Histopathological placental inflammation and umbilical cord blood inflammatory biomarkers in preterm infants: a prospective cohort study Authors: Krešimir Šantić, Ivana Bogojević, Ines Šahinović, Kristina Kralik, Jasmina Rajc, Silvija Pušeljić, Darjan Kardum Published: 2026 ABSTRACT BackgroundHistopathological placental inflammation provides objective evidence of intrauterine inflammatory exposure in preterm birth, but placental histopathology is not immediately available after delivery. We aimed to determine whether placental inflammatory pathology is associated with cord blood and postnatal inflammatory biomarkers in preterm infants.MethodsIn this prospective cohort study, 87 preterm infants with birth weight < 2500 g were enrolled. Cord blood collected at birth was analyzed for procalcitonin (PCT), presepsin (P-SEP), and high-sensitivity C-reactive protein (hs-CRP). Placentas were categorized as no pathological findings, inflammatory placental pathology, or other placental pathology. Exploratory lesion-specific analyses further classified inflammatory lesions as maternal inflammatory response only (MIR-only) or fetal inflammatory response present (FIR-present). Multiple linear regression models with log-transformed outcomes were used to examine associations between inflammatory placental pathology and cord blood PCT and postnatal CRP after adjustment for perinatal covariates.ResultsFifteen infants (17%) had inflammatory placental pathology. Cord blood PCT differed significantly between the primary placental pathology groups and was highest in infants with inflammatory placental pathology (P = 0.001), whereas cord blood P-SEP and hs-CRP did not differ significantly. Postnatal CRP was also higher in infants with inflammatory placental pathology (P = 0.004). In adjusted binary models, inflammatory placental pathology remained associated with higher cord blood PCT (adjusted fold change 2.31, 95% CI 1.50–3.55; P < 0.001) and higher postnatal CRP (adjusted fold change 4.66, 95% CI 2.23–9.76; P < 0.001). Exploratory lesion-specific analyses showed stepwise increases in cord blood PCT and postnatal CRP from no inflammatory lesion to MIR-only and the highest values in FIR-present cases (overall P = 0.003 and P = 0.002, respectively). Culture-confirmed or clinically diagnosed EOS and microbiological positivity were more frequent in infants with inflammatory placental pathology.ConclusionsHistopathological placental inflammation in preterm infants was associated with higher umbilical cord PCT and higher postnatal CRP. Cord blood PCT showed the most consistent association with inflammatory placental pathology, including in exploratory analyses of fetal inflammatory response lesions. These findings support the potential clinical relevance of clinically available inflammatory biomarkers in relation to placental inflammatory pathology, but larger studies are needed to confirm lesion-specific associations and clarify their clinical utility.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s12884-026-09516-0. Background Preterm birth remains a leading cause of neonatal morbidity and mortality worldwide. Intrauterine infection and placental inflammation are major contributors to spontaneous preterm birth and are associated with a broad range of adverse neonatal outcomes [1–3]. Histopathological inflammatory lesions of the placenta, including acute chorioamnionitis and funiculitis, provide objective evidence of intrauterine inflammatory exposure [4–6]. These lesions are clinically important because they may reflect inflammatory processes involving the maternal compartment, the fetal compartment, or both, and have been associated with adverse neonatal outcomes in preterm infants [4–7]. The placenta is increasingly recognized as an active immunological interface between mother and fetus rather than a passive organ [8, 9]. In this setting, inflammatory activation within placental tissues may influence both the timing of delivery and the fetal response to intrauterine exposure. Distinguishing maternal inflammatory response (MIR) lesions from fetal inflammatory response (FIR) lesions is particularly important because fetal involvement is more strongly associated with deeper intrauterine inflammatory exposure and with neonatal complications [5–7]. At the same time, placental inflammation is biologically heterogeneous and does not necessarily imply proven infection, as non-infectious or sterile inflammatory pathways may also contribute, including mechanisms linked to NETosis [10, 11]. Although placental histopathology provides valuable information, the results are usually not available immediately after birth [12, 13]. Cord blood, by contrast, can be obtained promptly, painlessly, and before substantial postnatal influences become established. For this reason, cord blood biomarkers may provide clinically useful information about inflammatory exposure at birth and help link placental histopathology with the newborn’s early biological response [14]. Several studies have shown that histological chorioamnionitis may be accompanied by measurable inflammatory changes in fetal or neonatal circulation. Howman et al. reported that histological chorioamnionitis was associated with inflammatory and hematological alterations in maternal, cord, and neonatal blood [15]. D’Alquen et al. demonstrated that chorioamnionitis with funiculitis was associated with activation of umbilical cord endothelial cells and increased inflammatory mediators in preterm infants [16]. Other studies have shown that fetal immune activation in placental inflammation may also be reflected in cord blood cellular and alarmin-associated responses [17, 18]. Among clinically available inflammatory biomarkers, procalcitonin (PCT), presepsin (P-SEP), and C-reactive protein (CRP) exhibit distinct biological kinetics and have potential translational relevance for neonatal inflammatory assessment [19–21]. PCT has been widely studied as an early marker of systemic inflammatory activation; presepsin has emerged as a promising biomarker for risk assessment of neonatal early-onset sepsis; and hs-CRP remains a readily available acute-phase reactant despite its known kinetic limitations at birth [19–21]. Importantly, prior studies have suggested that PCT may be associated with histological chorioamnionitis and fetal inflammatory involvement. Stranak et al. reported that PCT during chorioamnionitis is more likely to be released by the fetus than by placental tissue [22]. Horinouchi et al. showed that PCT in umbilical cord blood and amniotic fluid may help predict histological chorioamnionitis and neonatal outcomes [23]. Kaneko et al. further reported that umbilical venous PCT reflected chorioamnionitis severity [24]. However, data directly comparing clinically available cord blood biomarkers with histopathological placental inflammation in preterm infants remain limited, particularly when inflammatory lesions are considered in relation to maternal versus fetal inflammatory components. The primary objective of this study was to determine whether histopathological placental inflammation is associated with cord blood inflammatory biomarkers (PCT, P-SEP, and hs-CRP) in preterm infants with birth weight < 2500 g. A secondary objective was to examine whether placental inflammatory pathology is associated with early postnatal inflammatory markers, particularly CRP measured within the first 24 h of life. We hypothesized that histopathological placental inflammation, particularly lesions involving the fetal inflammatory compartment, would be associated with higher cord blood PCT and higher early postnatal CRP in preterm infants. Methods Study design and setting A prospective observational cohort study was conducted in a Level III Neonatal Intensive Care Unit over a predefined one-year period (April 2022 to May 2023). All consecutive eligible preterm infants meeting the inclusion criteria during this interval were enrolled. Because this was a non-interventional, prospective, observational cohort study, clinical trial registration was not performed. Participants Eligible participants were preterm infants who fulfilled both inclusion criteria: gestational age < 37 weeks and birth weight < 2500 g. Exclusion criteria were major congenital anomalies identified on physical examination, lack of written informed consent from parents, and inadequate cord blood samples for biomarker analysis. Infants were followed prospectively until hospital discharge. The derivation of the final cohort and reasons for exclusion are reported in the Results section and summarized in Fig. 1. Fig. 1Flow diagram of participant selection and classification according to placental histopathology. Legend: During the predefined study period, 1990 live births occurred, including 186 preterm infants. Among these, 109 also met the birth weight criterion (< 2500 g). After exclusion of infants without parental consent, with major congenital anomalies, or with inadequate cord blood samples, 87 infants were included in the final cohort and classified according to placental histopathology Flow diagram of participant selection and classification according to placental histopathology. Legend: During the predefined study period, 1990 live births occurred, including 186 preterm infants. Among these, 109 also met the birth weight criterion (< 2500 g). After exclusion of infants without parental consent, with major congenital anomalies, or with inadequate cord blood samples, 87 infants were included in the final cohort and classified according to placental histopathology Ethics approval and consent to participate The study was approved by the Local Ethics Committee of the University Hospital (approval no. R2-1457/2022; approved on February 8, 2022) and was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from parents or legal guardians before inclusion. Clinical data collection Maternal and perinatal data were extracted from obstetric medical records and neonatal charts. Recorded variables included gestational age, birth weight, sex, duration of rupture of membranes (ROM, hours), intrapartum antibiotic exposure, antenatal corticosteroid administration, mode of delivery, and selected neonatal outcomes. Gestational age was obtained from obstetric records. In routine obstetric care in Croatia, the expected date of delivery is primarily determined by the last menstrual period, and all pregnant women undergo at least one first-trimester ultrasound examination in primary gynecologic care. However, ultrasound-based dating parameters were not systematically available for separate extraction from the study database. The available growth-restriction variable in the database was intrauterine growth restriction (IUGR), whereas small-for-gestational-age status was not separately coded. Additional neonatal infectious outcomes extracted from the database included culture-confirmed or clinically diagnosed early-onset sepsis (EOS), suspected EOS, positive microbiological findings, antibiotic exposure, and antibiotic duration. Additional neonatal morbidity outcomes included bronchopulmonary dysplasia (BPD), intracranial hemorrhage (ICH), retinopathy of prematurity (ROP), duration of oxygen therapy, and length of hospital stay. Intracranial hemorrhage was diagnosed by cranial ultrasound and graded according to the Papile classification. For additional descriptive analysis, hemorrhages were categorized as low grade (grade I–II) or high grade (grade III–IV). Periventricular leukomalacia (PVL) was not systematically available in the study database and therefore could not be analyzed. These variables were selected a priori as clinically relevant perinatal factors that could be associated with both placental inflammatory exposure and neonatal inflammatory biomarker levels. Clinical chorioamnionitis and preterm prelabor rupture of membranes (PPROM) were not systematically coded as separate variables in the study database. Review of available clinical records identified intrapartum maternal fever in only one case, precluding meaningful comparative analysis. Therefore, the duration of rupture of membranes was used as the available clinically relevant variable related to membrane rupture. Placental culture results were not available for analysis. Definitions of neonatal infection and microbiological findings Early-onset sepsis (EOS) was defined as sepsis occurring within the first 72 h of life. Culture-confirmed EOS was defined as clinical signs consistent with neonatal sepsis together with a positive blood culture obtained within the first 72 h of life. Clinically diagnosed EOS was defined as clinical presentation consistent with neonatal sepsis, accompanied by increased inflammatory markers and/or relevant perinatal risk factors, and antimicrobial treatment continued for at least 5 days according to the institutional neonatal sepsis protocol, regardless of culture confirmation. Culture-confirmed or clinically diagnosed EOS included both culture-confirmed EOS and clinically diagnosed EOS. Suspected EOS was defined as clinical suspicion of EOS leading to empirical antimicrobial treatment but not fulfilling criteria for culture-confirmed or clinically diagnosed EOS. Positive microbiological findings were recorded separately from EOS classification and defined as at least one positive neonatal microbiological sample recorded during hospitalization. This variable included microbiological findings recorded both within and after the first 72 h of life; culture-confirmed EOS was defined separately by positive blood culture within the first 72 h of life. Available microbiological data included blood cultures, catheter tip cultures, surveillance swabs, gastric aspirates, urine cultures, and other clinically obtained samples. Findings labelled as late-onset sepsis in the database were identified descriptively. Potential contamination was assessed from the available clinical and microbiological records where possible. Because the database did not allow uniform adjudication of all positive isolates according to timing, site, and contamination status, this variable was retained as a descriptive measure of neonatal microbiological positivity and was not equated with culture-confirmed EOS or microbiologically confirmed placental infection. Umbilical cord blood sampling and processing Umbilical venous cord blood was obtained in the delivery room immediately after cord clamping. Blood was collected into two 6 mL plastic tubes without anticoagulant (BD Vacutainer, Becton, Dickinson, and Company, Franklin Lakes, New York, USA). Samples were centrifuged at 1370 g for 10 min immediately after collection. Aliquots of serum were separated within 2 h and stored at − 80 °C until analysis. Samples were divided into three aliquots to avoid repeated freeze–thaw cycles. Biomarker analyses were performed after recruitment had been completed. Postnatal blood sampling Within the first 24 h of life, capillary blood micro-samples obtained by heel prick were collected into K3-EDTA microtainers (150–200 µL) and serum separator microtainers (150–200 µL) (BD Vacutainer, Becton, Dickinson and Company, Franklin Lakes, NJ, USA). Postnatal CRP sampling was performed as part of routine clinical care within the first 24 h of life and was not protocolized to a uniform postnatal time point. The exact postnatal sampling time in hours was recorded and used for descriptive and sensitivity analyses. Laboratory measurements Postnatal complete blood count was performed from K3-EDTA anticoagulated blood using a Sysmex XN-2000 hematology analyzer (Sysmex Corporation, Kobe, Japan). Umbilical cord blood high-sensitivity C-reactive protein (hs-CRP) and early postnatal C-reactive protein (CRP) were measured using the Beckman Coulter CRP Latex reagent OSR6 × 99 on an Olympus AU 680 biochemistry analyzer (Beckman Coulter, Brea, CA, USA). Results were reported in mg/L. Umbilical cord blood procalcitonin (PCT) was measured using an electrochemiluminescence immunoassay on a Cobas e411 immunochemistry analyzer (Roche Diagnostics, Basel, Switzerland). Results were reported in ng/mL. Umbilical cord blood presepsin (P-SEP) was measured by ELISA using the Human Presepsin ELISA Kit (Abbexa Ltd, Cambridge, UK) on an Eti-Max 3000 microplate reader (DiaSorin, Saluggia, Italy). Results were reported in pg/mL. Placental histopathology Placentas were submitted to the Department of Pathology and Forensic Medicine for routine histopathological examination. Placental sampling, lesion definitions, and reporting were guided by the Amsterdam Placental Workshop Group Consensus Statement and subsequent consensus recommendations [12, 13]. For the primary descriptive analysis, placental findings were categorized into three groups: (1) no pathological findings, (2) inflammatory placental pathology, and (3) other placental pathology. Inflammatory placental pathology included lesions consistent with a maternal or fetal inflammatory response, as described in pathology reports, including acute chorioamnionitis, funiculitis, and other inflammatory lesions such as villitis. No inflammatory placental abnormalities were grouped as „other placental pathology“ because of their heterogeneity and included lesions such as maternal vascular malperfusion, abruption-related changes, and other structural abnormalities. Pathologists were unaware of the biomarker results and study analyses. For additional exploratory analyses, inflammatory placental lesions were further classified according to the lesion type documented in the pathology report. Acute chorioamnionitis without funiculitis was considered consistent with isolated maternal inflammatory response (MIR-only), whereas any case with funiculitis was considered evidence of fetal inflammatory response (FIR-present). Chorioamnionitis with villitis was recorded separately. Because only one infant had chorioamnionitis with villitis, that subgroup was described descriptively but was not included in formal subgroup hypothesis testing. One twin pair shared an inflammatory placental diagnosis; the main analyses were performed at the child level, and a sensitivity analysis excluding one twin was additionally performed. Statistical analysis Categorical variables are presented as counts and percentages, and continuous variables as medians with interquartile ranges. Differences between groups were assessed using the chi-square test or Fisher’s exact test for categorical variables and the Kruskal–Wallis test with post hoc Conover testing for continuous variables. For two-group comparisons of continuous variables, the Mann–Whitney U test was used. Normality of distribution was assessed using the Shapiro–Wilk test. Because of skewed distributions, cord blood PCT and early postnatal CRP were natural log-transformed before regression analysis. Duration of rupture of membranes was modeled as ln(hours + 1). Primary descriptive analyses used the three prespecified placental pathology groups: no pathological findings, inflammatory placental pathology, and other placental pathology. To provide a more clinically direct comparison of inflammatory versus no inflammatory placental pathology, additional exploratory analyses recoded placental findings as either no inflammatory placental lesion or inflammatory placental pathology. Inflammatory cases were further subtyped as MIR-only, FIR-present, or chorioamnionitis with villitis. Multiple linear regression models with log-transformed outcomes were used to examine whether biomarker levels differed by binary inflammatory placental pathology status, after adjustment for selected perinatal covariates. The main explanatory variable was inflammatory placental pathology (yes/no). The selected covariates were gestational age, birth weight, duration of rupture of membranes, intrapartum antibiotic exposure, and antenatal corticosteroid exposure. These covariates were chosen a priori based on clinical plausibility and their potential association with both placental inflammation and neonatal inflammatory biomarker levels. Regression coefficients were exponentiated and are presented as fold changes with 95% confidence intervals. Model fit was assessed using R², adjusted R², and the F-statistic. Exploratory lesion-specific subgroup analyses compared biomarker levels across the no inflammatory placental lesion, MIR-only, and FIR-present groups. Because the number of infants with FIR-present lesions was small, these analyses were interpreted cautiously. Additional exploratory analyses examined neonatal infectious outcomes, microbiological findings, and neonatal morbidities according to inflammatory placental pathology status. Receiver operating characteristic (ROC) curve analysis was used to assess the discriminatory performance of inflammatory biomarkers for inflammatory placental pathology. Exploratory cut-off values were derived using the Youden index. Because postnatal CRP was not sampled at a protocolized time point, the distribution of CRP sampling times was described, and a sensitivity analysis was performed in the subgroup sampled between 16 and 23 h of life. A sensitivity analysis excluding one infant from the twin pair sharing an inflammatory placental diagnosis was also performed. Additional sensitivity analyses were performed to address microbiological and placental comparator issues. First, cord blood PCT was compared between inflammatory and no inflammatory placental lesions after exclusion of infants with positive microbiological findings. Second, adjusted regression models were repeated using the original three-category placental pathology variable, with no placental pathology as the reference group. Cases with missing covariate data were excluded from adjusted analyses. Given the modest number of infants with inflammatory placental pathology, all additional analyses were considered exploratory. All P values were two-sided, and statistical significance was defined as α = 0.05. Because this was an exploratory prospective cohort study that included all eligible infants during the study period, no formal a priori sample size calculation was performed. Statistical analyses were conducted using MedCalc Statistical Software v23.4.8 (MedCalc Software Ltd, Ostend, Belgium) and IBM SPSS Statistics version 26.0.0.0. Reporting The study report was prepared in accordance with the STROBE guidelines for observational studies, as recommended by the EQUATOR Network [25]. Results During the predefined study period, 1990 live births occurred, of which 186 were preterm. Among these, 109 infants also met the birth weight criterion (< 2500 g). After excluding infants without parental consent, those with major congenital anomalies, or those with inadequate cord blood samples, 87 infants were included in the final cohort (Fig. 1). In the primary descriptive analysis, 33 infants (38%) had no placental pathology, 15 (17%) had inflammatory placental pathology, and 39 (45%) had other placental pathology. Infants with inflammatory placental pathology had significantly lower gestational age (Kruskal–Wallis test, P = 0.04) and birth weight (Kruskal–Wallis test, P = 0.03) than the other groups. They were also more likely to be delivered vaginally (Chi-square test, P = 0.005). Length of hospital stay was longer in infants with inflammatory placental pathology than in the other groups (Kruskal–Wallis test, P = 0.006) (Table 1). No statistically significant differences were observed between groups for maternal age, sex, Apgar scores, intrauterine growth restriction, mortality before discharge, need for invasive mechanical ventilation, duration of mechanical ventilation, or duration of supplemental oxygen therapy. Table 1Clinical characteristics and outcomes by placental histopathology groupRisk factorA - No placental pathology(n = 33)B - Inflammatory placental pathology (n = 15)C - Other placental pathology(n = 39)P valueBirth weight (g), median (IQR)1965 (1720–2443)1750 (1456–2132)2190 (1912.5–2420) 0.03* § Gestational age (completed weeks + days), median (IQR)34 + 0 (32 + 3–35 + 3)31 + 0 (29 + 6–34 + 2)34 + 2 (32 + 6–35 + 3) 0.04* ǁ Maternal age (years), Median (IQR)30 (27–35)28 (24–33)31 (27–35)0.37*1-min Apgar score, median (IQR)10 (9–10)10 (9–10)10 (9–10)0.69*Vaginal delivery, n (%)15 (45.5)14 (93.3)19 (48.7) 0.005† IUGR, n (%)5 (15.15)1 (6.67)2 (5.13)0.38‡5-min Apgar score, median (IQR)10 (9–10)10 (9–10)10 (9–10)0.63*Sex - female, n (%)17 (51.52)8 (53.33)20 (51.28)0.99†Death until hospital discharge, n (%)002 (5.1)0.66‡Any invasive mechanical ventilation, n (%)2 (6.1)3 (20)3 (7.7)0.33‡Duration of mechanical ventilation (days), median (IQR) – in ventilated patients2 (1–2)5 (3–53)1 (1–5)0.63*Supplemental oxygen therapy (days), median (IQR)4 (2–8)13 (5–27)3 (1–7)0.47*Duration of hospital stay (days), median (IQR)13 (8–31)26 (16–44)10 (7–16) 0.006* ǁ Abbreviations: IQR interquartile range, IUGR intrauterine growth restrictionBold values denote statistical significance*Kruskal Wallis test (post hoc test Conover)†Chi-square test‡Fisher’s Exact test§significant differences (P < 0.05) for groups: (B) vs. (C)ǁsignificant differences (P < 0.05) for groups: (B) vs. (A, C) Table 1Clinical characteristics and outcomes by placental histopathology groupRisk factorA - No placental pathology(n = 33)B - Inflammatory placental pathology (n = 15)C - Other placental pathology(n = 39)P valueBirth weight (g), median (IQR)1965 (1720–2443)1750 (1456–2132)2190 (1912.5–2420) 0.03* § Gestational age (completed weeks + days), median (IQR)34 + 0 (32 + 3–35 + 3)31 + 0 (29 + 6–34 + 2)34 + 2 (32 + 6–35 + 3) 0.04* ǁ Maternal age (years), Median (IQR)30 (27–35)28 (24–33)31 (27–35)0.37*1-min Apgar score, median (IQR)10 (9–10)10 (9–10)10 (9–10)0.69*Vaginal delivery, n (%)15 (45.5)14 (93.3)19 (48.7) 0.005† IUGR, n (%)5 (15.15)1 (6.67)2 (5.13)0.38‡5-min Apgar score, median (IQR)10 (9–10)10 (9–10)10 (9–10)0.63*Sex - female, n (%)17 (51.52)8 (53.33)20 (51.28)0.99†Death until hospital discharge, n (%)002 (5.1)0.66‡Any invasive mechanical ventilation, n (%)2 (6.1)3 (20)3 (7.7)0.33‡Duration of mechanical ventilation (days), median (IQR) – in ventilated patients2 (1–2)5 (3–53)1 (1–5)0.63*Supplemental oxygen therapy (days), median (IQR)4 (2–8)13 (5–27)3 (1–7)0.47*Duration of hospital stay (days), median (IQR)13 (8–31)26 (16–44)10 (7–16) 0.006* ǁ Abbreviations: IQR interquartile range, IUGR intrauterine growth restrictionBold values denote statistical significance*Kruskal Wallis test (post hoc test Conover)†Chi-square test‡Fisher’s Exact test§significant differences (P < 0.05) for groups: (B) vs. (C)ǁsignificant differences (P < 0.05) for groups: (B) vs. (A, C) Clinical characteristics and outcomes by placental histopathology group Abbreviations: IQR interquartile range, IUGR intrauterine growth restriction Bold values denote statistical significance *Kruskal Wallis test (post hoc test Conover) †Chi-square test ‡Fisher’s Exact test §significant differences (P < 0.05) for groups: (B) vs. (C) ǁsignificant differences (P < 0.05) for groups: (B) vs. (A, C) Cord blood PCT levels differed significantly between the three primary placental pathology groups (Kruskal–Wallis test, P = 0.001), with the highest levels observed in infants with inflammatory placental pathology. No significant differences were observed between groups for cord blood presepsin (P-SEP) or hs-CRP. Early postnatal CRP levels were also higher in infants with inflammatory placental pathology (Kruskal–Wallis test, P = 0.004). No significant between-group differences were observed for postnatal leukocyte count, leukocyte extremes, or immature-to-total neutrophil ratio (Table 2). Table 2Umbilical cord blood biomarkers, early neonatal inflammatory markers, and placental histopathologyA - No placental pathology(n = 33)B - Inflammatory placental pathology (n = 15)C - Other placental pathology(n = 39)P valueUmbilical cord P-SEP (value), median (IQR)78.50 (39.5–112)39.50 (30.25–537.75)58.50 (43.25–128.25)0.96‡Umbilical cord PCT (value), median (IQR)0.10 (0.09–0.18)0.29 (0.165–0.78)0.12 (0.085–0.16) 0.001 ‡§ Umbilical cord hs-CRP (value), median (IQR)0.061 (0.034–0.123)0.069 (0.04–0.199)0.045 (0.033–0.1045)0.44‡Postnatal leukocyte count (value), median (IQR)12.2 (10.1–17.4)13.3 (11.95–17.95)11.8 (10.1–14.8)0.42‡Leukocytes < 5 or > 25 (yes)2 (6.1)1 (6.7)2 (5.1)> 0.99†Immature to total neutrophil ratio (I/T ratio), median (IQR)0.04 (0.02–0.06)0.04 (0.01–0.06)0.04 (0.02–0.08)0.49‡Postnatal C-reactive protein (value), median (IQR)0.7 (0.4–1.3)4.5 (1–10.7)1.1 (0.5–3.5) 0.004 ‡ǁ Abbreviations: IQR interquartile range, P-SEP presepsin (soluble CD14 subtype; sCD14-ST), PCT procalcitonin, hs-CRP high-sensitivity C-reactive protein, CRP C-reactive protein, I/T ratio immature-to-total neutrophil ratio, Units: P-SEP (pg/mL), PCT (ng/mL), hs-CRP and CRP (mg/L), leukocytes (×10⁹/L)Bold values denote statistical significance†Fisher’s Exact test‡Kruskal Wallis test (post hoc test Conover)§significant differences (P < 0.05) for groups: (B) vs. (A, C)ǁsignificant differences (P < 0.05) for groups: (A) vs. (B, C) Table 2Umbilical cord blood biomarkers, early neonatal inflammatory markers, and placental histopathologyA - No placental pathology(n = 33)B - Inflammatory placental pathology (n = 15)C - Other placental pathology(n = 39)P valueUmbilical cord P-SEP (value), median (IQR)78.50 (39.5–112)39.50 (30.25–537.75)58.50 (43.25–128.25)0.96‡Umbilical cord PCT (value), median (IQR)0.10 (0.09–0.18)0.29 (0.165–0.78)0.12 (0.085–0.16) 0.001 ‡§ Umbilical cord hs-CRP (value), median (IQR)0.061 (0.034–0.123)0.069 (0.04–0.199)0.045 (0.033–0.1045)0.44‡Postnatal leukocyte count (value), median (IQR)12.2 (10.1–17.4)13.3 (11.95–17.95)11.8 (10.1–14.8)0.42‡Leukocytes < 5 or > 25 (yes)2 (6.1)1 (6.7)2 (5.1)> 0.99†Immature to total neutrophil ratio (I/T ratio), median (IQR)0.04 (0.02–0.06)0.04 (0.01–0.06)0.04 (0.02–0.08)0.49‡Postnatal C-reactive protein (value), median (IQR)0.7 (0.4–1.3)4.5 (1–10.7)1.1 (0.5–3.5) 0.004 ‡ǁ Abbreviations: IQR interquartile range, P-SEP presepsin (soluble CD14 subtype; sCD14-ST), PCT procalcitonin, hs-CRP high-sensitivity C-reactive protein, CRP C-reactive protein, I/T ratio immature-to-total neutrophil ratio, Units: P-SEP (pg/mL), PCT (ng/mL), hs-CRP and CRP (mg/L), leukocytes (×10⁹/L)Bold values denote statistical significance†Fisher’s Exact test‡Kruskal Wallis test (post hoc test Conover)§significant differences (P < 0.05) for groups: (B) vs. (A, C)ǁsignificant differences (P < 0.05) for groups: (A) vs. (B, C) Umbilical cord blood biomarkers, early neonatal inflammatory markers, and placental histopathology Abbreviations: IQR interquartile range, P-SEP presepsin (soluble CD14 subtype; sCD14-ST), PCT procalcitonin, hs-CRP high-sensitivity C-reactive protein, CRP C-reactive protein, I/T ratio immature-to-total neutrophil ratio, Units: P-SEP (pg/mL), PCT (ng/mL), hs-CRP and CRP (mg/L), leukocytes (×10⁹/L) Bold values denote statistical significance †Fisher’s Exact test ‡Kruskal Wallis test (post hoc test Conover) §significant differences (P < 0.05) for groups: (B) vs. (A, C) ǁsignificant differences (P < 0.05) for groups: (A) vs. (B, C) For exploratory inflammatory-lesion analyses, placental findings were recoded as no inflammatory placental lesion (n = 72), MIR-only (n = 10), FIR-present (n = 4), and chorioamnionitis with villitis (n = 1). Cord blood PCT differed significantly across the no inflammation, MIR-only, and FIR-present subgroups (Kruskal–Wallis test, P = 0.003), with the highest concentrations observed in FIR-present cases. Median cord blood PCT was 0.115 ng/mL (IQR 0.09–0.18) in infants without inflammatory placental lesions, 0.23 ng/mL (IQR 0.138–0.412) in MIR-only cases, and 0.78 ng/mL (IQR 0.427–6.82) in FIR-present cases. Pairwise comparisons showed higher PCT in MIR-only than in no inflammatory placental lesions (P = 0.015) and higher PCT in FIR-present than in no inflammatory placental lesions (P = 0.011). Cord blood P-SEP did not differ significantly across lesion subtypes (P = 0.767). Cord blood hs-CRP did not show a significant overall difference (P = 0.082), although values were higher in the FIR-present subgroup than in the no inflammatory placental lesion group in exploratory pairwise comparison (P = 0.036). Early postnatal CRP also differed significantly across lesion subtypes (Kruskal–Wallis test, P = 0.002), with median values of 0.8 mg/L (IQR 0.5–1.9) in infants without inflammatory placental lesions, 4.3 mg/L (IQR 1.0–8.8) in MIR-only cases, and 14.3 mg/L (IQR 8.97–17.98) in FIR-present cases (Table 3). Table 3Inflammatory biomarkers by placental inflammatory lesion subtypeBiomarkerNo inflammation(n = 72)MIR-only(n = 10)FIR-present(n = 4)OverallP valueUmbilical cord PCT (ng/mL), median (IQR)0.115 (0.09–0.18)0.23 (0.138–0.412)0.78 (0.427–6.82)0.003Umbilical cord P-SEP (pg/mL), median (IQR)71.8 (40.6–121.4)75.0 (30.6–1567.4)191.8 (45.1–437.1)0.767Umbilical cord hs-CRP (mg/L), median (IQR)0.049 (0.034–0.097)0.074 (0.037–0.196)0.58 (0.165–1.38)0.082Early postnatal CRP (mg/L), median (IQR)0.8 (0.5–1.9)4.3 (1.0–8.8)14.3 (8.97–17.98)0.002Exploratory subgroup analysis of inflammatory biomarkers according to lesion subtype. Acute chorioamnionitis without funiculitis was classified as MIR-only, and any case with funiculitis was classified as FIR-present. Because only one infant had chorioamnionitis with villitis, that case was described separately and was not included in formal subgroup hypothesis testing. Continuous variables are presented as medians with interquartile rangesOverall comparisons were performed using the Kruskal–Wallis test. Exploratory pairwise comparisons were performed only where clinically meaningful and are reported in the text or footnotes. Abbreviations: FIR fetal inflammatory response, hs-CRP high-sensitivity C-reactive protein, IQR interquartile range, MIR maternal inflammatory response, PCT procalcitonin, P-SEP presepsin. One additional case of chorioamnionitis with villitis is presented descriptively only and was not included in formal subgroup testing because of the very small subgroup size Table 3Inflammatory biomarkers by placental inflammatory lesion subtypeBiomarkerNo inflammation(n = 72)MIR-only(n = 10)FIR-present(n = 4)OverallP valueUmbilical cord PCT (ng/mL), median (IQR)0.115 (0.09–0.18)0.23 (0.138–0.412)0.78 (0.427–6.82)0.003Umbilical cord P-SEP (pg/mL), median (IQR)71.8 (40.6–121.4)75.0 (30.6–1567.4)191.8 (45.1–437.1)0.767Umbilical cord hs-CRP (mg/L), median (IQR)0.049 (0.034–0.097)0.074 (0.037–0.196)0.58 (0.165–1.38)0.082Early postnatal CRP (mg/L), median (IQR)0.8 (0.5–1.9)4.3 (1.0–8.8)14.3 (8.97–17.98)0.002Exploratory subgroup analysis of inflammatory biomarkers according to lesion subtype. Acute chorioamnionitis without funiculitis was classified as MIR-only, and any case with funiculitis was classified as FIR-present. Because only one infant had chorioamnionitis with villitis, that case was described separately and was not included in formal subgroup hypothesis testing. Continuous variables are presented as medians with interquartile rangesOverall comparisons were performed using the Kruskal–Wallis test. Exploratory pairwise comparisons were performed only where clinically meaningful and are reported in the text or footnotes. Abbreviations: FIR fetal inflammatory response, hs-CRP high-sensitivity C-reactive protein, IQR interquartile range, MIR maternal inflammatory response, PCT procalcitonin, P-SEP presepsin. One additional case of chorioamnionitis with villitis is presented descriptively only and was not included in formal subgroup testing because of the very small subgroup size Inflammatory biomarkers by placental inflammatory lesion subtype Exploratory subgroup analysis of inflammatory biomarkers according to lesion subtype. Acute chorioamnionitis without funiculitis was classified as MIR-only, and any case with funiculitis was classified as FIR-present. Because only one infant had chorioamnionitis with villitis, that case was described separately and was not included in formal subgroup hypothesis testing. Continuous variables are presented as medians with interquartile ranges Overall comparisons were performed using the Kruskal–Wallis test. Exploratory pairwise comparisons were performed only where clinically meaningful and are reported in the text or footnotes. Abbreviations: FIR fetal inflammatory response, hs-CRP high-sensitivity C-reactive protein, IQR interquartile range, MIR maternal inflammatory response, PCT procalcitonin, P-SEP presepsin. One additional case of chorioamnionitis with villitis is presented descriptively only and was not included in formal subgroup testing because of the very small subgroup size In binary comparisons, any inflammatory placental pathology was associated with higher cord blood PCT and higher early postnatal CRP than in the absence of inflammatory placental lesions. Median cord blood PCT was 0.29 ng/mL (IQR 0.145–0.78) in inflammatory placental pathology versus 0.115 ng/mL (IQR 0.09–0.18) in no inflammatory placental lesion (Mann–Whitney U test, P < 0.001). Early postnatal CRP was 4.9 mg/L (IQR 1.8–10.9) versus 0.8 mg/L (IQR 0.5–1.9), respectively (P < 0.001). No significant differences were observed for cord blood P-SEP (P = 0.906) or hs-CRP (P = 0.101). In adjusted binary models, inflammatory placental pathology remained independently associated with higher cord blood PCT and higher early postnatal CRP. In the cord blood PCT model, inflammatory placental pathology was associated with a 2.31-fold higher PCT concentration (95% CI 1.50–3.55, P < 0.001; R² = 0.415, adjusted R² = 0.372). In the early postnatal CRP model, inflammatory placental pathology was associated with a 4.66-fold higher CRP concentration (95% CI 2.23–9.76, P < 0.001; R² = 0.250, adjusted R² = 0.194). Gestational age was independently associated with lower early postnatal CRP, whereas higher birth weight was associated with higher early postnatal CRP (Tables 4 and 5). Table 4Adjusted binary model for cord blood PCT: inflammatory vs. non-inflammatory placental pathologyPredictorAdjusted fold change (expβ)95% CIP valueInflammatory placental pathology(yes vs. no inflammatory lesion)2.311.50–3.55< 0.001Gestational age (per 1 week)0.910.81–1.030.146Birth weight (per 100 g)0.990.92–1.070.785Duration of ROM (log[hours + 1])1.000.92–1.100.921Intrapartum antibiotics (yes)0.830.57–1.200.321Antenatal corticosteroids (yes)1.460.97–2.190.069Multiple linear regression model with log-transformed umbilical cord PCT as the dependent variable. The main explanatory variable was inflammatory placental pathology (yes/no). Covariates included gestational age, birth weight, duration of rupture of membranes, intrapartum antibiotic exposure, and antenatal corticosteroid exposure. Regression coefficients are presented as exponentiated values (adjusted fold change) with 95% confidence intervals. Model fit statistics are provided to indicate the proportion of variance explained by the modelDependent variable: ln(umbilical cord PCT [ng/mL]). Gestational age is expressed per 1-week increase; birth weight per 100 g increase. Duration of rupture of membranes (ROM) was modeled as ln(ROM hours + 1). Abbreviations: CI confidence interval, PCT procalcitonin, ROM rupture of membranes. Model fit: R2 = 0.415; adjusted R2 = 0.372; F-statistic = 9.48; model P < 0.001 Table 4Adjusted binary model for cord blood PCT: inflammatory vs. non-inflammatory placental pathologyPredictorAdjusted fold change (expβ)95% CIP valueInflammatory placental pathology(yes vs. no inflammatory lesion)2.311.50–3.55< 0.001Gestational age (per 1 week)0.910.81–1.030.146Birth weight (per 100 g)0.990.92–1.070.785Duration of ROM (log[hours + 1])1.000.92–1.100.921Intrapartum antibiotics (yes)0.830.57–1.200.321Antenatal corticosteroids (yes)1.460.97–2.190.069Multiple linear regression model with log-transformed umbilical cord PCT as the dependent variable. The main explanatory variable was inflammatory placental pathology (yes/no). Covariates included gestational age, birth weight, duration of rupture of membranes, intrapartum antibiotic exposure, and antenatal corticosteroid exposure. Regression coefficients are presented as exponentiated values (adjusted fold change) with 95% confidence intervals. Model fit statistics are provided to indicate the proportion of variance explained by the modelDependent variable: ln(umbilical cord PCT [ng/mL]). Gestational age is expressed per 1-week increase; birth weight per 100 g increase. Duration of rupture of membranes (ROM) was modeled as ln(ROM hours + 1). Abbreviations: CI confidence interval, PCT procalcitonin, ROM rupture of membranes. Model fit: R2 = 0.415; adjusted R2 = 0.372; F-statistic = 9.48; model P < 0.001 Adjusted binary model for cord blood PCT: inflammatory vs. non-inflammatory placental pathology Inflammatory placental pathology (yes vs. no inflammatory lesion) Multiple linear regression model with log-transformed umbilical cord PCT as the dependent variable. The main explanatory variable was inflammatory placental pathology (yes/no). Covariates included gestational age, birth weight, duration of rupture of membranes, intrapartum antibiotic exposure, and antenatal corticosteroid exposure. Regression coefficients are presented as exponentiated values (adjusted fold change) with 95% confidence intervals. Model fit statistics are provided to indicate the proportion of variance explained by the model Dependent variable: ln(umbilical cord PCT [ng/mL]). Gestational age is expressed per 1-week increase; birth weight per 100 g increase. Duration of rupture of membranes (ROM) was modeled as ln(ROM hours + 1). Abbreviations: CI confidence interval, PCT procalcitonin, ROM rupture of membranes. Model fit: R2 = 0.415; adjusted R2 = 0.372; F-statistic = 9.48; model P < 0.001 Table 5Association of inflammatory placental pathology with early postnatal CRP in adjusted analysisPredictorAdjusted fold change (expβ)95% CIP valueInflammatory placental pathology(yes vs. no inflammatory lesion)4.662.23–9.76< 0.001Gestational age (per 1 week)0.770.62–0.960.018Birth weight (per 100 g)1.201.06–1.370.004Duration of ROM (log[hours + 1])0.880.76–1.020.097Intrapartum antibiotics (yes)0.990.53–1.860.968Antenatal corticosteroids (yes)0.910.45–1.810.779Multiple linear regression model with log-transformed early postnatal CRP as the dependent variable. The main explanatory variable was inflammatory placental pathology (yes/no). Covariates included gestational age, birth weight, duration of rupture of membranes, intrapartum antibiotic exposure, and antenatal corticosteroid exposure. Regression coefficients are presented as exponentiated values (adjusted fold change) with 95% confidence intervals. Model fit statistics are provided to indicate the proportion of variance explained by the modelDependent variable: ln(early postnatal CRP [mg/L]; first value obtained within 24 hours of life). Gestational age is expressed per 1-week increase; birth weight per 100 g increase. Duration of rupture of membranes (ROM) was modeled as ln(ROM hours + 1). Abbreviations: CI confidence interval, CRP C-reactive protein, ROM rupture of membranes. Model fit: R2 = 0.250; adjusted R2 = 0.194; F-statistic = 4.45; model P < 0.001 Table 5Association of inflammatory placental pathology with early postnatal CRP in adjusted analysisPredictorAdjusted fold change (expβ)95% CIP valueInflammatory placental pathology(yes vs. no inflammatory lesion)4.662.23–9.76< 0.001Gestational age (per 1 week)0.770.62–0.960.018Birth weight (per 100 g)1.201.06–1.370.004Duration of ROM (log[hours + 1])0.880.76–1.020.097Intrapartum antibiotics (yes)0.990.53–1.860.968Antenatal corticosteroids (yes)0.910.45–1.810.779Multiple linear regression model with log-transformed early postnatal CRP as the dependent variable. The main explanatory variable was inflammatory placental pathology (yes/no). Covariates included gestational age, birth weight, duration of rupture of membranes, intrapartum antibiotic exposure, and antenatal corticosteroid exposure. Regression coefficients are presented as exponentiated values (adjusted fold change) with 95% confidence intervals. Model fit statistics are provided to indicate the proportion of variance explained by the modelDependent variable: ln(early postnatal CRP [mg/L]; first value obtained within 24 hours of life). Gestational age is expressed per 1-week increase; birth weight per 100 g increase. Duration of rupture of membranes (ROM) was modeled as ln(ROM hours + 1). Abbreviations: CI confidence interval, CRP C-reactive protein, ROM rupture of membranes. Model fit: R2 = 0.250; adjusted R2 = 0.194; F-statistic = 4.45; model P < 0.001 Association of inflammatory placental pathology with early postnatal CRP in adjusted analysis Inflammatory placental pathology (yes vs. no inflammatory lesion) Multiple linear regression model with log-transformed early postnatal CRP as the dependent variable. The main explanatory variable was inflammatory placental pathology (yes/no). Covariates included gestational age, birth weight, duration of rupture of membranes, intrapartum antibiotic exposure, and antenatal corticosteroid exposure. Regression coefficients are presented as exponentiated values (adjusted fold change) with 95% confidence intervals. Model fit statistics are provided to indicate the proportion of variance explained by the model Dependent variable: ln(early postnatal CRP [mg/L]; first value obtained within 24 hours of life). Gestational age is expressed per 1-week increase; birth weight per 100 g increase. Duration of rupture of membranes (ROM) was modeled as ln(ROM hours + 1). Abbreviations: CI confidence interval, CRP C-reactive protein, ROM rupture of membranes. Model fit: R2 = 0.250; adjusted R2 = 0.194; F-statistic = 4.45; model P < 0.001 Infants with inflammatory placental pathology had a higher burden of infectious outcomes than those without inflammatory placental lesions. Culture-confirmed or clinically diagnosed EOS occurred in 7/15 (46.7%) infants with inflammatory placental pathology versus 10/72 (13.9%) infants without inflammatory placental lesions (Fisher’s exact test, P = 0.008). When suspected EOS was also included, EOS-related clinical categories were more frequent in inflammatory placental pathology than in no inflammatory placental lesions (12/15, 80.0% vs. 37/72, 51.4%; P = 0.049). Positive microbiological findings were more frequent in inflammatory placental pathology (9/15, 60.0%) than in no inflammatory placental lesions (10/72, 13.9%) (P < 0.001), and antibiotic treatment duration was longer in the inflammatory group (median 9 vs. 6 days, P = 0.005) (Table 6). Among infants with positive microbiological findings, the most frequently recorded organisms or findings included methicillin-resistant Staphylococcus epidermidis, Escherichia coli, Enterococcus faecalis, Enterococcus faecium, Klebsiella pneumoniae, Staphylococcus aureus, and group B Streptococcus. Table 6Neonatal infectious outcomes by inflammatory placental pathologyOutcomeNo inflammatory placental lesion (n = 72)Any inflammatory placental pathology (n = 15)P valueCulture-confirmed or clinically diagnosed EOS, n (%)10/72 (13.9%)7/15 (46.7%)0.008EOS-related clinical category including suspected EOS, n (%)37/72 (51.4%)12/15 (80.0%)0.049Positive microbiological findings, n (%)10/72 (13.9%)9/15 (60.0%)< 0.001Antibiotic treatment, n (%)55/72 (76.4%)14/15 (93.3%)0.179Antibiotic duration (days), median (IQR)6 (1–7)9 (7–11.5)0.005Comparison of neonatal infectious outcomes and microbiological findings between infants with inflammatory placental pathology and those without inflammatory placental lesions. Categorical variables are presented as counts and percentages, and continuous variables as medians with interquartile rangesCulture-confirmed EOS was defined as clinical signs consistent with neonatal sepsis together with a positive blood culture obtained within the first 72 hours of life. Clinically diagnosed EOS was defined as clinical presentation consistent with neonatal sepsis, increased inflammatory markers and/or relevant perinatal risk factors, and antimicrobial treatment continued for at least 5 days according to the institutional neonatal sepsis protocol, regardless of culture confirmation. Suspected EOS was defined as clinical suspicion of EOS leading to empirical antimicrobial treatment without fulfilment of criteria for culture-confirmed or clinically diagnosed EOS. Positive microbiological findings were recorded separately from EOS classification and defined as at least one positive neonatal microbiological sample recorded during hospitalization. Categorical variables were compared using Fisher’s exact test; continuous variables were compared using the Mann–Whitney U test. Abbreviations: EOS early-onset sepsis, IQR interquartile range Table 6Neonatal infectious outcomes by inflammatory placental pathologyOutcomeNo inflammatory placental lesion (n = 72)Any inflammatory placental pathology (n = 15)P valueCulture-confirmed or clinically diagnosed EOS, n (%)10/72 (13.9%)7/15 (46.7%)0.008EOS-related clinical category including suspected EOS, n (%)37/72 (51.4%)12/15 (80.0%)0.049Positive microbiological findings, n (%)10/72 (13.9%)9/15 (60.0%)< 0.001Antibiotic treatment, n (%)55/72 (76.4%)14/15 (93.3%)0.179Antibiotic duration (days), median (IQR)6 (1–7)9 (7–11.5)0.005Comparison of neonatal infectious outcomes and microbiological findings between infants with inflammatory placental pathology and those without inflammatory placental lesions. Categorical variables are presented as counts and percentages, and continuous variables as medians with interquartile rangesCulture-confirmed EOS was defined as clinical signs consistent with neonatal sepsis together with a positive blood culture obtained within the first 72 hours of life. Clinically diagnosed EOS was defined as clinical presentation consistent with neonatal sepsis, increased inflammatory markers and/or relevant perinatal risk factors, and antimicrobial treatment continued for at least 5 days according to the institutional neonatal sepsis protocol, regardless of culture confirmation. Suspected EOS was defined as clinical suspicion of EOS leading to empirical antimicrobial treatment without fulfilment of criteria for culture-confirmed or clinically diagnosed EOS. Positive microbiological findings were recorded separately from EOS classification and defined as at least one positive neonatal microbiological sample recorded during hospitalization. Categorical variables were compared using Fisher’s exact test; continuous variables were compared using the Mann–Whitney U test. Abbreviations: EOS early-onset sepsis, IQR interquartile range Neonatal infectious outcomes by inflammatory placental pathology Comparison of neonatal infectious outcomes and microbiological findings between infants with inflammatory placental pathology and those without inflammatory placental lesions. Categorical variables are presented as counts and percentages, and continuous variables as medians with interquartile ranges Culture-confirmed EOS was defined as clinical signs consistent with neonatal sepsis together with a positive blood culture obtained within the first 72 hours of life. Clinically diagnosed EOS was defined as clinical presentation consistent with neonatal sepsis, increased inflammatory markers and/or relevant perinatal risk factors, and antimicrobial treatment continued for at least 5 days according to the institutional neonatal sepsis protocol, regardless of culture confirmation. Suspected EOS was defined as clinical suspicion of EOS leading to empirical antimicrobial treatment without fulfilment of criteria for culture-confirmed or clinically diagnosed EOS. Positive microbiological findings were recorded separately from EOS classification and defined as at least one positive neonatal microbiological sample recorded during hospitalization. Categorical variables were compared using Fisher’s exact test; continuous variables were compared using the Mann–Whitney U test. Abbreviations: EOS early-onset sepsis, IQR interquartile range Neonatal morbidity analyses showed no statistically significant differences in the incidence of BPD or ROP. Overall ICH was more frequent in infants with inflammatory placental pathology, but the difference did not reach conventional statistical significance (13/15, 86.7% vs. 43/72, 59.7%; P = 0.073). Most hemorrhages were low grade. Grade I–II ICH occurred in 11/15 (73.3%) infants with inflammatory placental pathology and 41/72 (56.9%) infants without inflammatory placental lesions (P = 0.386), whereas grade III–IV ICH occurred in 2/15 (13.3%) and 2/72 (2.8%), respectively (P = 0.136). Overall, four infants had grade III ICH, and no grade IV ICH was recorded. ROP occurred in 1/15 (6.7%) versus 5/72 (6.9%) (P = 1.0). However, infants with inflammatory placental pathology had a longer duration of oxygen therapy (median 13 days [IQR 5–26.5] vs. 4 days [IQR 1–8], P = 0.014) and longer hospitalization (median 18 days [IQR 16–43.5] vs. 11 days [IQR 7–21], P = 0.008) (Table 7). Table 7Neonatal morbidities by inflammatory placental pathologyOutcomeNo inflammatory placental lesion (n = 72)Any inflammatory placental pathology (n = 15)P valueBPD, n (%)1/72 (1.4%)1/15 (6.7%)0.317Any ICH, n (%)43/72 (59.7%)13/15 (86.7%)0.073ICH grade I-II, n (%)41/72 (56.9%)11/15 (73.3%)0.386ICH grade III-IV, n (%)2/72 (2.8%)2/15 (13.3%)0.136ROP, n (%)5/72 (6.9%)1/15 (6.7%)1.000Oxygen therapy (days), median (IQR)4 (1–8)13 (5–26.5)0.014Hospital stay (days), median (IQR)11 (7–21)18 (16–43.5)0.008Comparison of selected neonatal morbidities and clinical course indicators between infants with inflammatory placental pathology and those without inflammatory placental lesions. Categorical variables are presented as counts and percentages, and continuous variables as medians with interquartile ranges. Intracranial hemorrhage was diagnosed by cranial ultrasound and stratified as grade I–II or grade III–IV according to the Papile classificationCategorical variables were compared using Fisher’s exact test; continuous variables were compared using the Mann–Whitney U test. Abbreviations: BPD bronchopulmonary dysplasia, ICH intracranial hemorrhage, IQR interquartile range, ROP retinopathy of prematurity Table 7Neonatal morbidities by inflammatory placental pathologyOutcomeNo inflammatory placental lesion (n = 72)Any inflammatory placental pathology (n = 15)P valueBPD, n (%)1/72 (1.4%)1/15 (6.7%)0.317Any ICH, n (%)43/72 (59.7%)13/15 (86.7%)0.073ICH grade I-II, n (%)41/72 (56.9%)11/15 (73.3%)0.386ICH grade III-IV, n (%)2/72 (2.8%)2/15 (13.3%)0.136ROP, n (%)5/72 (6.9%)1/15 (6.7%)1.000Oxygen therapy (days), median (IQR)4 (1–8)13 (5–26.5)0.014Hospital stay (days), median (IQR)11 (7–21)18 (16–43.5)0.008Comparison of selected neonatal morbidities and clinical course indicators between infants with inflammatory placental pathology and those without inflammatory placental lesions. Categorical variables are presented as counts and percentages, and continuous variables as medians with interquartile ranges. Intracranial hemorrhage was diagnosed by cranial ultrasound and stratified as grade I–II or grade III–IV according to the Papile classificationCategorical variables were compared using Fisher’s exact test; continuous variables were compared using the Mann–Whitney U test. Abbreviations: BPD bronchopulmonary dysplasia, ICH intracranial hemorrhage, IQR interquartile range, ROP retinopathy of prematurity Neonatal morbidities by inflammatory placental pathology Comparison of selected neonatal morbidities and clinical course indicators between infants with inflammatory placental pathology and those without inflammatory placental lesions. Categorical variables are presented as counts and percentages, and continuous variables as medians with interquartile ranges. Intracranial hemorrhage was diagnosed by cranial ultrasound and stratified as grade I–II or grade III–IV according to the Papile classification Categorical variables were compared using Fisher’s exact test; continuous variables were compared using the Mann–Whitney U test. Abbreviations: BPD bronchopulmonary dysplasia, ICH intracranial hemorrhage, IQR interquartile range, ROP retinopathy of prematurity Postnatal CRP was sampled between 13 and 23 h of life, with a median sampling time of 18 h (IQR 15.5–20). Sampling time did not differ significantly between inflammatory and no inflammatory placental groups (18 h [IQR 15–18.5] vs. 18 h [IQR 16–20], P = 0.384). In a sensitivity analysis restricted to samples obtained between 16 and 23 h of life (n = 65; inflammatory placental pathology n = 10), early postnatal CRP remained higher in inflammatory placental pathology than in no inflammatory placental lesion (5.05 mg/L [IQR 1.88–10.52] vs. 0.7 mg/L [IQR 0.45–1.95], P = 0.009). In an adjusted sensitivity model restricted to this time window, inflammatory placental pathology remained associated with higher early postnatal CRP (adjusted fold change 3.76, 95% CI 1.53–9.24, P = 0.005) (Supplementary Table S1). ROC analyses showed that cord blood PCT and early postnatal CRP had the best discriminatory performance for inflammatory placental pathology. The AUC was 0.787 (95% CI 0.628–0.924) for cord blood PCT and 0.785 (95% CI 0.661–0.891) for early postnatal CRP. Discrimination was poor for cord blood P-SEP (AUC 0.510, 95% CI 0.304–0.719) and modest for cord blood hs-CRP (AUC 0.636, 95% CI 0.475–0.790) (Supplementary Table S2). Exploratory Youden-based cutoffs were 0.25 ng/mL for cord blood PCT and 2.6 mg/L for early postnatal CRP. Finally, a sensitivity analysis excluding one infant from the twin pair sharing an inflammatory placental diagnosis did not materially change the association between placental inflammation and either cord blood PCT or early postnatal CRP (Supplementary Table S3). After exclusion of infants with positive microbiological findings, cord blood PCT remained higher in inflammatory placental pathology than in no inflammatory placental lesions (0.23 [IQR 0.163–0.402] vs. 0.10 [IQR 0.09–0.158] ng/mL; P = 0.038; Supplementary Table S4). In sensitivity analyses using the original three-category placental pathology variable and no placental pathology as the reference group, inflammatory placental pathology remained associated with higher cord blood PCT (adjusted fold change 2.21, 95% CI 1.39–3.53; P = 0.001) and higher early postnatal CRP (adjusted fold change 5.41, 95% CI 2.44–11.97; P < 0.001). Other placental pathology was not associated with comparable biomarker elevation (Supplementary Table S5). Discussion Intrauterine inflammation is a major contributor to preterm birth and adverse neonatal outcomes, but its biological basis is heterogeneous and extends beyond the classical concept of overt infection [1–3]. Histopathological inflammatory lesions of the placenta provide objective evidence of intrauterine inflammatory exposure and may reflect processes involving the maternal compartment, the fetal compartment, or both [4–6]. In this context, there is growing interest in identifying clinically accessible biomarkers that can be measured at birth, and that may complement placental histopathology in the early assessment of preterm infants. The principal finding of this study is that histopathological placental inflammation was associated with higher umbilical cord blood PCT levels in preterm infants. This association remained significant after adjustment for selected perinatal covariates, with inflammatory placental pathology associated with an approximately 2.3-fold higher cord blood PCT concentration in the adjusted model. In parallel, inflammatory placental pathology was associated with an approximately 4.7-fold higher early postnatal CRP concentration. Because PCT and CRP are clinically available biomarkers that are already used in neonatal inflammatory evaluation, these findings may be of translational interest. However, our results should be interpreted as evidence of association rather than as proof of a specific fetal inflammatory syndrome or as a stand-alone basis for treatment decisions. An important limitation in this regard is that umbilical cord IL-6, a commonly used reference biomarker of fetal inflammatory response syndrome (FIRS), was not measured in our cohort [7]. Therefore, elevated cord blood PCT cannot be interpreted as a surrogate diagnostic marker of FIRS. Rather, our findings indicate that higher cord blood PCT was associated with placental inflammatory pathology in this preterm population. This distinction is important because placental inflammatory lesions are biologically heterogeneous and do not necessarily correspond to the same degree or type of fetal inflammatory activation. A major strength of the revised analysis is that inflammatory placental lesions can be further explored by lesion subtype. In exploratory subgroup analyses, cord blood PCT showed a stepwise increase from infants without inflammatory placental lesions to those with isolated maternal inflammatory response (MIR-only) and was highest in those with fetal inflammatory response (FIR-present) lesions. Early postnatal CRP showed a similar gradient. Although the number of FIR-present cases was small, this lesion-specific pattern strengthens the biological plausibility of cord blood PCT as a marker of fetal involvement rather than merely non-specific placental inflammation. This is clinically relevant because the distinction between MIR and FIR has prognostic implications, with fetal inflammatory involvement generally considered the more advanced and clinically consequential stage of intrauterine inflammatory activation. We also found that infants with inflammatory placental pathology had a substantially higher burden of neonatal infectious outcomes. Culture-confirmed or clinically diagnosed EOS occurred more frequently in infants with inflammatory placental pathology than in those without inflammatory placental lesions, and this difference persisted even when suspected EOS was included. Positive microbiological findings were also markedly more common in the inflammatory group, and antibiotic treatment duration was longer. These results suggest that the observed inflammatory biomarker signal was not merely an isolated laboratory phenomenon but was associated with a clinically relevant infectious phenotype in the neonatal period. At the same time, EOS classification in this study included both culture-confirmed and clinically diagnosed cases treated according to the institutional neonatal sepsis protocol, while positive microbiological findings were recorded separately from EOS classification. Therefore, positive microbiological findings should be interpreted as a broader descriptive measure of neonatal microbiological positivity rather than as a microbiological gold standard for placental infection. In addition, placental culture results were not available, and clinical chorioamnionitis was not systematically coded in the study database. Therefore, we could not directly distinguish placental inflammation caused by proven microbial invasion from sterile placental inflammation. Importantly, the association between inflammatory placental pathology and higher cord blood PCT levels persisted after excluding infants with positive neonatal microbiological findings. This suggests that the observed PCT signal was not solely driven by microbiologically positive neonatal infection, although the small number of remaining infants with inflammatory placental pathology means that this sensitivity analysis should be interpreted cautiously. We also addressed the potential heterogeneity of the pooled binary reference group. In the main binary models, infants with no placental pathology and those with other non-inflammatory placental pathology were combined as the reference group in order to evaluate the clinically direct contrast between inflammatory and no inflammatory placental lesions. However, this reference group is biologically heterogeneous. Therefore, adjusted models were repeated using the original three-category placental pathology variable, with no placental pathology as the reference category. In these sensitivity models, inflammatory placental pathology remained associated with higher cord blood PCT and higher early postnatal CRP, whereas other placental pathology did not show a comparable biomarker pattern. These findings support the robustness of the main association while acknowledging that non-inflammatory placental pathology represents a heterogeneous comparator. In contrast, cord blood P-SEP and hs-CRP did not differ significantly across overall placental pathology categories, and P-SEP also showed poor discriminatory performance in ROC analysis. Several explanations are possible. First, placental inflammation is not synonymous with bacterial infection, and sterile or non-infectious inflammatory pathways, including NETosis-related mechanisms, may produce biomarker-specific response patterns [10, 11, 26–31]. Second, although cord blood presepsin has shown promise in assessing early-onset neonatal sepsis risk [20], it may not reflect placental inflammatory pathology as effectively as PCT. P-SEP is more closely linked to monocyte/macrophage CD14-related pathways and may therefore be less consistently elevated in a heterogeneous placental inflammatory environment that includes non-infectious lesions. Third, the relatively small number of infants with inflammatory placental pathology may have limited our ability to detect more modest associations for biomarkers with greater inter-individual variability. These negative findings should therefore be interpreted cautiously. The postnatal CRP finding deserves particular comment. CRP sampling was not protocolized to a single postnatal time point, which was an important concern raised by the reviewers. However, in the revised analysis, we found that postnatal CRP was sampled within a relatively narrow time window, between 13 and 23 h of life, with a median of 18 h, and sampling time did not differ significantly between inflammatory and no inflammatory placental groups. Moreover, in a sensitivity analysis restricted to samples obtained between 16 and 23 h of life, the association between inflammatory placental pathology and higher early postnatal CRP remained significant. This does not eliminate all timing-related confounding, but it strengthens confidence that the observed CRP signal was not driven simply by systematic differences in sampling time. Our ROC analyses further support the differential performance of the evaluated biomarkers. Cord blood PCT showed the best discrimination for inflammatory placental pathology, while early postnatal CRP showed similar discrimination. By contrast, discrimination was poor for cord blood P-SEP and only modest for cord blood hs-CRP. These findings are consistent with the lesion-specific analyses and support the interpretation that PCT was the biomarker most consistently linked to placental inflammatory pathology in this cohort. Nevertheless, our study was not designed to define clinically actionable cord blood PCT cutoffs for placental inflammatory pathology, and the present results should therefore be interpreted at the group level rather than as a basis for diagnostic threshold definition. Any cutoffs derived from ROC analysis should be regarded as exploratory only. Our results are also broadly consistent with prior work showing that histological placental inflammation can be accompanied by measurable inflammatory changes on the fetal side of the maternal-fetal interface. Howman et al. reported that histological chorioamnionitis was associated with moderate systemic inflammation detectable in maternal and cord blood, while also emphasizing that early postnatal factors may amplify or obscure antenatal inflammatory signals [15]. D’Alquen et al. demonstrated that chorioamnionitis with funiculitis was associated with activation of umbilical cord endothelial cells and increased inflammatory mediators in cord blood [16]. More specifically, prior studies evaluating PCT in the setting of histological chorioamnionitis have shown that cord and early neonatal PCT levels are higher in affected preterm deliveries, that cord blood PCT may help predict histological chorioamnionitis and neonatal outcomes, and that umbilical venous PCT may reflect the severity of chorioamnionitis [22–24]. Taken together, these studies support the biological plausibility of our findings. From a clinical perspective, our data do not establish a new treatment strategy triggered by high cord blood PCT or early postnatal CRP alone. Rather, they suggest that these clinically available biomarkers may improve early recognition of intrauterine inflammatory exposure and may help contextualize neonatal laboratory findings before placental histopathology becomes available. This may be relevant for risk stratification, closer monitoring, and interpretation of early neonatal inflammatory status, particularly in preterm infants at higher risk of EOS-related clinical courses and prolonged respiratory support. Any treatment decisions, however, must continue to rely on the full clinical context, microbiological evaluation, local neonatal sepsis protocols, and serial clinical and laboratory assessment. Although inflammatory placental pathology was associated with longer oxygen therapy and longer hospitalization, we did not observe statistically significant differences in the incidence of BPD or ROP. The apparent difference in overall ICH should also be interpreted cautiously because most hemorrhages were low grade. Grade I–II hemorrhages accounted for the majority of ICH cases, while only four infants in the entire cohort had grade III ICH and no grade IV ICH was recorded. Therefore, the observed numerical difference in overall ICH appears to be driven mainly by grade I–II hemorrhages, and its clinical significance remains uncertain. These results should be interpreted cautiously. First, the number of inflammatory cases was limited, which reduces statistical power for clinical outcome analyses. Second, the cohort included a relatively broad gestational age range, and infants with inflammatory placental pathology were of lower gestational age. Although gestational age was included in adjusted biomarker models, residual confounding by gestational maturity remains possible and may also have influenced neonatal morbidity comparisons. Third, some important neonatal outcomes, such as PVL, were not systematically available in the study database. Therefore, the morbidity analyses should be viewed as descriptive and exploratory rather than definitive. A strength of this study is its prospective cohort design and the use of umbilical cord blood collected immediately after birth, before substantial postnatal influences become established. Placental histopathology was based on consensus recommendations, and the pathologists were unaware of biomarker results, reducing the likelihood of differential classification bias [12, 13]. In addition, the study focused on clinically available biomarkers that may be more feasible for routine neonatal use than specialized cytokine, transcriptomic, or alarmin-based assays. The revised analysis also enabled lesion-specific exploration of MIR-only and FIR-present cases, the inclusion of neonatal infectious outcomes and morbidities, and a sensitivity analysis of the timing of postnatal CRP sampling, microbiological positivity, and the choice of placental reference group. Several limitations should be acknowledged. First, this was a single-center study with a modest sample size of 15 infants with inflammatory placental pathology; therefore, adjusted analyses should be regarded as exploratory and hypothesis-generating. Second, although lesion-specific subgrouping was possible, the number of FIR-present cases was small, and one additional inflammatory case represented chorioamnionitis with villitis, which was described separately rather than included in formal subgroup testing. Third, one inflammatory placental diagnosis was shared by a twin pair; however, sensitivity analysis excluding one twin did not materially change the main biomarker associations. Fourth, umbilical cord IL-6 was not measured, preventing direct comparison with an established biomarker of fetal inflammatory response. Fifth, clinical chorioamnionitis and PPROM were not systematically coded as distinct variables in the study database. Maternal intrapartum fever was identified in only one case, precluding meaningful analysis of that variable, and placental culture data were unavailable. Positive neonatal microbiological findings could not be used as a microbiological gold standard for placental infection. In addition, although positive microbiological findings were recorded, the database did not allow uniform retrospective adjudication of all isolates according to timing, sample type, and contamination status. Therefore, microbiological findings were retained as a descriptive variable and interpreted separately from EOS classification. Sixth, postnatal CRP sampling, although relatively narrowly distributed in practice, was not fully protocolized. Seventh, gestational age was obtained from obstetric records in a setting where dating is primarily based on the last menstrual period, with routine first-trimester ultrasound performed in primary gynecologic care; however, ultrasound-based dating parameters were not available separately for standardized extraction from the study database. Eighth, because gestational age and birth weight are closely related in preterm populations, residual collinearity may have influenced adjusted estimates. Finally, our study was not designed to establish clinically useful cutoff values for cord blood PCT associated with placental inflammatory pathology. Overall, our findings suggest that histopathological placental inflammation in preterm infants is associated with higher umbilical cord PCT and higher early postnatal CRP. Among the biomarkers evaluated, cord blood PCT showed the most consistent association with inflammatory placental pathology, including in exploratory analyses of fetal inflammatory response lesions. The association persisted in sensitivity analyses addressing microbiological positivity and the choice of placental reference group. These findings support the potential clinical relevance of clinically available inflammatory biomarkers in relation to placental inflammatory pathology, but larger multicenter studies with lesion-specific placental phenotyping, standardized postnatal sampling, and clinically meaningful neonatal outcomes are needed before their clinical utility can be more clearly defined. Conclusions In preterm infants, histopathological placental inflammation was associated with higher umbilical cord PCT and higher early postnatal CRP. Among the biomarkers evaluated, cord blood PCT showed the most consistent association with inflammatory placental pathology, including in exploratory analyses of fetal inflammatory response lesions. Inflammatory placental pathology was also associated with a higher burden of neonatal infectious outcomes, including culture-confirmed or clinically diagnosed EOS and positive microbiological findings. These findings support the potential clinical relevance of clinically available inflammatory biomarkers for early recognition of intrauterine inflammatory exposure, but larger multicenter studies with lesion-specific placental phenotyping, standardized postnatal sampling, and clinically meaningful neonatal outcomes are needed before their clinical utility can be more clearly defined. Supplementary Information Additional files 1: Supplementary Table S1. Timing of postnatal CRP sampling and sensitivity analysis restricted to samples obtained between 16 and 23 hours of life. Supplementary Table S2. ROC analysis of inflammatory biomarkers for placental inflammatory pathology. Supplementary Table S3. Sensitivity analysis excluding one twin from the pair sharing an inflammatory placental diagnosis. Supplementary Table S4. Sensitivity analysis excluding infants with positive microbiological findings. Supplementary Table S5. Adjusted sensitivity regression using no placental pathology as the reference group. Additional files 1: Supplementary Table S1. Timing of postnatal CRP sampling and sensitivity analysis restricted to samples obtained between 16 and 23 hours of life. Supplementary Table S2. ROC analysis of inflammatory biomarkers for placental inflammatory pathology. Supplementary Table S3. Sensitivity analysis excluding one twin from the pair sharing an inflammatory placental diagnosis. Supplementary Table S4. Sensitivity analysis excluding infants with positive microbiological findings. Supplementary Table S5. Adjusted sensitivity regression using no placental pathology as the reference group.