Correlation between Escherichia coli infection during pregnancy and maternal-fetal outcomes: a retrospective analysis (2025)

Background

There has been an increase in infections caused by Escherichia coli during pregnancy. E. coli is a major pathogen posing substantial risks to maternal and fetal health. However, knowledge of the impact of these infections on maternal and fetal outcomes is limited. This study aimed to investigate the high-risk factors related to Escherichia coli infection in pregnant women and associated maternal and fetal outcomes.

Methods

This retrospective analysis was performed using clinical data of pregnant women hospitalized between January 2020 and December 2022 for conditions such as premature rupture of membranes (PROM), threatened preterm labor, threatened miscarriage, and fever, all of whom underwent microbiological testing. Patients were categorized into E. coli or non-E. coli infection groups based on bacterial culture results. High-risk factors for E. coli infection and associated pregnancy outcomes were assessed.

Results

Microbiological analysis of 7,213 clinical specimens revealed 925 positive cases, with E. coli infection accounting for 202 cases. Primiparity (odds ratio = 2.981, 95% CI 1.877–4.709, P < 0.001) and a pre-pregnancy body mass index of ≥ 24kg/m² (odds ratio = 1.149, 95% CI 1.062–1.898, P = 0.017) were identified as independent risk factors for E. coli infection. E. coli infection was significantly associated with an increased incidence of preterm PROM. The E. coli infection group exhibited a higher rate of miscarriages before 28 weeks, a lower rate of full-term deliveries, and higher incidences of fetal distress during labor than the non-E. coli infection group.

Conclusions

The prevalence of E. coli is high among pregnant women. This observational study suggests an association between maternal E. coli infection and adverse pregnancy outcomes in high-risk populations.

Common pathogens responsible for infections during pregnancy include Group B Streptococcus (GBS) and Escherichia coli. With the introduction of prenatal GBS screening and prophylactic antibiotic use during labor, neonatal infections caused by GBS have decreased; however, infections due to E. coli have increased.

E coli is a Gram-negative bacterium that naturally colonizes the intestinal mucosa. Despite being typically a gut commensal, its high genomic plasticity enables diverse pathogenic traits. E. coli strains are classified into seven intestinal pathovars “InPEC” and three extraintestinal pathovars “ExPEC” [1, 2], each associated with distinct mechanisms of pathogenesis, driven by virulence-factor gene expression [3]. Colonization rates of E. coli in pregnant women vary across studies, ranging from 13 to 35.8% [4, 5]. A 2021 meta-analysis reported an overall detection rate of 29% in combined samples (urine, stool, and vaginal swabs), with a vaginal colonization rate of up to 32% [6]. E. coli has been implicated in perinatal complications, such as fever, chorioamnionitis, miscarriage, preterm birth, and stillbirth [7]. Additionally, it is a major cause of neonatal encephalitis and sepsis, posing significant risks to both maternal and fetal health [8, 9]. Currently, effective screening and preventive measures are lacking, with limited research on the maternal and fetal outcomes of these infections.

Therefore, this study aimed to investigate the high-risk factors for E. coli infection, as well as their impacts on maternal and fetal outcomes, with the goal of improving the management of E. coli infections in pregnant women.

Study population and design

All procedures in this study were approved by the Human Research Ethics Committee of the Obstetrics and Gynecology Hospital, Fudan University (grant no. Kyy2018-57). The requirement for patient consent was waived owing to the retrospective nature of this study. This retrospective analysis was performed in patients hospitalized at the Yangpu campus of the Obstetrics and Gynecology Hospital of Fudan University between January 2020 and December 2022 owing to premature rupture of membranes (PROM), threatened preterm labor, threatened miscarriage, or fever.

The inclusion criteria were (1) patients who were hospitalized for PROM, threatened preterm labor, threatened miscarriage, or fever and underwent microbiological testing; (2) those who had an established prenatal care record and attended regular check-ups at the Obstetrics and Gynecology Hospital of Fudan University, with complete clinical data; and (3) pregnancies without considerable fetal malformations detected during prenatal screening.

The exclusion criteria were (1) patients who received antibiotics within the past 2 weeks prior to the culture examination; (2) pregnant women with acute infectious diseases, defined as microbiologically confirmed infectious diseases requiring isolation, including respiratory (e.g., influenza, COVID-19), gastrointestinal (e.g., viral hepatitis), and systemic infections (e.g., initial symptoms of sepsis, septicemia); and (3) pregnant women with severe underlying conditions.

Data collection

During the study, 32,982 patients were hospitalized at the Obstetrics and Gynecology Hospital of Fudan University. Clinical samples (n = 7,213) were systematically collected. After removing duplicate samples and excluding those positive for other bacterial pathogens, E. coli was detected in 221 samples. Following the exclusion of duplicate patient entries, 6,337 unique patients underwent microbiological sampling. The study population was stratified based on E. coli infection status: 202 culture-confirmed E. coli-positive cases versus 5,412 E. coli-negative controls (Fig.1).

The two groups were subsequently compared in terms of age, obstetric history, body mass index (BMI), history of multiple abortions, in vitro fertilization-embryo transfer (IVF-ET), twin pregnancies, pregnancy complications, and comorbidities. Additionally, pairwise comparisons of pregnancy outcomes, such as PROM, fever, postpartum hemorrhage, and neonatal asphyxia, were performed. Pregnancy complications and comorbidities included gestational diabetes, pregnancy-related hypertensive disorders, and intrahepatic cholestasis of pregnancy.

Study definitions/diagnostic criteria

Threatened miscarriage

In pregnant women at < 28 weeks of gestation, threatened miscarriage is characterized by minor vaginal bleeding and/or lower abdominal pain following a missed period, with a closed cervical os, intact fetal membranes, no expulsion of pregnancy tissue, and uterine size corresponding to gestational age.

Threatened preterm labor

This condition occurs between 28 and 37 weeks of gestation, presenting with regular uterine contractions (4 contractions within 20min or 8 contractions within 60min) without cervical dilation.

Fever

Fever includes the following conditions: prenatal, intrapartum, and postpartum fever, as well as chorioamnionitis.

E. coli infection

Microbiological analyses were conducted on specimens collected from vaginal discharge, intrauterine content, and blood of pregnant women presenting with threatened abortion and preterm labor, PROM, and fever. Specimens obtained from sterile sites (e.g., blood) underwent strain identification and antimicrobial susceptibility testing. For non-sterile site samples (e.g., vaginal swabs), bacterial cultures were performed, followed by identification and susceptibility profiling of predominant bacterial isolates. Detection of E. coli in these specimens was interpreted as E. coli infection. Strict criteria were applied to confirm E. coli infections, excluding potential contaminants (e.g., skin flora) and asymptomatic colonization, which was defined as a positive culture without infection-related symptoms or inflammatory biomarkers.

Bacterial identification

The culture, isolation, and identification of bacterial strains were conducted following the protocols outlined in the “National Clinical Laboratory Procedures.” Identification was performed using matrix-assisted laser desorption ionization time-of-flight mass spectrometry, with quality control strains including E. coli ATCC25922, Pseudomonas aeruginosa ATCC28753, Candida albicans ATCC90028, and Staphylococcus aureus ATCC25923, which were all supplied by the Shanghai Clinical Research Center. Microbiological specimens collected from sterile sites underwent comprehensive strain identification and antimicrobial susceptibility testing (AST). For specimens obtained from non-sterile sites, strain identification and AST were performed only when cultures exhibited predominant bacterial growth. To avoid duplication, repeated isolation of strains from multiple specimens of the same patient was systematically recorded as a single isolate.

Statistical analysis

Categorical data are expressed as frequencies and percentages, and comparisons between the groups were made using the χ² or Fisher exact test. Univariate analysis was performed using the χ² test to identify potential risk factors for E. coli infection, and multivariate logistic regression was used to analyze factors associated with E. coli infection. Statistical analysis was conducted using SPSS 20.0 (IBM Corp.), and a P-value of < 0.05 was considered statistically significant.

Results of E. coli detection

A total of 7,213 clinical samples were analyzed, comprising 1,487 vaginal discharge samples, 5,491 intrauterine content samples, and 235 blood samples. E. coli was detected in 221 samples, with the highest detection rate observed in blood cultures (8.51%). Microbial analysis revealed confirmed 925 pathogenic isolates from 6,337 unique patients after duplicate exclusions, while 5,412 cases showed no detectable pathogens. The three most frequently identified pathogens were Candida albicans (n = 231, 24.97%), E. coli (n = 202, 21.84%), and Streptococcus agalactiae (n = 171, 18.47%), collectively accounting for 65.28% of all identified microorganisms. E. coli was detected in 202 patients, yielding an overall detection rate of 3.19%. E. coli exhibited the highest detection rate during the puerperium (n = 23, 25.00%). Detailed results are presented in Tables1 and 2.

Risk factors for E. coli infection

In analyzing the general data of the two groups of pregnant women, factors such as age, parity, pre-pregnancy BMI, history of multiple abortions, IVF-ET, pregnancy complications, and comorbidities were used as covariates, whereas E. coli infection was used as the dependent variable. Primiparity and pre-pregnancy BMI of ≥ 24kg/m² were independent risk factors for E. coli infection in pregnant women, with odds ratios (95% confidence intervals [CIs]) of 2.981 (1.887–4.709) and 1.419 (1.062–1.898), respectively. Detailed results are presented in Tables3 and 4.

Relationship between E. coli infection and adverse maternal and neonatal outcomes

The numbers of patients with delivery outcomes during hospitalization were 179 and 5,343 in the E. coli and non-E. coli infection groups, respectively. Subsequently, the maternal and neonatal outcomes, including fever, preterm PROM (PPROM), gestational age at delivery, delivery method, postpartum hemorrhage, fetal distress, and neonatal asphyxia, were compared between the groups. The analysis revealed that E. coli infection showed a significant positive correlation with PPROM (P = 0.007). The E. coli infection group exhibited a higher proportion of miscarriages before 28 weeks, a lower proportion of full-term deliveries, and fewer cesarean sections than the non-E. coli infection group (all, P < 0.0001). Additionally, the incidences of fetal distress were significantly higher in the E. coli infection group than in the non-E. coli infection group (P < 0.0001). In the E. coli infection group, one case of maternal sepsis was identified (1/179, 0.6%), whereas no cases occurred in the control group (0/5343). Fisher’s exact test indicated statistical significance (P = 0.003); however, owing to the extremely low event rate, we have refrained from tabulating this outcome to avoid overinterpretation. No significant differences were observed in intrapartum fever, postpartum hemorrhage, and neonatal asphyxia between the groups (all, P > 0.05). However, these observed associations should be interpreted with caution. The details are depicted in Table5.

In this study, the clinical data from pregnant women were used to identify high-risk factors for E. coli infection and assess the impact of this infection on maternal and neonatal outcomes, thereby providing insights for managing E. coli infections during pregnancy.

At the Obstetrics and Gynecology Hospital of Fudan University, pathogen testing is routinely performed for pregnant women presenting with symptoms such as fever, PROM, threatened miscarriage, or threatened preterm labor. Among infections in pregnancy, excluding genital fungal infections, E. coli was the most prevalent, accounting for 21.84% (202/925) of cases. The overall detection rate of E. coli in the tested population was 3.19%. Substantial geographic variation in E. coli detection rates has been documented, ranging from 9.8 to 48.6% across different regions [10,11,12]. A systematic review and meta-analysis of 82 cohort studies demonstrated that the global prevalence of E. coli colonization in pregnant women reached 29% (95% CI 23–36%) [6]. The low detection rate in our study may be attributed to two reasons: first, as E. coli is both a commensal gut microbiota and opportunistic pathogen, asymptomatic colonization could lead to overestimation in population-level screening studies; second, our protocol exclusively analyzed symptomatic cases and excluded urine and fecal samples that inherently harbor high baseline colonization rates. This targeted approach enhances clinical relevance by distinguishing pathogenic infections from incidental carriage. Among different types of specimens, the detection rates were 8.51% and only 1.91% in blood and intrauterine content cultures, respectively. This variation is possibly related to the influence of clinical sampling practices. Specifically, at the Obstetrics and Gynecology Hospital of Fudan University, intrauterine content cultures were routinely performed postpartum for women who had PROM and a body temperature of ≥ 37.8°C, whereas blood cultures were conducted only when body temperature reached ≥ 38.5°C. Additionally, E. coli was detected in 6.2% of patients with a gestational age of < 28 weeks and in only 3.0% of those with a gestational age of ≥ 28 weeks. This difference may be attributed to the increasing estrogen levels as pregnancy progresses, which enhance lactobacilli activity and proliferation by increasing glycogen utilization, consequently leading to a more favorable vaginal microbiome [13, 14]. A retrospective study included 593 singleton pregnancies and categorized the study population into an early pregnancy group (n = 221), a mid-pregnancy group (n = 138), and a late pregnancy group (n = 234). The results suggested that the rate of Gram-negative bacterial infections, particularly E. coli infections, significantly decreased with advancing gestational age, with rates of 6.3%, 3.6%, and 1.7% in early, mid, and late pregnancy, respectively (P = 0.011) [15]. Furthermore, E. coli infection is associated with adverse outcomes, such as PROM and threatened miscarriage, explaining the higher detection rates in women at < 28 weeks of gestation [16]. Clinically, for patients with threatened miscarriage or preterm PROM, actively investigating the underlying cause and, once identified, treating E. coli infections promptly are important to avoid adverse outcomes. During the postpartum period, the positivity rate for E. coli was 25.0%, likely owing to decreased immunity following childbirth and poor personal hygiene practices. Women with PROM, gestational diabetes, or gestational hypertension may experience an increased risk of postpartum infection [17]. For women with these high-risk factors, postpartum hygiene education is essential to prevent E. coli infections during the puerperium.

Factors associated with E. coli infection include age, history of multiple miscarriages, parity, education level, and the use of antibiotics during pregnancy [4, 18, 19]. This study also identified being a primipara and having a pre-pregnancy BMI ≥ 24kg/m² as independent risk factors for E. coli infection in pregnant women. The fact that multiparous women exhibited a lower risk is possibly owing to permanent protective structural changes caused by previous pregnancies [20]. Animal studies have demonstrated that adaptive immune responses polarized toward Th17/Th1 phenotypes enhance the innate immune system’s capacity to eliminate bacterial pathogens during the postpartum period [21]. Compared to multiparous women, primiparous women are more likely to undergo intrapartum interventions (e.g., artificial rupture of membranes, repeated vaginal examinations), which elevates their risk of infection exposure [22]. Additionally, primiparous women are at a substantially elevated risk of requiring episiotomy, experiencing perineal trauma (including second-degree or higher lacerations), and developing dystocia during labor. These interventions exacerbate tissue trauma and necrosis, impair local microcirculation, and suppress neutrophil recruitment, collectively contributing to the increased likelihood of bacterial contamination in primiparous pregnancy [23]. Alternatively, an increased BMI may suppress immune function, increase glycogen levels in vaginal epithelial cells, and alter the vaginal microbiome [24]. Additionally, a higher BMI can reduce prostacyclin secretion, increase peroxidase activity, and promote metabolic abnormalities and inflammatory responses through adipokine secretion, thereby contributing to a higher likelihood of perinatal infections [25]. Maternal overweight and obesity are related to increased risks of early-onset sepsis caused by GBS and Escherichia coli in dose-response manners [26].

Pregnant mice infected with E. coli show a higher differentiation of macrophages into the M1 subtype than controls, whereas fewer macrophages are differentiated into the M2 subtype. This leads to a substantial down regulation of chemokine expression in placental and uterine tissues, resulting in adverse maternal-fetal outcomes [27]. In this study, pairwise comparison of the pregnancy outcomes suggested that E. coli infection was positively correlated with PPROM. This may be because E. coli suppresses collagen content in the membranes by promoting matrix metalloproteinase activity, thereby disrupting their normal structural function and leading to PPROM [28, 29]. The E. coli infection group exhibited a higher rate of miscarriages before 28 weeks and a lower rate of full-term deliveries than the non-E. coli infection group. E. coli can ascend from the vagina through the cervix and even penetrate intact membranes, causing amniotic fluid infection [30]. In the amniotic cavity, E. coli stimulates the production of cytokines, chemokines, inflammatory mediators, and proteases, which promote myometrial contractions, leading to miscarriage and preterm birth. Additionally, intrauterine infection may result in fetal growth restriction and increased placental vascular resistance, compromising the ability of the fetus to cope with stress during delivery and, consequently, leading to fetal distress. This distress may increase the rate of forceps delivery. In this study, the lower cesarean section rate observed in the E. coli-positive group (31.8% vs. 41.0%) is likely attributable to more proactive intrapartum interventions (e.g., higher rates of forceps delivery) and a higher rate of preterm births in the infection cohort. For pregnant women with PROM or intrapartum fever, clinical protocols prioritized infection risk mitigation through expedited labor progression rather than surgical delivery. E. coli in the amniotic fluid can cause complications, such as intrapartum fever, chorioamnionitis, and maternal sepsis [31, 32]. In this study, no significant differences were observed between the groups in the rates of intrapartum fever, chorioamnionitis, postpartum hemorrhage, or neonatal asphyxia. In our clinical practice, antibiotics are administered routinely to prevent infection in patients with PPROM. Moreover, for women with a body temperature exceeding 37.5°C, acetaminophen and physical cooling methods are used in addition to appropriate antibiotics. These measures, together with close monitoring and active management of the labor process, prevent prolonged labor, thereby reducing the incidence of intrapartum fever, chorioamnionitis, and other complications, such as postpartum hemorrhage and neonatal asphyxia. In clinical practice, for primiparous women with a BMI ≥ 24kg/m² who present with symptoms such as fever, PPROM, or threatened preterm labor, considering the possibility of E. coli infection, promptly performing pathogen testing, and initiating empirical treatment are crucial to prevent adverse maternal and neonatal outcomes.

This study has some limitations that warrant careful consideration of their potential impact on the observed associations. Since this single-center study employed a retrospective analysis, our findings may be influenced by selection bias and institutional practice patterns, such as higher rates of intrapartum interventions in E. coli-positive cases, potentially overestimating the pathogen-specific effects on outcomes. The restricted demographic diversity further limits generalizability to populations with different genetic predispositions or healthcare protocols. Owing to the limitations of the retrospective design, key factors influencing outcomes—such as antibiotic use, hygiene practices, pregnancy complications, and labor duration—were not fully analyzed, yielding unresolved potential residual confounding. For example, incomplete documentation of antibiotic exposures (particularly intrapartum prophylaxis) and labor duration metrics (e.g., rupture-to-delivery interval) may have obscured protective effects against E. coli-associated complications. Additionally, unmeasured confounders (such as socioeconomic status influencing hygiene practices or undocumented prenatal care intensity) could have inflated the perceived association between E. coli colonization and adverse outcomes. Furthermore, although microbiological tests were clinically validated, they may have limited sensitivity or specificity, impacting the accuracy of bacterial colonization assessments. Finally, although our findings demonstrate a significant correlation between E. coli infection and adverse pregnancy outcomes, the retrospective nature of this study precludes definitive conclusions regarding causality. The observed relationships may reflect either a direct pathogenic role of E. coli, a marker of underlying risk factors, or consequences of infection-related interventions.

These factors suggest that the observed associations warrant cautious interpretation. To clarify causal relationships and inform evidence-based prevention strategies, future research should prioritize multicenter, randomized controlled trials with longitudinal follow-up. Such studies should incorporate standardized protocols, systematic documentation of confounders (e.g., intrapartum practices, antimicrobial use, and prenatal care differences), and patient-reported outcomes. These elements are essential for validating the epidemiological links and developing targeted management guidelines for high-risk populations.

PROM:

Premature rupture of membranes

GBS:

Group B Streptococcus

BMI:

Body mass index

IVF-ET:

In vitro fertilization-embryo transfer

PPROM:

Preterm premature rupture of membranes

TPROM:

Term premature rupture of membranes

Author notes

1. Dongdong Shi, Chang Liu these authors have contributed equally to this work and share the first authorship.

Authors and Affiliations

1. Department of Obstetrics, Obstetrics and Gynecology Hospital of Fudan University, Shanghai, 200082, China

   Dongdong Shi,Chang Liu,Yan Cheng,Haidong Cheng&Qingying Zhang

Contributions

DDS organized and analyzed the data and wrote the manuscript. CL analyzed the data. YC supervised the study and revised the manuscript. HDC supervised the study. QYZ designed the study and revised the manuscript.All authors reviewed the manuscript.

Corresponding author

Correspondence to Qingying Zhang.

Ethics approval and consent to participate

This study received ethical approval from the Human Research Ethics Committee of the Obstetrics and Gynecology Hospital, Fudan University (Approval No. Kyy2018-57). The requirement for written informed consent was formally waived by the same Institutional Review Board (IRB) due to the retrospective, non-interventional design that utilized exclusively pre-existing, de-identified medical records. This waiver aligns with Article 39 of the Chinese Regulations on the Ethical Review of Biomedical Research Involving Human Subjects (2016), which permits exemption from informed consent for studies using anonymized archival data that pose no risk to participants. The research complied with the ethical standards of the 1964 Declaration of Helsinki and its subsequent amendments. All direct identifiers (e.g., names, ID numbers) were removed during data extraction to ensure patient confidentiality. No additional clinical interventions, biospecimen collections, or analyses involving identifiable information were performed.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Clinical trial number

Not applicable.

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions