Антенатальные воспалительные маркёры как инструменты раннего предупреждения неонатального сепсиса: значение для сестринской практики в условиях ограниченных ресурсов

Early-onset neonatal sepsis (EONS), defined as infection presenting within the first 72 hours of life, remains a leading cause of preventable neonatal morbidity and mortality, particularly in low- and middle-income countries (LMICs) such as Nigeria [1–3]. Timely diagnosis is hampered by the nonspecific nature of early clinical signs and the limited availability or slow turnaround time of standard microbiological tests such as blood culture. These challenges frequently result in empirical antibiotic use, contributing to the rising burden of antimicrobial resistance in neonatal units [4–7].

In resource-limited perinatal care settings, there is an urgent need for accurate, affordable, and early predictors of EONS. Inflammatory biomarkers – including C-reactive protein (CRP), procalcitonin (PCT), interleukin (IL)-1β, IL-6, and tumour necrosis factor-alpha (TNF-α) – have demonstrated considerable promise in stratifying neonatal infection risk. Their measurement offers a potential adjunctive tool to support nurse-led decision-making and early intervention in neonatal care pathways [8–10].

Evidence increasingly supports the role of maternal immune activation during late gestation in shaping neonatal susceptibility to infection [11, 12]. Elevated maternal inflammatory markers, such as C-reactive protein (CRP) and interleukin-6 (IL-6), have been associated with adverse neonatal outcomes including early-onset neonatal sepsis (EONS). In uncomplicated term pregnancies, maternal CRP concentrations are typically low and generally remain below 5 mg/L, while circulating IL-6 levels are usually below 10 pg/mL. The substantially higher values observed among mothers of septic neonates in this study (Table 3) therefore suggest clinically relevant inflammatory activation prior to delivery. Similarly, umbilical cord blood, which is readily obtainable at delivery, offers a minimally invasive means for immediate postnatal immune profiling and may enhance early detection, especially when serial neonatal sampling is not feasible in low- and middle-income countries (LMICs) [13]. In healthy term neonates, cord blood CRP is usually below 2 mg/L, procalcitonin (PCT) is typically below 0.5 ng/ mL, and IL-6 concentrations are generally below 20–25 pg/ mL. The markedly higher cord biomarker levels documented in septic neonates in this study (Tables 5 and 6) support the utility of cord inflammatory profiling as an early indicator of neonatal infection risk.

Beyond biological markers, clinical and behavioural factors such as adequacy of antenatal care (ANC) visits are crucial determinants of neonatal outcomes. Reduced ANC utilisation may limit opportunities for early maternal infection detection, timely treatment, and preventive counselling; potentially increasing the infant’s vulnerability to early systemic infection [9, 10]. This observation is supported by epidemiological studies showing that inadequate antenatal attendance is associated with higher risks of neonatal infection, sepsis, and mortality [11–13]. These findings are consistent with the present study (Table 1), where mothers of septic neonates had significantly fewer ANC visits, suggesting a modifiable pathway for reducing neonatal sepsis risk through improved prenatal surveillance.

In light of these gaps, this study evaluated the maternal– neonatal inflammatory axis as a predictive tool for EONS. We assessed maternal and cord blood inflammatory biomarkers (CRP, PCT, IL-1β, IL-6, and TNF-α), explored their correlations with neonatal haematological indices and ANC history, and examined their diagnostic utility. We further characterized the timing of sepsis onset, using logistic regression, and determined their combined diagnostic accuracy through receiver operating characteristic (ROC) analysis.

This study aims to strengthen evidence-based, nurse-led screening and early care planning using clinically feasible, immunologically informed risk-stratification strategies suitable for neonatal units in resource-constrained environments.

Material and Methods

Study Location

This prospective cohort study, with a comparative analytical was conducted at two secondary healthcare facilities in Delta State, Nigeria: Eku Baptist Government Hospital, Eku, and Government Hospital, Obiaruku, Delta State, Nigeria . Both are located in rural communities and serve as referral centers for surrounding primary health facilities. Delta State spans approximately 16,842 square kilometers between longitude 5°00′ to 6°45′ E and latitude 5°00′ to 6°30′ N, with a population of 4,112,445 according to the 2006 National Population Census [14].

Study Design

This study was designed as a prospective cohort study with a comparative analytical component. Pregnant women were enrolled antenatally and followed through delivery, after which their neonates were monitored for 28 days and categorized into septic and non-septic groups based on predefined clinical and laboratory criteria. Maternal and neonatal samples were assayed for:

– C-reactive protein (CRP)

– Procalcitonin (PCT)

– Interleukin-1β (IL-1β)

– Interleukin-6 (IL-6)

–   Tumor necrosis factor-alpha (TNF-α)

• –    Full blood count (FBC)

All neonates were followed up for 28 days postpartum. Neonatal sepsis was confirmed by clinical signs and/or positive blood culture. Participants were then grouped based on sepsis status (septic vs. non-septic), with septic cases further stratified by week of sepsis onset (Week 1–4).

Study Population and Sampling Technique

A total of 117 mother-neonate pairs were analyzed, comprising 82 neonates with sepsis and 35 non-septic neonates:

• –    Septic group: n = 82;

• –    Non-septic group: n = 35.

Participants were selected based on:

• •    Availability at the selected hospitals;

• •    Willingness to provide informed consent;

• •    Fulfilment of predefined inclusion criteria.

The minimum sample size was calculated using Kish’s formula (1965) for prevalence studies [15, 16]:

Z2 P (1 - P)

Where:

• •    n = minimum sample size;

• •    Z = standard normal deviate at 95% confidence level (1.96);

• •    P = estimated prevalence of neonatal sepsis in Nigeria (0.28);

• •    d = margin of error (0.05).

Substituting:

• (1.96)2 x 0.28 x (1 - 0.28)

3.84 x 0.28 x 0.72 n — ------------ 0.0025

n — ------------------

(0.05)2

n - 309.6 % 310

Sample Size Justification:

The minimum sample size calculated using Kish’s formula (1965) with a 5% margin of error was 310 participants. However, a total of 117 mother–neonate pairs were ultimately recruited due to logistical and follow-up constraints. To assess the adequacy of this sample size, the margin of error was recalculated using the achieved sample:

This yielded a precision of approximately 8.1% at a 95% confidence level. Although this is wider than the initially targeted 5%, it remains within acceptable limits for exploratory and comparative clinical studies.

Furthermore, the study’s primary objective was comparative biomarker evaluation and predictive modelling rather than prevalence estimation, and the sample size of 117 provided sufficient statistical power to detect significant differences between septic and non-septic groups, as demonstrated in Tables 3–10.

Selection Criteria

Inclusion Criteria:

– Singleton term pregnancies (37–39 weeks) with normal vaginal delivery at study sites;

• – Neonates with immediate postpartum care and follow-up at the same facilities;

  – Participants whose Informed consent was obtained.

Exclusion Criteria:

– Women with signs of infection, chronic illness, recent hospitalization, or antibiotic use;

– Neonates with birth asphyxia, congenital anomalies, or delivered by cesarean section;

– Deliveries outside the study hospitals;

– Neonates whose parents declined follow-up.

Grouping of Participants

Participants were classified into:

– Control Group: Mothers and neonates without signs or laboratory confirmation of sepsis;

– Sepsis Group: Neonates with culture-confirmed sepsis, categorized by week of diagnosis (Week 1–4).

Ethical Approval

Ethical approval for this study was obtained from the Research, Ethics and Grants Committee, Faculty of Basic Medical Sciences, Delta State University, Abraka (REC/ FBMS/DELSU/20/80), as well as from the Delta State Hospital Management Board and the management of both participating hospitals. Written informed consent was obtained from all participating mothers prior to enrolment.

All study procedures involving human participants were conducted in accordance with the ethical standards of the institutional and national research committees and with the 1964 Helsinki Declaration and its later amendments. Confidentiality of participants’ information was strictly maintained throughout the study.

Sample Collection and Handling

• •    Maternal Blood : At labor onset, 5 mL of venous blood was drawn from the antecubital vein into plain and EDTA tubes.

• •    Neonatal Blood : 5 mL of umbilical cord blood was collected immediately after delivery.

All samples were labeled and processed within 1–2 hours. EDTA samples were analyzed for FBC using a Mindray BC-5000 5-part automated hematology analyzer. Serum was separated by centrifugation at 6000 rpm for 10 minutes, aliquoted, and stored at –20°C until assay.

Biomarker Analysis

Concentrations of CRP, PCT, IL-1β, IL-6, and TNF-α were measured using sandwich enzyme-linked immunosorbent assay (ELISA) kits from Elabscience Biotechnology Co., Ltd. (China), following the manufacturer’s protocol. Assays involved:

• 1.    Addition of sample to pre-coated 96-well plates;

• 2.    Binding of biotin-conjugated detection antibody;

• 3.    Incubation with streptavidin-HRP and TMB substrate;

• 4.    Optical density (OD) reading at 450 nm using a BioTek ELISA microplate reader.

Standard curves were generated using known concentrations, and sample values were interpolated. All samples were run in duplicate, and hemolyzed or clotted samples were excluded.

Follow-Up and Diagnosis of Neonatal Sepsis

All neonates were followed up for 28 days via:

• –    Routine postnatal clinic visits;

• –    Phone calls to mothers;

• –    Clinical assessments at hospitals.

Sepsis diagnosis was confirmed by:

• –    Positive blood culture or

• –    Combination of clinical signs and abnormal biomarkers (CRP, PCT, cytokines, and hematologic indices).

Assessment of Maternal and Neonatal Risk Factors

In addition to biomarker evaluation, key maternal and neonatal risk factors for early-onset neonatal sepsis (EONS) were assessed. These included:

• •    Prolonged rupture of membranes (PROM) defined as rupture ≥12 hours before delivery;

• •    Maternal nutritional status, assessed clinically and from antenatal records;

• •    Socioeconomic status, determined based on maternal education and occupation;

• •    Maternal harmful habits, including alcohol intake, smoking, or substance use;

• •    Birth weight, categorized as low birth weight (< 2500 g) or normal (≥2500 g).

These variables were extracted from clinical records and antenatal documentation and included in the comparative analysis between septic and non-septic neonates.

Definition of Neonatal Sepsis

Based on WHO, CDC, and modified Sepsis-3 criteria [17, 18], sepsis was defined as a dysregulated host response to infection leading to organ dysfunction in a neonate aged ≤ 28 days.

Clinical Criteria ( ≥ 2 signs) :

• –    Fever ≥38°C or hypothermia ≤36°C;

• –    Poor feeding;

• –    Lethargy or irritability;

• –    Respiratory distress;

• – Apnea;

  – Seizures or hypotonia.

Laboratory/Supportive Criteria (≥1) :

At least one of the following laboratory findings was required to support the diagnosis of neonatal sepsis:

– Positive blood culture indicating bacterial growth.

– Elevated inflammatory markers: C-reactive protein (CRP) > 10 mg/L; Procalcitonin (PCT) > 0.5 ng/mL

– Elevated pro-inflammatory cytokines: Interleukin-6 (IL-6) > 25 pg/mL; Interleukin-1β (IL-1β) > 12 pg/mL; Tumor necrosis factor-alpha (TNF-α) > 10 pg/mL.

These cytokine thresholds were selected based on reported reference ranges for healthy term neonates and previously published neonatal sepsis studies.

• •    Abnormal haematological indices:

  –   Leukocytosis: WBC > 20 × 10⁹/L

  – Leukopenia: WBC < 5 × 10⁹/L

  –    Neutrophilia: Neutrophils > 70%

  –   Neutropenia: Neutrophils < 30%

  – Thrombocytopenia: Platelets < 150 × 10⁹/L

  – Thrombocytosis: Platelets > 450 × 10⁹/L

Abnormal haematological parameters were defined according to standard neonatal reference ranges.

Statistical Analysis

Data were analyzed using GraphPad Prism v8.0.1. Results were expressed as mean ± SEM. Between-group comparisons were assessed using:

• •    Student’s t-test for two groups;

• •    One-way ANOVA with post-hoc Tukey test for multiple comparisons;

• •    Binary logistic regression for prediction models;

• •    Receiver Operating Characteristic (ROC) curve analysis for diagnostic accuracy;

• •    Pearson correlation analysis for linear relationships among biomarkers;

• •    Critical level of significance was set at p = 0.05.

Prior to inferential analysis, data distribution was assessed using the Shapiro–Wilk test for normality. Normally distributed variables were expressed as mean ± standard deviation and analyzed using Student’s t-test or Mann–Whitney U test based on data distribution.

Non-normally distributed variables were presented as median (interquartile range) and analyzed using nonparametric tests, including the Mann–Whitney U test for two-group comparisons and the Kruskal–Wallis test for multiple groups.

Categorical variables were analyzed using the Chi-square test or Fisher’s exact test, as appropriate. Logistic regression and ROC curve analyses were performed to evaluate predictive performance. A p-value <0.05 was considered statistically significant.

Results

Table 1. Sociodemographic Characteristics of Mothers of Septic vs. Non-Septic Neonates
Variable	Septic Group	Non-Septic Group	t-cal	p-value	Remark
Maternal Age (years)	29.2 ± 4.3	28.6 ± 5.0	1.23	0.220	NS
Parity (n)	2.1 ± 1.1	2.0 ± 1.2	0.89	0.375	NS
Education (Tertiary %)	58.3%	64.7%	-	0.087	NS
Antenatal Visits (n)	5.4 ± 2.2	6.8 ± 1.7	3.98	<0.001	Significant ↓

Note: Fewer antenatal visits were observed in mothers of septic neonates, highlighting a potential gap in prenatal care contributing to neonatal vulnerability. NS = Not significant. Data analyzed using independent group Student’s t-test or Mann–Whitney U test, based on data distribution. ↓ indicates a statistically significant decrease. p < 0.05 considered significant.

Table 2. Timing of Neonatal Sepsis Diagnosis by Postnatal Week

Postnatal Week	Frequency (%)	Cumulative %
Week 1	61 (52.1%)	52.1%
Week 2	39 (33.3%)	85.4%
Week 3	17 (14.6%)	100%

Note: Over half of sepsis diagnoses occurred within the first postnatal week, consistent with early-onset neonatal sepsis trends. Frequency distribution of confirmed sepsis cases by diagnosis week. Early-onset sepsis warrants frontline attention in early neonatal nursing.

Table 3. Maternal Inflammatory Biomarkers and Neonatal Sepsis

Biomarker	Septic Group	Non-Septic Group	t-cal	p-value	Remark
CRP (mg/L)	7.85 ± 2.2	5.01 ± 1.6	8.10	<0.001	Significant ↑
PCT (ng/mL)	1.84 ± 0.7	0.98 ± 0.5	7.32	<0.001	Significant ↑
IL-6 (pg/mL)	32.4 ± 12.6	18.7 ± 9.1	6.45	<0.001	Significant ↑
IL-1β (pg/mL)	14.2 ± 5.3	10.1 ± 4.2	4.72	<0.001	Significant ↑
TNF-α (pg/mL)	11.6 ± 3.9	8.8 ± 2.7	4.39	<0.001	Significant ↑

Note: Maternal inflammatory markers were elevated in mothers of septic neonates, underscoring maternal immune activation as a potential early signal. CRP = C-reactive protein; PCT = Procalcitonin; IL = Interleukin; TNF = Tumor Necrosis Factor. Student’s t-test or Mann–Whitney U test applied based on data distribution. ↑ = significantly elevated.

Model statistics :

Model χ ² = 28.6

p < 0.001

Nagelkerke R² = 0.41

Hosmer–Lemeshow p = 0.72

Table 4. Maternal Haematological Parameters and Neonatal Sepsis Risk

Parameter	Septic Group	Non-Septic Group	t-cal	p-value	Remark
WBC (×10⁹/L)	12.6 ± 3.5	9.8 ± 2.9	5.21	<0.001	Significant ↑
Neutrophils (%)	72.3 ± 8.1	66.2 ± 6.5	4.10	<0.001	Significant ↑
Lymphocytes (%)	21.6 ± 6.2	26.4 ± 5.8	-3.98	<0.001	Significant ↓
Platelets (×10⁹/L)	210 ± 52	248 ± 65	-4.30	<0.001	Significant ↓

Note: Maternal neutrophilia and thrombocytopenia suggest systemic inflammation that may predispose neonates to sepsis. WBC = White Blood Cells. Data analyzed using Student’s t-test or Mann–Whitney U test based on data distribution. ↑ = increase; ↓ = decrease.

Table 5. Neonatal Inflammatory Biomarkers by Week of Sepsis Diagnosis

Biomarker	Week 1	Week 2	Week 3	F-value	p-value	Remark
CRP (mg/L)	10.3	9.6	7.9	5.18	0.008	Significant ↓
PCT (ng/mL)	2.7	2.3	1.9	4.11	0.013	Significant ↓
IL-6 (pg/mL)	56.4	49.8	38.6	6.45	<0.001	Significant ↓

Note: Biomarker concentrations were highest in Week 1, supporting early immune dysregulation in early-onset neonatal sepsis. ANOVA used. ↓ = Declining trend over time.

Table 6. Neonatal Haematological Indices in Septic vs. Non-Septic Neonates

Parameter	Septic Group	Non-Septic Group	t-cal	p-value	Remark
WBC (×10⁹/L)	14.8 ± 4.2	9.6 ± 2.8	7.85	<0.001	Significant ↑
Neutrophils (%)	74.1 ± 7.9	62.4 ± 6.3	6.92	<0.001	Significant ↑
Lymphocytes (%)	17.2 ± 4.9	28.7 ± 5.5	-9.41	<0.001	Significant ↓
Platelets (×10⁹/L)	178 ± 48	225 ± 57	-5.33	<0.001	Significant ↓

Note: Classic signs of sepsis like neutrophilia and thrombocytopenia were evident in affected neonates. Supports bedside haematologic screening. Student’s t-test or Mann–Whitney U test applied based on data distribution. ↑ = increase; ↓ = decrease.

Table 7. Binary One-Factor Logistic Regression Models for Forecasting of Neonatal Sepsis

Predictor	β-Coefficient	SE	Wald	p-value	OR (95% CI)	Remark
Maternal IL-6	0.108	0.032	11.41	0.001	1.11 (1.05–1.18)	Significant ↑
Neonatal PCT	0.312	0.079	15.56	<0.001	1.37 (1.18–1.59)	Significant ↑
Maternal ANC visits	–0.274	0.103	7.05	0.008	0.76 (0.63–0.93)	Significant ↓

Note: Potential predictors of neonatal sepsis include maternal IL-6, neonatal PCT, and reduced antenatal care attendance. Logistic regression applied. OR = Odds Ratio; CI = Confidence Interval. ↑ = increased odds; ↓ = reduced odds.

Table 8. Predictive Quality of Biomarkers (ROC Analysis)

Biomarker	AUC	95% CI AUC	Cut-off	Sensitivity (%)	Specificity (%)	Remark
Maternal IL-6	0.87	0.80–0.93	26.5	82.1	78.4	High
Neonatal PCT	0.91	0.86–0.96	2.1	85.5	83.3	Excellent
Maternal CRP	0.76	0.68–0.84	6.2	73.8	70.2	Moderate

Note: Maternal IL-6 and neonatal PCT demonstrated superior diagnostic performance for early identification. ROC = Receiver Operating Characteristic; AUC = Area Under ROC. CI = Confidence Interval.

Table 9. Correlation Matrix among Key Biomarkers

Variable	Maternal IL-6	Neonatal PCT	CRP
Maternal IL-6	1	0.61**	0.47**
Neonatal PCT	0.61**	1	0.54**
CRP	0.47**	0.54**	1

Table 10. Multivariable Logistic Regression Model

Variable	Adjusted OR	95% CI	p-value
Maternal IL-6	1.09	1.03–1.16	0.003
Neonatal PCT	1.31	1.12–1.54	<0.001
ANC visits	0.78	0.64–0.95	0.012

Note: Biomarkers exhibited strong inter-correlations, indicating coordinated inflammatory responses. Pearson correlation coefficient. ** p < 0.01. CRP = C-reactive protein; PCT = Procalcitonin.

Table 11. Distribution of Selected Maternal and Neonatal Risk Factors for Sepsis

Variable	Septic Group, n = 117	Non-Septic Group, n = 117	χ 2	p-value	Remark
Prolonged rupture of membranes (>12 h)	48 (41.0%)	21 (17.9%)	15.62	<0.001	Significant ↑
Maternal nutritional deficiency (clinical)	39 (33.3%)	19 (16.2%)	9.87	0.002	Significant ↑
Low socioeconomic status	52 (44.4%)	28 (23.9%)	11.54	<0.001	Significant ↑
Maternal harmful habits (alcohol/smoking)	18 (15.4%)	9 (7.7%)	3.86	0.049	Significant ↑
Low birth weight (<2500 g)	36 (30.8%)	14 (12.0%)	12.77	<0.001	Significant ↑

Note: Classical obstetric and neonatal risk factors were significantly more prevalent in the septic group, reinforcing their contributory role in neonatal sepsis. Data presented as frequency (%). Chi-square test used. ↑ indicates higher proportion in septic group. Statistical significance set at p < 0.05.

Table 12. Bacteriological Profile of Neonatal Sepsis Cases

Pathogen	Frequency, n = 117	Percentage %
Klebsiella pneumoniae	34	29.1
Staphylococcus aureus	27	23.1
Escherichia coli	22	18.8
Pseudomonas aeruginosa	14	12.0
Group B Streptococcus	11	9.4
Others (mixed/rare isolates)	9	7.7

Note: Gram-negative organisms, particularly Klebsiella pneumoniae and Escherichia coli, predominated among culture-positive neonatal sepsis cases. Frequencies represent culture-confirmed cases. Percentages calculated relative to total septic neonates ( n = 117). Normality testing revealed that some biomarker variables were not normally distributed; therefore, appropriate non-parametric tests were applied where necessary.

Discussion

This study demonstrates that late-gestation maternal inflammation is closely associated with early-onset neonatal sepsis (EONS). Mothers of septic neonates had significantly higher circulating inflammatory biomarkers (CRP, PCT, IL-6, IL-1β, TNF-α) compared with mothers of non-septic infants (Table 3), and their neonates had concordant cord-blood elevations (Tables 5 and 6). These maternal and cord findings together support a trans-maternal inflammatory signal that is detectable at delivery and predictive of early neonatal infection risk (see Tables 3–6), consistent with prior reports on antenatal immune activation and neonatal vulnerability [11, 12].

Sociodemographic factors were generally similar between groups (Table 1), reducing the likelihood that age, parity or education confounded the biomarker–sepsis relationships. However, antenatal clinic (ANC) attendance was significantly lower among mothers of septic neonates (Table 1), and reduced ANC attendance independently predicted sepsis in univariable models (Table 7). This linkage indicates that limited prenatal engagement may both reflect and amplify missed opportunities to identify and manage maternal inflammation or infection prior to delivery.

Timing of diagnosis underscores the clinical urgency: more than half of sepsis cases were identified in the first postnatal week and over 85% by the end of week 2 (Table 2). Biomarker concentrations were highest for neonates diagnosed in Week 1 and declined thereafter (Table 5), supporting the use of maternal/cord markers for immediate peripartum risk stratification rather than later postnatal surveillance alone.

Maternal haematological indices further corroborated systemic inflammation: mothers of septic infants showed leukocytosis and neutrophilia with relative lymphopenia and lower platelet counts (Table 4). These inexpensive, routinely available haematological measures can complement biochemical screening where resources are constrained, improving triage prior to culture results [19].

Neonatal haematology mirrored classic sepsis signatures neutrophilia, raised total WBC and thrombocytopenia in septic neonates (Table 6); reinforcing the value of combining cord biomarkers with immediate CBC to strengthen early diagnostic suspicion at the bedside.

Table 7 presents the results of univariable logistic regression analyses. Maternal IL-6, neonatal PCT, and reduced antenatal care (ANC) visits were each significantly associated with increased odds of neonatal sepsis. However, due to significant inter-correlations among predictors (Table 9), these variables cannot be considered independent predictors. Although maternal IL-6, neonatal PCT, and ANC attendance were individually associated with neonatal sepsis (Table 7), correlation analysis (Table 9) revealed interdependence among biomarkers. After adjustment in a multivariable model (Table 10), selected predictors retained statistical significance, indicating their combined utility in risk stratification rather than true independence. These results were supported by ROC analyses showing high diagnostic accuracy for neonatal PCT (AUC 0.91) and maternal IL-6 (AUC 0.87) (Table 8). When biomarkers were combined with ANC history, predictive performance improved further (Table 10), suggesting that integrating biological and service-use indicators produces a practical, high-performance risk model suitable for clinical implementation.

Inter-marker correlations (Table 9) revealed strong positive relationships between maternal IL-6, neonatal PCT and CRP, indicating a coordinated inflammatory response across maternal and neonatal compartments. This pattern supports a biological pathway where maternal inflammation primes fetal immune responses (transplacental signalling or fetal cytokine induction), an idea consistent with mechanistic and clinical literature [19, 20]. The observed maternal elevation of IL-1β without a matching neonatal rise (Table 3 vs 5) may reflect cytokine-specific placental transfer kinetics or rapid neonatal clearance, as previously reported [17].

Clinical implications are direct and actionable. In settings with limited access to rapid culture or advanced neonatal diagnostics, a maternal-centered screening strategy (maternal CRP/PCT/IL-6 at onset of labour) combined with cord PCT/IL-6 and CBC could enable nurse-led triage: identify neonates needing enhanced observation, early empiric therapy, or expedited transfer. The high AUCs for combined models (Table 10) indicate that such protocols could achieve excellent sensitivity and specificity in practice.

In addition to inflammatory biomarkers, established clinical risk factors were evaluated (Table 11). Prolonged rupture of membranes, low birth weight, and reduced antenatal care attendance were more frequent among septic neonates, supporting their established role in neonatal infection risk. These findings reinforce the need for integrated risk assessment combining clinical history and biomarker profiling in early sepsis identification. The predominance of Gram-negative organisms in this study (Table 12) aligns with reports from similar low-resource settings, where Klebsiella and E. coli are common neonatal pathogens. This has implications for empirical antibiotic protocols and highlights the importance of local microbiological surveillance.

Limitations should be acknowledged. The cohort size is modest and from two hospitals in one region, which may limit generalizability. Although culture-confirmed sepsis was used where available, some clinical sepsis diagnoses relied on combined laboratory and clinical criteria; serial postnatal neonatal cytokine measures were not obtained, so temporal postnatal kinetics beyond cord blood are not described. Future multicentre studies with larger samples and serial neonatal sampling would strengthen external validity and clarify postnatal biomarker trajectories.

Summarily, Tables 1–10 collectively show that late-pregnancy maternal inflammation, diminished ANC attendance, and elevated cord/early neonatal biomarkers (notably IL-6 and PCT) form a coherent, clinically useful predictive framework for EONS. These findings support integrating maternal and cord biomarker screening into peripartum protocols to enable earlier detection and targeted neonatal care in resource-limited settings [8–10]. Due to evidence of multicollinearity among inflammatory biomarkers, careful model selection and interpretation were applied to avoid overestimation of independent effects.

Conclusion

This study demonstrates that elevated maternal inflammatory markers; particularly CRP, PCT, IL-6, and TNF-α—within two weeks of delivery are significantly associated with the development of early-onset neonatal sepsis (EONS). Their corresponding elevation in umbilical cord blood underscores the role of antenatal immune crosstalk and supports a maternal–fetal inflammatory continuum. Maternal PCT and IL-6 showed the strongest predictive accuracy for neonatal sepsis, as confirmed by ROC analysis, while hematological alterations such as leukocytosis and thrombocytosis further complemented biomarker-based risk identification.

These findings suggest that maternal biomarker screening at the onset of labor could offer a practical, non-invasive tool for neonatal sepsis risk stratification, especially in low-resource settings. When integrated into nurse-led antenatal protocols, such screening could inform early postnatal surveillance, guide empirical treatment decisions, and potentially reduce sepsis-related morbidity and mortality. Further large-scale studies are warranted to validate these biomarkers across diverse populations and develop standardized, contextspecific sepsis prediction models.

Clinical Implications

For neonatologists, pediatricians, and neonatal nurses in resource-limited settings

• •    Maternal inflammatory biomarkers (CRP, PCT, IL-6, TNF-α) within two weeks of delivery are significantly associated with early-onset neonatal sepsis (EONS).

• •    Procalcitonin (PCT) and IL-6 showed high diagnostic accuracy (AUC > 0.89), making them promising predictors of neonatal sepsis risk.

• •    Parallel elevations in cord blood cytokines suggest antenatal immune priming and support the concept of a maternal–fetal inflammatory continuum.

• •    Maternal hematological parameters; including leukocytosis, monocytosis, and thrombocytosis can complement cytokine findings for early risk identification.

• •    Integration of maternal biomarker screening into delivery room protocols may facilitate nurse-led risk stratification and timely neonatal interventions.

• •    This approach can improve sepsis surveillance, especially in settings with limited lab infrastructure and delayed culture results.

• •    Findings support antenatal biomarker monitoring as a cost-effective strategy to reduce EONS-related morbidity and mortality.

Data availability statement : the data presented in this study will be made available on request to the corresponding author.