal mortality due to hemorrhage was associated with the following causes: ectopic pregnancy; early pregnancy bleeding; malformations of the placenta; placenta previa; abruptio placentae; antepartum hemorrhage; intrapartum hemorrhage; postpartum hemorrhage. It was considered important to explore the causes of hemorrhage, because hemorrhage is merely a symptom of a disease. Therefore, it is the underlying cause, such as uterine atony or abruptio placentae, which will present with hemorrhage, but each of these conditions has a unique etiology. The risk of hemorrhage increases in cases of multiple pregnancies, polyhydramnios, macrosomia, precipitous labor or prolonged labor, chorioamnionitis, or simply the inability to contract the uterine muscle, due to the use of tocolytics or general anesthesia. This is also important because, in order to prevent, control, or intervene effectively in an obstetrical hemorrhage, it is necessary to know the underlying cause [4,5].
Although it is a physiological process, pregnancy, depending on the woman's life history, can reach varying degrees of risk, especially when there is a juxtaposition of several factors such as age, adverse events in previous pregnancies, multiple gestation, polyhydramnios, fetal macrosomia, great-grand multiparity, obesity and hypertension. The severity can also be expanded by the assistance that is provided to women, whether in the prenatal, birth and postpartum periods, associated with the infrastructure of the services, the quali cation of professionals, the distances and the access to intensive care units [5,6].
Maternal deaths are mostly preventable, constituting markers of a Nation's socioeconomic development level. Most of these deaths could be prevented by improving access, in a timely manner, to quali ed health services. Although the Maternal Mortality Ratio varies from country to country, the causes, in general, are mainly: hemorrhage (25% of maternal deaths), sepsis (15%), abortion complications (13%), eclampsia (12%) and obstructed labor (8.0%); and can be associated with each other [7,8].
Other research reported that it is di cult to understand why maternal mortality, most of the causes of which are preventable, does not receive adequate attention from health professionals, politicians and policy makers. Among the professionals, it is worth mentioning the obstetricians, who have been particularly negligent in relation to the reduction of maternal mortality. It is even observed that most obstetricians are concentrated in subspecialties that place emphasis on high technology [9]. Therefore, the theme of this Systematic Review (SR) is current, since identifying the risks will be contribute to actions that result in the reduction of maternal deaths. The reduction of maternal deaths was assumed as a global commitment, guided by the 2030 Agenda, in the context of the Sustainable Development Goals [10].
Carrying out a SR of primary studies on the risk for hemorrhage in the third stage of labor, described in the literature from 2000 to 2020, provides the analysis of risk factors that achieve greater relevance and the application of prevention and control measures that are feasible to put into practice. The obtained results will serve as subsidies for the education of professionals, as a support for the planning of clinical guidelines, as well as for the rede nition of public policies, so that care actions are offered equitably and safely to women. The risk factors make it possible to determine which women are at risk of postpartum hemorrhage also to propose guidelines for health services to qualify assistance.
The risk factors that obtain more frequency will be applied for validation to the reality of Huambo/Angola

Review question
This protocol for a Systematic Review and Meta-Analysis was developed with support from the PRISMA-P. The report of the review will be held in line with the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [11,12,13].
The research question for conducting the review was structured according to the mnemonic PEOS (P -Populations: Women aged between 10 and 49 years, in labor; E -Exposure (s): risk factors for hemorrhage in the third stage of labor; O -Outcome: women with hemorrhage during birth and postpartum; S -Study design: observational studies (case-control; cohort). Therefore, studies classi ed as follows will be included in this review: a) regarding the type of participants, those that address women in labor and birth; b) regarding the exposure, those that address risk factors for postpartum hemorrhage/third period of labor; c) regarding the outcome of interest, women who had hemorrhage during labor; postpartum/third period of labor, survivors or who died from this cause; d) regarding the type of studies, observational studies will be selected, including prospective and retrospective cohort studies, as well as case control [14,15,16]. Accordingly, the question was de ned: What are the risk factors for hemorrhage in the third stage of labor described in the literature from 2000 to 2020?
Search strategy Controlled descriptors in English, Portuguese and Spanish were used, in the plural and singular possibilities for each descriptor, in addition to recovering their synonyms. To this end, terms indexed in the Medical Subject Headings (MeSH) were retrieved as "MeSH terms" and "All Fields", also in the Descritores em Ciências da Saúde (DeCS) as descriptors and synonyms. Therefore, the following Search Strategy Medline (Table 01) was established. In addition, the Boolean operators AND and OR were adopted in the search strategies. Full texts of primary studies, written in Portuguese, English or Spanish will be identi ed and retrieved by screening titles and abstracts.
In addition, two independent researchers will select the articles, reading each of these articles by title, by summary and by the full text, according to the following inclusion criteria: primary scienti c articles, published between 2000 and 2020, in English, Portuguese or Spanish that address the Risk Factors for Hemorrhage in the Third Period of Labor in observational analytical studies: cohort studies, case control, available in the selected databases. The following exclusion criteria will be considered: theses, dissertations, editorials, integrative and systematic reviews and qualitative studies.

Process and tool for obtaining data
The collected data will be exported to a reference manager, EndNote. The search results in the databases will be exported to Rayyan®, an electronic tool that allows independent evaluation of studies by reviewers. Through it, studies indexed in more than one database will also be identi ed, and then duplicate publications will be removed [17]. The titles and abstracts of articles will be analyzed by two reviewers, independently, according to the eligibility criteria.
In order to assess the level of agreement between the reviewers, the Kappa coe cient will be calculated to obtain the proportion of agreement between the researchers, removing those given by chance [18]. The following classi cation will be adopted: <0.00, poor agreement; 0.00-0.20, slight agreement; 0.21-0.40, reasonable agreement; 0.41-0.60, moderate agreement; 0.61-0.80, substantial agreement; and 0.81-1.00, almost perfect [19].
Articles that do not reach consensus between the two reviewers will be assessed by a third reviewer for inclusion or disposal. The selected articles will be integrated into this SR for full reading and extracting data.

Grading of evidence
The level of evidence identi ed in the analyzed articles will be classi ed according to the GRADE [20,21], a system considered sensitive for grading the quality of evidence. In this system, the quality of evidence is describe in four levels: high, moderate, low and very low. Evidence from randomized clinical trials starts with a high level and evidence from observational studies with a low level.

Data extraction
Only published data will be extracted: article characterization, method design, intervention, outcome, results, statistic test and conclusions. Moreover, a peer-analysis of the extracted data will then be performed. The extracted data will be systematized by creating tables.

Risk of bias assessment
The GRADE criteria will be applied for the risk of bias assessments and will be carried out individually by two independent researchers by using the online tool GRADEpro GDT. The evidence will be classify as high, moderate, low and very low. The main conclusions drawn from the synthesis of the included studies will be given based on the criteria that lower the level of evidence, those being: risk of bias, inconsistency, indirect evidence, inaccuracy and publication bias, a large magnitude of effect, the dose-response gradient and residual confounders. In addition, it will be used, independently, assessed risk of bias by using the Newcastle-Ottawa Scale for evaluating the quality of nonrandomized studies [21,22]. Three factors will be considered to score the quality of included studies: (1) selection, including representativeness of the exposed, selection of the non-exposed, ascertainment of exposure and demonstration that at the start of the study the outcome of interest was not present; (2) comparability, assessed on the basis of study design and analysis, and whether any confounding variables were adjusted for; and (3) outcomes. We will rate the quality of the studies in three levels (good, fair and poor) by awarding stars in each domain following the guidelines.
In this review, based on the adopted classi cation to assess the quality of the evidence, the risk of bias in observational studies -cohort studies and case control studies will be considered in relation to the limitations identi ed in the description of the method referring to the design or execution of individual studies. Evidence from observational studies can be elevated by allocation con dentiality, blinding, complete segment, outcome reporting and su cient information to assess whether there is a risk of bias. For each of these domains, the risk of bias will be assessed and classi ed as high risk, uncertain and low risk.
In the third stage, after reading the full texts, the data will be systematized for all studies included in the analysis by reconciling the results between the independent researchers.
In order to compile and synthesize the results of the different studies included in the review, tables, charts and gures will be designed to compose the presentation of the results.
Strategy for data synthesis Narrative synthesis will be made with meta-analysis and will be carried out by using a framework that consists of: 1. Developing a preliminary synthesis of ndings of included studies 2. Exploring relationships within and between studies 3. Assessing the robustness of the synthesis. If possible, a statistical meta-analysis will be carried out by using Review Manager software (Revman The Cochrane Collaboration, Oxford, UK). In addition, a random-effect meta-analysis will be used for combining data where appropriate, at certain time points and by the outcome [23].

Discussion
Despite initiatives to reduce the maternal mortality risk in many places around the world, this problem is very important. The hemorrhage is the principal cause of maternal mortality when the health service not is good and the women don't receive the proper care [4,8,9].
The strengths of this review will be the use of guidelines such as PRISMA to support a standardized approach to review methods, well-de ned inclusion and exclusion criteria. The quality of the evidence of all studies registered in the consulted databases will be evaluated with the support of the GRADE, with the online tool -GRADEpro GDT. In addition, you will be using other computerized supports: EndNote, Rayyan®, Review Manager and Newcastle-Ottawa Scale. There are no competing interests to report.

Funding
Not applicable

Authors' contributions
All authors participated in the study meetings for the planning of this protocol and the formulation of this manuscript.
MLS, AEFC, and EPT conceived the study, developed the criteria and searched the literature, and wrote and register the protocol in PROSPERO. ACRHS, RCTR, IAPC, CMLB, HLC assisted in protocol design, study the support of the literature for suggest tools and performed data analysis. DRB, JFN study references to write the introduction of this protocol. MLS, ZBS reviewed all the authors' suggestions and the journal's norms to compose the manuscript on the protocol. Reading and nal review of the manuscript and submission to BMC. All authors read and approved the nal manuscript.