Commonly defined as loss of intra-amniotic fluid (AF) through the maternal genital tract, premature rupture of membranes (PROM) results from an opening of the membranes of an egg at any time during pregnancy, before the onset of uterine contractions [1, 2]. Differentiating between the two types of PROM is possible, depending on the moment of onset during pregnancy: the preterm premature rupture of membranes (PPROM) occurs before 37 weeks of gestation (WG), whereas a term premature rupture of membranes (TPROM) occurs after 37 WG [1, 2]. The frequency of the condition across all pregnancies is estimated to be between 1.0% and 10% worldwide [2, 3, 4, 5, 6, 7], while in some African series the frequency ranges from 1% to 18% [8, 9, 10, 11]. The majority of PROM cases occur beyond 37 WG [12]. Moreover, PROM is rarely observed before 28 weeks, representing 1 to 7/1000 pregnancies; however, a neonatal prognosis is extremely critical due to prematurity, infection, and pulmonary hypoplasia [12]. In low- and middle-income countries (LMICs), PROM is not easily diagnosed due to the lack of diagnostic tests [12, 13, 14, 15]. Only clinical features can guide the diagnosis, often leading to errors, as these clinical elements are non-specific, and include the vulvar discharge aspects (spontaneous, abundant, permanent, clear, and odorless) as well as the visualization of a spontaneous or provoked liquid discharge of endocervical origin and/or in the posterior vaginal fornix. Furthermore, this diagnosis is often made when the amount of AF decreases on the ultrasound observation [16]; however, the possibility of false-negative results is noted, with an overestimation frequency of 12% [16, 17].

Efficient management is crucial in preventing serious maternal and perinatal complications. However, this management requires a multidisciplinary medical team collaboration, alongside an individualized approach that considers gestational age, cervical condition, and fetal status. Additionally, this management strategy must be a continuous search for a suitable compromise between potential infectious risks and those related to prematurity [8]. All these aspects are challenging to implement in LMICs, where many health centers have inadequate technical equipment (including the absence of neonatal intensive care units) and unqualified human resources for neonatal resuscitation [17, 18, 19]. Furthermore, PROM is an important cause of perinatal morbidity and mortality. The main neonatal risks of PPROM are respiratory distress syndrome, sepsis, intraventricular hemorrhage, and death, which are related to prematurity in 24% to 42% [20], and require management in a well-equipped tertiary preventive healthcare [21]. Expectant management refers to the decision to allow the pregnancy to continue before 37⁺⁰ WG. The main risks are ascending infection and umbilical cord compression [22]. Premature delivery must always occur in a level 3 maternity hospital with a neonatal intensive care unit (NICU) [2, 22]. However, LMICs face numerous difficulties that hinder the improvement of medical data [10, 23]. In fact, LMICs continue to struggle with inadequate healthcare infrastructure, insufficient technical equipment, and a shortage of well-qualified healthcare providers, as well as the high cost of medical care, the lack of health insurance, and the unavailability of some treatments, such as synthetic surfactant [2, 11, 12, 13]. All these factors make PROM management more difficult and further compromise early neonatal prognosis, with high morbidity and mortality rates [20, 24, 25, 26, 27]. A pregnant woman with PROM has a risk of intra-amniotic infection, postpartum infection, endometritis, and death [24, 28]. Neonatal outcomes after expectant management are well described [26, 27, 28]. In contrast, limited research has addressed maternal outcomes associated with expectant management compared to the termination of pregnancy. Expectant management for PPROM before 24 WG was associated with a significantly higher risk of maternal morbidity when compared to TPROM. Thus, the maternal prognosis is related to the possible occurrence of some maternal complications, particularly infectious complications (sepsis and septic shock), which may result in mortality in 1% of cases [24], and an increase in morbidity of 58 to 63% [29, 30, 31].

Due to the lack of standard diagnostic methods, limited therapeutic options, and the scarcity of studies on prognostic factors in LMICs, this study aimed to evaluate the management and identify maternal and neonatal risk factors in a reference maternity unit.

We conducted a cross-sectional study with an analytical focus in a single tertiary perinatal center, from January 2019 to December 2023 (five years). Singleton or twin pregnancies with PROM that occurred after 28 WG or 6 months, including both PPROM and TPROM, were included. Pregnant women who delivered within 12 h after the onset of PROM, suspected urinary incontinence or physiological leukorrhea, or those with major fetal anomalies or fetal death in utero were not included.

PROM diagnoses were suspected in the presence of a spontaneous vulvar discharge, clear, continuous, odorless, as well as the speculum visualization into the intracervical and/or vaginal fornix. This intracervical discharge was amplified by coughing, mobilization of the abdomen, and abdominal efforts in a pregnant woman with no uterine contractions or cervical changes. Due to the absence of diagnostic tests in Cote d’Ivoire, the PROM diagnosis was also suspected in the absence of fetal membranes on vaginal examination or the decrease in AF on ultrasound (amniotic fluid index [AFI] value of less than 25 cm). Pregnancies after 28 WG or 6 months, and before 37 WG, with clear spontaneous vulvar discharge without uterine contractions for more than 12 hours before the onset of labor, were also included in the study. We did not consider clear vulvar discharges that quickly dried up with normal AF on the ultrasound. An ultrasonography scan performed at admission evaluated fetal presentation and weight, as well as AFI values. The maternal infectious assessment included a C-reactive protein (CRP) (used to assess inflammatory parameters, which were considered normal at values below 5 mg/L, and positive above 10 mg/L), and a white blood cell count was recorded every 3 days. Pregnant women received an intramuscular (IM) injection of corticosteroid or betamethasone at 12 mg, repeated 24 h later, to prevent hyaline membrane disease, as well as a beta-lactamase-stable broad-spectrum cephalosporin: ceftriaxone 2 g intravenous (IV) until birth. Infant outcomes included gestational age, birth weight, Apgar score at 5 minutes, neonatal transfer, hospital length of stay, neonatal outcomes (death or survival), respiratory distress, prematurity, and neonatal infection. Data were obtained from the medical records, delivery registers, operative reports, and maternal and neonatal death reports for the pregnant women. An anonymous questionnaire was used to collect the data. We selected all files recovered during the study period based on the inclusion and exclusion criteria. The parameters studied were sociodemographic characteristics, history, age of pregnancy, admission data (time of rupture, presence or absence of uterine contractions, vaginal bleeding, uterine height, fetal heartbeat, fetal presentation, diagnosis retained), time from water break to admission, obstetric decision, time from opening of membranes to delivery, mode of delivery, Apgar score at 5 minutes of life, birth weight, resuscitation of newborn, evacuation to neonatology, maternal–fetal complications. Fetal growth restriction (FGR) was defined as an estimated fetal weight (EFW) that was less than 1.5 standard deviations. Chorioamnionitis was suspected in the presence of a maternal body temperature $>$38 °C, tachycardia $>$70 bpm, uterine contractions, fetid vaginal discharge, and hyperleukocytosis at 15,000 elements/mm³, as well as neutrophil counts $>$70%. The confidentiality of the information collected during the survey was guaranteed by assigning an anonymous number to each survey sheet. Data were analyzed using Epi Info software (Atlanta, GA, USA, version 3.5.1). Logistic regression analysis was used to identify maternal and neonatal prognostic factors associated with PROM. The statistical tests used were the chi-square and the Student’s t-test at a significance level of 5% or p 0.05. Results are presented as the mean $\pm$ standard deviation (SD). A p-value of less than 0.25 at bivariate and 0.05 during multivariate logistic regression at a 95% confidence interval (CI) was considered statistically significant.

PROM remains a major challenge in obstetric care in Cote d’Ivoire, as in many other sub-Saharan African LMICs [5, 8, 9, 10, 21]. The complications observed from maternal and fetal infections are most often exacerbated by delayed consultation after the onset of PROM. However, antibiotics combined with antenatal corticosteroid therapy have proven to be effective in managing these complications, highlighting the essential role of antibiotics in patient care. This study offers valuable insights into the specific challenges encountered in managing PROM, particularly in the resource-limited settings of LMICs. This study also highlights existing gaps in research on managing PROM in Africa. Our analysis also highlights the importance of these findings for future research, underscoring the urgent need to develop care protocols that are more effectively tailored to local realities. Meanwhile, these findings also highlight the need to strengthen training for healthcare professionals and improve access to care, particularly in rural and underserved urban areas.

The frequency of PROM was relatively low in our study (3.5%). Despite this low rate, the occurrence of PROM should not be underestimated due to the numerous consequences of this condition for the mother and child. The global prevalence of PROM is estimated to be between 1.0 and 10% [2, 3, 4, 5, 6, 7], while an African series reported rates ranging from 1% to 18% of pregnancies [5, 8, 9, 10, 21, 28]. Our results corroborate those found in African and international literature, but also reveal specificities unique to the Ivorian context, influenced by socioeconomic, logistical, and health factors. This difference can be attributed to several factors, including the possibility of cases being underreported in our facilities due to a lack of systematic monitoring of pregnancies or late clinical diagnosis, potential exclusion of the earliest cases, and patients who were unable to access hospital care, as well as a preference for traditional or community-based care in certain areas, which limits the recording of cases in hospitals. The etiology of the disease remains poorly understood [22, 32, 33, 34]; however, some authors have suggested risk factors such as low socioeconomic status, obstetric history of cervical incompetence, repeated spontaneous miscarriages, or previous PROM [1, 4, 7, 9, 27, 32]. In our study, although some of these factors were present, their statistical significance was not always evident, which could be attributed to the lack of systematic screening (genital infections, cervical incompetence) during pregnancy and to limitations in the collection of gynecological and obstetric historical data, due to the lack of comprehensive medical records and missing data.

A high number of neonatal deaths was observed (11.2%) in our study. The main identified risk factors for morbidity and mortality were a delay in membrane rupture of $>$72 hours (p 0.001) and a birth weight of 1500 g (p 0.001).

Some contextual conditions in Cote d’Ivoire may explain the increased severity of these outcomes, including a shortage of neonatal resuscitation equipment, particularly in peripheral areas or primary care hospitals, the scarcity of fully functional NICUs, and the lack of continuing education for healthcare personnel in neonatal resuscitation. Similar results have been reported in the literature from many LMICs in Africa [12, 13, 26, 27, 29], as well as in India [32]. Prolonged membrane rupture exposes the fetus to several risks, including pulmonary hypoplasia and hyaline membrane disease. The consequences after birth include respiratory distress that requires care in a NICU, which is not always available in many LMICs [26, 27]. Prolonged rupture also exposes the fetus to infections, which sometimes require neonatal resuscitation and prolonged hospitalization.

The benefits of antibiotic prophylaxis and corticosteroid therapy have been demonstrated in the literature [9, 18]. However, the underuse of these therapies in our study (correlated with neonatal deaths) reflects the high cost of these drugs, which are most often paid for by the patients themselves because the medical centers/countries do not have social security, their frequent unavailability in hospital pharmacies, and the lack of standardized protocols that are accessible and applied in all hospitals in Cote d’Ivoire.

Most neonatal deaths were significantly associated with vaginal delivery (p 0.001), absence of betamethasone administration to the mother (p 0.001), and a low birth weight of less than 1500 g (p 0.001). Similar observations have been reported in the literature for LMICs [1, 3, 11, 12] compared to high-resource countries [4, 8, 14]. This might be explained by delayed decisions in performing a cesarean section in cases of fetal distress, inadequate management of labor in premature births, and the frequent lack of continuous monitoring of the fetal heart rate. Thus, improvements in healthcare infrastructure and equipment, expanded medical coverage, and public awareness campaigns could help reduce these perinatal mortality rates.

Among the maternal complications observed were intra-amniotic infections (chorioamnionitis), postpartum endometritis, and an increased frequency of cesarean sections. These complications have been widely described in previous studies [1, 13, 19, 20, 29]. Meanwhile, their impact on maternal morbidity may be exacerbated in Cote d’Ivoire by the lack of rapid biological diagnosis (CRP, blood cultures) due to the low purchasing power of the population, delays in surgical referral, especially in peripheral areas, and inadequate postpartum follow-ups, particularly after early discharge.

Furthermore, cases of PROM occurring before the threshold of fetal viability (24 WG) pose a particular challenge in terms of maternal care. Studies by Dotters-Katz et al. [29] and Sklar et al. [30] reported high rates of maternal infection, sepsis, and prolonged hospitalizations in these situations, findings consistent with the high rates of early maternal complications observed in our study. These findings, observed in our series, highlight the unavailability of appropriate respiratory support and the lack of a clear national consensus on the management of extreme PROM in Cote d’Ivoire.

PROM is also associated with a higher rate of cesarean section, often due to fetal distress, non-cephalic presentation, or intrauterine infection [16, 22, 25]. While some guidelines recommend a wait-and-see approach to allow fetal maturation [15, 32], others recommend early delivery in cases of infection [13, 22]. Our data support the latter strategy, as the rate of cesarean sections was significantly higher in cases where signs of disease or fetal distress were observed.

Scholarly societies such as the American College of Obstetricians and Gynecologists (ACOG) and the Royal College of Obstetricians and Gynaecologists (RCOG) recommend individualized management strategies based on gestational age, clinical signs of infection, and fetal well-being [13, 22, 32]. However, the application of these strategies in Cote d’Ivoire is hindered by a lack of diagnostic tools (such as ultrasound and biological markers), a shortage of contextualized protocols available in health facilities, and unequal access to care between urban and rural areas.

Furthermore, in the context of other low-resource countries, implementing these recommendations remains difficult due to limited access to diagnostic tools and specialized obstetric care [9, 23, 28]. Therefore, maternal morbidity related to PROM remains a major concern in these settings. Finally, although certain biomarkers, such as CRP levels, have been proposed for the early diagnosis of intra-amniotic infections, the clinical utility of these biomarkers remains controversial [16]. In our context, the availability of tests to evaluate these biomarkers is often limited, which represents a potential area for improvement in maternal care.

PROM is common during pregnancy and remains a major obstetric emergency, associated with significant maternal and neonatal morbidity despite therapeutic advances. In our practice, we encounter numerous difficulties in management due to the high number of diagnostic errors, low levels of education, and limited income among families, as well as insufficient technical resources in our healthcare facilities. This study highlights gaps in the management of PROM, particularly in terms of late diagnosis and the lack of standardized protocols. These findings emphasize the importance of high-quality prenatal care and the early identification and prevention of risk factors. In low-resource settings, it is imperative to review management strategies and strengthen the capacity of the health system. Future research should focus on evaluating protocols adapted to local contexts and identifying socioeconomic factors influencing the management of PROM, with the ultimate goal of effectively reducing maternal and neonatal morbidity and mortality.

Data supporting the results of this study are available from the corresponding author, but restrictions apply to their availability. The data were used under license for the current study, and are therefore not publicly available. However, the data are available from the authors upon reasonable request and with permission.

Authors LNO, AKK, RK, JCB, GAZ, AY, DBM performed the research and contributed to data extraction and the manuscript draft. LNO, AKK, RK, DBM have made the analysis of data, manuscript revision, design and revision, final statistical analysis. All authors read and approve this final manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Ethics Committee of Felix Houphouet Boigny University (N347678-CI/2024). All subjects gave their informed consent for inclusion before they participated in the study.

This research received no external funding.

The authors declare no conflict of interest.