Maternal death is a worldwide public health problem, especially in developing countries where the maternal mortality rate remains very high. According to the World Health Organization (WHO), 303,000 women died worldwide in 2015 due to problems related to pregnancy or childbirth. Approximately 99% of these maternal deaths occurred in developing countries [1] [2] . In sub-Saharan Africa, the situation is alarming, where 66% of global maternal deaths occur and the probability that a young woman will one day die from a cause related to pregnancy or childbirth is 1 in 36 compared to 1 in 4,900 in developed countries [1] [3] . These high rates can be explained by several factors, including hemorrhages, high blood pressure and its complications, infections, dystocia, and complicated abortions. In Côte d’Ivoire, the maternal mortality rate remains high; in fact, according to DHS 2021, the maternal mortality ratio is estimated at 385 deaths per 100,000 live births [4] . Among the obstetric complications involved in maternal deaths, uterine rupture (UR) is certainly one of the most serious, as it immediately jeopardizes maternal and fetal life, constituting a common obstetric drama in our delivery rooms in countries with limited resources, demonstrating poor quality obstetric care and therefore constituting an unmet obstetric need [5] . Uterine rupture, with the hemorrhage it causes, is the main cause of maternal and fetal death in developing countries [6] . Nowadays, uterine rupture is exceptional in developed countries. The reported incidence of uterine rupture was 12 out of 36,000 births in developed countries and these occurred mainly in patients with healed uteri [7] . In developing countries, a maternal mortality rate of 11.26% and fetal mortality of 100% were noted in a study conducted from 1996 to 2001 at the University Hospital of Bouaké [8] . In the Republic of Côte d’Ivoire in general and particularly in Bouaké, we note little epidemiological data on the factors associated with maternal mortality during this very serious and much-feared obstetric pathology, hence the importance of this study.

This was a cross-sectional study with a descriptive and analytical aim with prospective data collection over a period of four years from January 1, 2020 to December 31, 2023 at the University Hospital (UH) of Bouaké, in the Obstetrics and Gynecology Department. All patients with uterine rupture diagnosed before, during labour or in the post-partum period during the study period and managed in the department were included. Not included were patients with uterine perforation, cervical tear, patients who arrived dead from uterine rupture, patients operated on for rupture and admitted for management of complications. All cases of uterine rupture were confirmed in the operating theatre during laparotomy. The data collection sources were the admission, delivery and operating room registers, the death register and then the patient files. The variables studied were the epidemiological characteristics (age, parity, ANC, etiological factors, origin of the patients), therapeutic aspects and factors associated with maternal deaths. The study was carried out with the authorization of the administrative authorities of the University Hospital of Bouaké. Confidentiality was respected by assigning an anonymous code to each survey form, so that no one could know the exact identity of the patients. Data entry and processing used Epi-Info7 software. Quantitative variables were expressed as means and extreme values and qualitative variables as proportions. The proportion analysis used the KHI2 or FISCHER test depending on the validity conditions. The ordds ratio (OR) was calculated and presented with its limits in the 95% confidence interval (95% CI) to assess the strength of the association between the variables studied and maternal mortality during uterine ruptures and the significance threshold was set at p < 0.05.

In order to contribute effectively to the reduction of maternal mortality in pregnant women with uterine rupture at the University Hospital of Bouaké, we conducted this prospective study in the obstetrics and gynaecology department. It enabled us to describe the sociodemographic and clinical characteristics of pregnant women, to determine the prevalence of uterine rupture and to identify the factors associated with mortality during uterine rupture. The limitations of this study could be information bias due to the method of collecting data from the respondents using a self-administered questionnaire and the impossibility of verifying the information provided.

The prevalence of 0.63%, corresponding to 6 uterine ruptures for 10 deliveries. Our result is close to that of Huyghe et al. [9] in the Central African Republic who noted 0.6%. On the other hand, our result is higher than that of Gueye et al. [10] who found a prevalence of 0.2%, and lower than those of Fané et al. in Mali [11] and Imen et al. [12] in Tunisia who noted 3.7% and 2.173% respectively. In developed countries, UR is exceptional. Sheng et al. [13] in China noted a prevalence of 0.02%. The disparity between industrialized and developing countries can be explained, on the one hand, by limited access to health facilities and, on the other hand, by the poor organization of the referral and evacuation system, as well as the insufficient quality of care, which are particularly notable in Africa.

The patients were young with an average age of 32 years +/− 6. Patients aged 35 years and over accounted for a third (33.68%) of cases. Our result is close to that of Fané et al. in Mali [11] who noted an average age of 30 years with patients aged over 35 years representing 44.2% of cases. Also Kitenge et al. [14] in the Democratic Republic and Belinga et al [15] in Cameroon noted a similar result with an average age of 29.5 years and 30.98 years respectively. This result could be explained by the fact that this age corresponds to the period when reproductive activity is intense. According to Hochler et al [16] , women aged 35 years or overrun almost 3 times the risk of uterine rupture. Pauciparous and multiparous women represented 41.05% and 40.0% respectively. This high frequency of pauciparous and multiparous women can be explained by the early marriage in Africa and the high multiparity of often young women [5] . 21.58% of cases of uterine rupture occurred on a scarred uterus. This result is similar to that of Kitenge et al. [14] who noted a history of cesarean section in 20.9% of cases. Indeed, the uterine scar is the weak link of the myometrium, which, subjected to the stress of contractions, can suddenly rupture. However, in our study, more than 3/4 of uterine ruptures occurred on healthy uteri. This could be explained by the use of traditional oxytocic during labor. Indeed, in our study, the onset of labor was spontaneous, but accelerated by the use of traditional oxytocics in more than half (55.79%) of the cases, unlike the studies of Kitenge et al. [14] and Huyghe et al. [12] who noted that the onset of labor was spontaneous in 90.7% and 75% of the cases respectively. The indiscriminate use of traditional oxytocics to accelerate labour is a common practice in our context, with harmful consequences for the health of mothers and newborns. These products are used without any study of the dose or route of administration. The use of these products, combined with other risk factors such as abdominal pressure at full dilation in the event of insufficient expulsive effort, trauma following certain obstetric procedures and delayed diagnosis of certain dystocies, could explain the frequency of spontaneous ruptures. The patients had performed less than 4 ANCs in 78.42% of the cases. For Fane et al. [11] and Gueye et al. [10] the patients had not performed ANC in 36.06% and 30.8% of the cases respectively. In the study of Huyghe et al. [9] , no difference was noted between non-participation in antenatal consultations and the occurrence of UR. However, ANC should be encouraged, given that they constitute opportunities for early screening of UR risk factors. The diagnosis of uterine rupture was made per partum in 97.89% of cases. This result is close to that of Imen et al [12] , Huyghe et al. [9] and Kitenge et al. [14] who noted respectively 72.7%, 78.1% and 100% of cases. The lesion was complete in 100% of cases with the location of the lesion being segmento-corporeal in 74.21% of cases. Fané et al. [11] noted 79.1% complete UR; in 51.59% the lesion was located at the level of the lower segment and segmento-corporeal in 32.22% of cases. Kitenge et al. [14] , as well as Imen et al. [12] also noted complete ruptures in respectively 79.1% and 80% of cases. The higher frequency of complete uterine ruptures in our study would be related to the delay in admission of patients who for the most part stay several hours, even several days, by unqualified personnel ignoring the contraindications of vaginal delivery and abusing the use of oxytocics. Surgical treatment was conservative in 77.89% of cases. However, there was a high rate of hysterectomy in 21.05% of cases. The rate of hemostasis hysterectomies varies in the literature in cases of complete ruptures. Kitenge et al. [14] in their study noted conservative treatment in 95.3% of cases and hysterectomy in 4.6% of cases. For Huyghe et al. [9] , surgical treatment was conservative in 93.33% of cases and hysterectomy in 6.67% of cases. The high rate of hysterectomy in our context could be explained by the extent of uterine lesions (100% complete UR) as well as the occurrence of major complications such as hemorrhagic shock (20.62%) and coagulopathy in 20.62% of cases.

Our maternal death rate of 15.79% is similar to that reported by Baldé et al. [5] , who noted a rate of 14.6%. This high rate could be explained by the late evacuation of patients and the precarious state of patients upon admission. As for the fetal prognosis, it was poor with a stillbirth rate of 89.47% similar to that reported in the African literature [9] - [11] . Factors statistically associated with death were age ≥ 35 years (OR = 3.14), duration of labor ≥ 12 hours (OR = 5.8), multiparity (OR = 19.04), admission time beyond 2 h (OR = 4.36), hemoglobin level ≤ 7 g/dl (OR = 36.84), coma or obtundation on admission (OR = 71.82), shock on admission (OR = 243.94) and the occurrence of postoperative complications (OR = 76.45). These complications were hemorrhagic shock (OR = 118.5) and coagulopathy (OR = 31). According to Belinga et al. [15] , when the arrival time was more than 4 hours after the referral, it multiplied by 20.8 the risk of death by uterine rupture. In addition, compared to survivors, patients who died during uterine ruptures were between 11 and 20 times more likely to have presented agitation and/or obtundation [15] . Also, shock and coagulopathy multiplied by 12 and 162 times respectively the probability of dying during uterine rupture [15] .

The frequency of uterine ruptures remains significant in the department (0.63%) with maternal mortality still high (15.79%). The factors statistically associated with death were age ≥ 35 years, duration of labor ≥ 12 hours, multiparity, admission time beyond 2 hours, coma or obtundation on admission, shock on admission and the occurrence of postoperative complications.

Almost all of these uterine ruptures and complications could be avoided by: raising awareness among pregnant women, improving the management of obstetric emergencies, increasing access to health facilities and ensuring that gynaecologists and obstetricians are properly assigned throughout the country.

All authors participated intellectually in the preparation and revision of the manuscript before its submission.