This is a cross-sectional prospective, descriptive and analytical study that took place over a period of thirty-six (36) months from the 1^st of January 2019 to the 31^st of December 2021. It took place in the Obstetrics and Gynaecology department of the Bouaké University Teaching Hospital. The city of Bouaké is located in the centre of Ivory Coast and is the second most populous city after Abidjan, with a general population estimated at 1,542,000 according to the MICS 2021 [6] . The city has a hospital and university centre, which is a tertiary-level hospital according to the country’s health pyramid. This centre has an Obstetrics and Gynaecology department, which provides medical and surgical care for gynaecological and obstetric pathologies. It is a referral service, receiving patients referred from general and peripheral hospitals in the town of Bouaké, as well as from surrounding towns in the centre, north and west of the country. The maternity ward in the Gynaecology and Obstetrics department has 4 delivery cubicles, 6 post-natal beds and an operating theatre with two operating theatres. The daily team consists of a Gynaecologist Obstetrician, three medical doctors with a diploma in gynaecology and obstetrics, four midwives and two nurses.

The study population consisted of pregnant women, parturients and newborns admitted to the department. Sampling was exhaustive. All pregnant women and recent mothers admitted to the department presenting with a tonic-clonic seizure associated with systolic blood pressure greater than or equal to 140 mmHg and/or diastolic blood pressure greater than or equal to 90 mmHg associated with high proteinuria (greater than 0.30 g/24h) were included in the study. We did not include women with other causes of seizures during pregnancy not associated with the inclusion criteria, patients who had died and patients with hypertension without significant albuminuria.

The data sources were prenatal consultation books, delivery records and hospitalization records. The questionnaire was developed after reviewing different literatures. The variables studied were sociodemographic characteristics, the course of pregnancy and delivery, and neonatal parameters. The data were entered into a computer and analysed using EPI INFO software version 7.2.2.6 with the use of Chi-squared statistical tests for numbers ≥ 5 and Fisher for numbers < 5 with α = 5%. The significant variables were of value less than p < 0.05. Descriptive statistics were used to describe the characteristics of the study respondents by using means and standard deviations for numerical variables, frequencies along with percentages for categorical variables and table.

The prevalence of eclampsia observed in our study was 1.14%. This prevalence is higher than that observed by Vousden et al. in their study conducted in Uganda in 2019, which was 0.3% [7] . This is similar to that of Abdoul Aziz Diouf et al. in Senegal, who found a prevalence of 1.75% [4] , but lower than that of Horo et al. in Abidjan, who found a prevalence of 3.4% [8] . The high prevalence in our study can be explained by the fact that it was conducted in a referral centre that receives all cases of complicated pregnancy. The patients evacuated represented 74.27% of our study. In addition, the low rate of ANC coverage observed inwomen with eclampsiacould explain this prevalence. In our study, 64.31% of these women had fewer than or equal to 3ANCs, including 3.73% who had not undergone ANC. It has been clearly established in the literature that good-quality ANC improves maternal health by reducing the risk of anaemia, pregnancy-induced hypertensive disorders and preterm labour and improves the prognosis of the pregnancy.

The mean age of the population was 22 ± 7 years (range from 13 to 47 years). The age group ≤ 19 years represented 45.64%. These results are consistent with previous epidemiological data in the literature. Several studies carried out in different countries have confirmed our observation, in particular those reported by Koffi and Bonkoungou [9] [10] . For parity, 49.8% of the women in our study were nulliparous. The occurrence of eclampsia in young nulliparous women is linked to certain factors, such as the failure of the maternal body to adapt to haemodynamic and renal changes and uterine hypoplasia, which is common in young pregnant women. During the study, 97.93% of patients had a low socioeconomic level. This observation had already been made by several studies in Africa, notably that of Bohoussou M. [11] on maternal mortality in Abidjan and Chalumeau M. [12] on maternal health in French-speaking Africa. Nevertheless, we believe that the lack of financial resources may indirectly influence the increase in the percentage of unrecognized preeclampsia, and then eclampsia, by creating the conditions for poor pregnancy monitoring associated with a weakened body. The 3rd edition of the Demographic and Health Survey (EDS) in the Ivory Coast [13] showed that 46.9% of women were single, compared with 27.2% who were married. The same observation was made in our study, which found that 54.77% of women were single.

Seizures occurred in both the antepartum (56.84%) and postpartum (40%) periods. Coulibaly KT et al. [14] and Samake et al. [15] found a predominance in the antepartum period (59.4% and 69%, respectively). Our results can be explained by the fact that the seizures that occurred most frequently in theante and postpartum periods are linked to the non-availability of magnesium sulfate for almost the entire duration of our study. Gestational age greater than 36 weeks accounted for 62.2% of our respondents, which corroborates the data in the literature. This trend was similar to those reported by Mamadou Ibrahima Kampo et al. in Senegal and Abdoul Aziz Diouf, who found 57% and 59.7%, respectively [4] [16] . These results can be explained by the fact that eclampsia most often occurs at the end of pregnancy, with a better fetal prognosis if the fetus is extracted quickly. The preferred mode of delivery in our study was caesarean section, with a percentage of approximately 64.7%. Our results are similar to those of Mamadou Ibrahima Kampo et al. and Abdoul Aziz Diouf et al., who found 77.6% and 54.8%, respectively [4] [16] . The high caesarean section rate in our series could be explained by the fact that the majority of patients were at term and had convulsed in the antepartum period, and rapid extraction is an integral part of the treatment of eclampsia. In addition, delayed uterine evacuation is accompanied by complications that can jeopardize the mother’s prognosis, which are difficult to manage due to our inadequate technical facilities.

Maternal lethality varies between 14% and 15.8% in developing countries [17] . In the studies by Mamadou Ibrahim Kampo et al. and Tchaou et al., 4.3% and 11.1%, respectively, died [16] [18] . Our case fatality rate (7.88%) is similar to that of African studies. This high case-fatality rate could be explained by the seriousness of complications such as HELLP syndrome (45%), which makes patient management difficult by clouding the prognosis and delaying admission. In addition, the delay in management and the delay in starting treatment are influenced by the late purchase of medicines by parents and the inadequacy of the technical facilities in intensive care. The factors associated with maternal lethality were parity of less than 3 and evacuation. However, Mamadou Ibrahima Kampo et al. did not find a statistically significant association between maternal lethality and parity and mode of admission but rather with the Glasgow score ≤ 8 on admission [16] . The most frequent complication was HELLP syndrome (45%). As reported in the literature, HELLP syndrome is a complication of pregnancy affecting 2% - 30% of women with preeclampsia [17] .

The neonatal case fatality rate was high in our study at 18.95%. This corroborates the data in the literature, which mentions that the foetal outcome was less favourable than the maternal outcome. This was the case in the series by Beye et al., who reported 42.8% [19] . The same result was observed in the study of Mamadou Ibrahima Kampo et al., with a percentage of 21.5% [16] . In our study, we noted a statistically significant association between neonatal lethality and prematurity. The same observation was made in the study by Abdoul Aziz Diallo et al. [4] , who found a statistically significant association between lethality and prematurity.