Determinants of Maternal Mortality at the Community University Hospital of Bangui: Central African Republic

Introduction: Pregnancy is a physiological condition that can sometimes end in death. The death of a woman is a constant concern for the obstetrician and is considered a major public health problem in our developing countries. Objective: To identify the determinants that contribute to maternal morbidity and mortality in the Gynaecology and Obstetrics Department of the University Community Hospital. Methodology: We conducted a retrospective study from 1 January 2015 to 31 December 2019 (5 years) in the obstetrics and gynaecology department of the Centre Hospitalier Universitaire Com-munautaire (CHUC). It focused on the determinants of maternal mortality. Our study population consisted of all women who died during the period and met the World Health Organization (WHO) definition of maternal mortality (MOMA). We used non-probability sampling with the exhaustive choice technique. Results: The study


Introduction
The death of a woman during pregnancy, childbirth and the postnatal period is a major public health problem.Worldwide, an estimated 1500 women die every day from complications related to pregnancy or childbirth [1].Most (99%) of these deaths occur in low-income countries, half of them in sub-Saharan Africa and nearly a third in South Asia [2] [3].The situation in the Central African Republic (CAR) is very worrying.The maternal death rate in 2003 was 1355 per 100,000 live births (LBW), which corresponds to 1 maternal death every 5 hours [4] [5].In response to this problem, in 2011 the country drew up a strategic plan entitled "Strategies to reduce maternal mortality".Implementation of this plan has reduced the maternal death rate by 882 per 100,000 live births [6].With this in mind, we conducted this study to identify the determinants that contribute to maternal morbidity and mortality in the gynaecology and obstetrics department of the Centre Hospitalier Universitaire Communautaire (CHUC).And in specific ways: -Determine the maternal mortality ratio at the CHUC.
-Describe the socio-economic and demographic characteristics of pregnant women and women who have given birth who are victims of maternal death.
-To study the circumstances in which maternal deaths occur.

Patients and Method
We conducted a retrospective study from 1st January 2017 to 31 December 2021, over a period of 5 years in the obstetrics and gynaecology department of the CHUC.
The study population consisted of the records of women who died in the department during the period of our study.According to the World Health Organisation, maternal mortality is defined as the death of a woman during pregnancy or within 42 days of the end of pregnancy, irrespective of the duration or location of the pregnancy, from any cause determined or aggravated by the pregnancy or the care it required, but neither accidental nor fortuitous [7].
All women who died in the department during pregnancy, childbirth and the post-partum period up to 42 days after delivery were included in the study.
Women who died while not pregnant were not included, as were those who died outside the post-natal period and those whose medical records were not available.
We chose non-probability sampling using the exhaustive selection technique.
A pre-established questionnaire containing the parameters studied was used as a data collection tool.The data sources were patient files and registers found in the operating theatre, the obstetric emergency unit and the pathological postnatal unit.
The parameters studied included socio-demographic and economic characteristics, antecedents, causes of death and factors influencing maternal death.
After quality control, the data were immediately entered using Epi-info software version 7.2.1.0and recorded and processed.The results were analysed to calculate the mean and percentage.Relationships were studied using the chisquare test.A p value < 0.05 is statistically significant.
Ethics: A study protocol was validated by the ethics committee of the University of Bangui before the study was carried out.

Results
During the period of our study, we recorded 98 maternal deaths in the Gynaecology and Obstetrics Department of the CHUC.During the same period, 17,172 live births were recorded, giving a maternal mortality ratio of 570.70 per 100,000 live births.
The evolution of maternal mortality ratios over the last 5 years shows that after a significant decrease from 2018 to 2020, there is a slight increase in 2021.
The median age was 26 years, with extremes ranging from 16 to 41 years (Table 1).
Primiparity was associated with the occurrence of direct causes, with a statistically significant difference.
Age less than or equal to 25 years was associated with the occurrence of direct causes (Table 4) and abortion complications (Table 5), with statistically significant differences.
Patients who died of direct causes often died on immediate admission to hospital, with a statistically significant difference.

Discussion
Our study focused on the determinants of maternal mortality in the Gynaecology and Obstetrics Department of the Centre Hospitalier Universitaire Communautaire, which is a referral centre.A study of the records of patients who died during the study period enabled risk factors and common causes to be identified, with a view to identifying certain measures that could help reduce maternal mortality.
The Maternal Mortality Ratio (MMR) in our series was 570 per 100,000 live births.Changes in MMR over the last 5 years show that after a significant drop from 2018 to 2020 (987 to 250), there is a slight increase in 2021 (250 to 400).
This result is below the national rate but comparable to that reported by Mbaye et al. in Senegal, with a rate of 592/100,000 live births over 7 months [8].This rate is also comparable to that of other countries in the sub-region (500 to 999 per 100,000 live births) [9], although it is much higher than that of developed countries (9 per 100,000 live births) [10].
In our study, young girls aged under 25 were the most represented.The average age was 26, virtually the same as that found by Alida et al. in 2016 in the same department [11].The same trend was found by Kané in Mali who found an age range of 14 to 25 years [12] and Baldé with an age range of 14 to 16 years [13].On the other hand, Thiam et al. in Senegal reported ages ranging from 20 to 39 years, with an average age of 28.4 years [14].
Analysis in this study showed that age less than or equal to 25 years was associated with maternal causes of death, with a statistically significant difference (Chi2 = 5.9979; p < 0.014).Similarly, complications of abortion occur much more frequently in young women aged 25 or under, with a statistically significant difference (Chi2 = 5.5549; p < 0.018).This high maternal mortality rate among young girls is explained by the fact that the contraceptive prevalence rate among young people remains low (12%) [15], which more often than not exposes them to unwanted pregnancies with their attendant complications.A study conducted by Sépou in Bangui found that girls began having sex at the age of 14 [16] [17].This means that adolescent girls are often exposed to the risk of morbidity and mortality.
Primiparity and multiparity play an important role in the causes of maternal death [18].According to some authors, nulliparity and multiparity are a risk factor for maternal death: Friday E. et al. in Nigeria 31% and Diakariadia in Mali 24.24% [19] [20].As pointed out by another author, the two extreme parities are mainly affected, 11.8 per thousand for primiparous women and 16.7 per thousand for large multiparous women [21].Our series is dominated by primiparous women (29.59%) followed by poor women (24.48%).The analysis also showed that primiparity was associated with the causes of maternal death, with a statistically significant difference (Chi2 = 10.0855;p < 0.00014).Anki Yambare in Congo found a similar result with 40% of primiparous women [22].Baldé reports 67.3% of nulliparous women, whereas Diakariadia reports 24.4% of large multiparous women.Although women of extreme parity (primiparous and multiparous) are more represented, we believe that maternal mortality does not spare any parity category.
The antenatal consultation is an important part of pregnancy monitoring, as it enables anomalies to be detected and treated.It is also the ideal time to carry out certain preventive measures (screening for risk factors, anti-tetanus vaccination, etc.) and to educate women about the progress of pregnancy and childbirth [18].The new WHO recommendations recommend eight (8) contacts for better pregnancy monitoring [23].In our study, 53.06% of the women who died did not attend antenatal clinics.This result is comparable to those of Diakaradia (63.6%) and Baldé (57.7%) [13] [19].In contrast, Thiam et al. found that 75% of women attended antenatal clinics [14].This difference may be explained by the low level of education and unfavourable socio-economic conditions in the Central African Republic.
Direct causes were dominated by complications following unsafe abortions (31.34%), followed by puerperal infections (17.91%) and postpartum haemorrhage (16.41%).Indirect causes, on the other hand, were dominated by anaemia (45.16%).The increasing incidence of complications in high-risk abortions has been reported in some studies [11] [27].However, Fomulu found haemorrhage, eclampsia and infection in order of frequency [25].Baldé found that 32.6% of women died as a result of eclampsia [13].This difference may be explained by the young age of our patients, the early onset of sexual relations and the high frequency of clandestine abortions [16] [17].The majority of deaths occurred after childbirth (34.61%), followed by abortion (30.60%).This result is comparable to that of Bohousou et al. in Côte d'Ivoire (58.5% [28]) and Baldé in Mali (82.7% [13]).This high rate of death after childbirth could be explained by poor monitoring of newborns, leading to late diagnosis of postpartum haemorrhage, which is responsible for a large number of maternal deaths.
The referral and counter-referral system is a pillar in the fight against maternal mortality.In our study, most of the women who died had come from home without having been to a health facility.These women often come late because the decision to come to a health facility is often taken late (type 1 delay).Moreover, Sépou et al. [29] demonstrated in a study that late evacuation contributes to maintaining the maternal mortality rate at a high level.In our series, the referral was not related to the cause of death.

Conclusion
At the end of this study, complications of unsafe abortion and poverty are all factors contributing to the rise in the maternal mortality rate.Emphasis should be placed on promoting family planning, as this would make a major contribution to reducing maternal mortality.

Table 1 .
Distribution of deceased according to age group.

Table 2 .
Frequency of deaths according to gesture and parity.

Table 3 .
Frequency of deceased according to time of death.

Table 4 .
Distribution of the deceased according to age in relation to the causes.

Table 5 .
Distribution of deceased by age in relation to direct causes (n = 67).