Plasma inflammatory and immune proteins as predictors of intra-amniotic infection and spontaneous preterm delivery in women with preterm labor: a retrospective study

A retrospective cohort study was conducted on consecutive singleton pregnant women diagnosed with preterm labor (23 + 0 to 33 + 6 weeks of gestation) who were admitted to Seoul National University Bundang Hospital (Seongnamsi, Republic of Korea) from June 2004 to April 2015. The inclusion criteria were: (1) a live fetus; (2) an aliquot of maternal plasma available for analysis; (3) amniocentesis performed to evaluate the infectious and inflammatory status of AF at the time of enrollment; and (4) cervical length measured at the time of amniocentesis. The exclusion criteria were: (1) multifetal pregnancy; (2) preterm premature rupture of the membranes (pPROM); (3) prior or subsequent cerclage; (4) active labor at enrollment (defined as the presence of cervical dilatation greater than 3 cm); (5) evidence of clinical chorioamnionitis; and (6) major fetal congenital anomalies. Preterm labor was defined as the presence of regular uterine contractions, with a frequency of at least 2 contractions every 10 min, and cervical change (softening, effacement, or dilation) that required hospitalization. Gestational age was determined based on the last menstrual period and the first or second trimester (≤20 weeks) ultrasound results, when available. The ethics committee at Seoul National University Bundang Hospital approved the study (IRB no. B-1105/128–102). During the study period, amniocentesis for retrieval of AF and maternal blood sampling were immediately offered to patients who were admitted to our institution. Patients provided written informed consent for the collection and use of the blood and AF samples and for the use of their clinical information for research purposes. The primary and secondary outcome measures were positive AF cultures and SPTD within 48 h, respectively.

After obtaining informed consent, a transabdominal amniocentesis was performed under sonographic guidance with a 22-gauge spinal needle under aseptic conditions, and 15 to 20 mL of AF was aspirated. The AF was immediately sent to the laboratory for culture of aerobic/anaerobic bacteria and genital mycoplasmas (Ureaplasma urealyticum and Mycoplasma hominis) using previously described methods [20]. The remaining AF was centrifuged at 1500 g and 4 °C for 10 min, aliquoted, and stored at − 70 °C for further analysis. The IL-6 levels in the stored AF were measured using the enzyme-linked immunosorbent assay (ELISA) human DuoSet Kit (R&D Systems, Minneapolis, MN, USA). The range of the IL-6 standard curve was 7.8–600 pg/mL. The intra- and inter-assay coefficients of variation were < 10% each. CRP and cervical length by transvaginal ultrasound were measured in participants at the time of amniocentesis as part of the hospital protocol using methods that have been previously described [20, 21]. The results of AF culture, CRP, and cervical length measurement were available to the managing clinicians, but the AF IL-6 results were not.

Intra-amniotic infection was defined as the presence of a positive AF culture for microorganisms. The patients were hydrated and, if uterine contractions persisted, they were started on a regimen of intravenous tocolysis with magnesium sulfate, ritodrine, or atosiban. Corticosteroids were administered between 24 and 34 weeks of gestation. The decision to use prophylactic antibiotics was left to the discretion of the attending obstetrician. At our institution, antibiotic treatment of women with preterm labor for pregnancy prolongation is not recommended except for group B streptococcus prophylaxis or treatment of cases with clinically suspected (AF WBC count ≥ 50 cells/mm³) or diagnosed intra-amniotic infection, as previously described in detail [20]. Although the regimen for antibiotic prophylaxis in these patients varied according to the study period and the obstetrician, ampicillin and azithromycin (clarithromycin or erythromycin) were the main antibiotics used. For women with intra-amniotic infection, antibiotics are selected on the basis of the type of microorganisms found in the AF cultures. Medications such as antibiotics, corticosteroids, and tocolytics were started after sampling. A histologic diagnosis of chorioamnionitis was made in accordance with the definition previously described in detail [22]. Clinical chorioamnionitis was diagnosed in accordance with the criteria proposed by Gibbs et al. [23].

Data were analyzed using SPSS version 22.0 for Windows (IBM Corp., Armonk, NY, USA). The Shapiro–Wilk test was used to determine whether the data in groups were distributed normally. Univariate analysis was performed using the Student’s t-test or Mann–Whitney U test for continuous data and the χ²-test or Fisher’s exact test for dichotomous data. A multivariate logistic regression model was used to examine the relationship of the level of each protein in plasma to the outcome after adjusting for baseline variables, with a P value < 0.1 in univariate analysis. Independent variables associated with invasive amniocentesis were not included in the logistic regression model, because these were not baseline variables but were used for comparison. In the logistic regression model, continuous data were transformed into dichotomous data to decrease the problem of multicollinearity, especially between the serum CRP and plasma IL-6 (r = 0.495) or for prediction or decision-making purposes. Receiver-operating characteristic (ROC) curves were used to identify the best cut-off values for dichotomization. The optimal cut-off values were selected on the basis of the Youden index (maximum [sensitivity + specificity] − 1). The areas under the ROC curves (AUCs) for each protein and the cervical length were measured, and compared using the method described by DeLong et al. [24]. The sensitivity, specificity, and predictive values of plasma IL-6 and cervical length with reference to AF IL-6 and serum CRP levels were calculated and compared using the McNemar test. The sampling-to-delivery interval was assessed using Kaplan–Meier analysis and was compared between groups using the log-rank test. The Cox proportional model was used to determine whether higher levels of various proteins in plasma were associated with the sampling-to-delivery interval after adjusting for other prognostic variables. The Spearman rank correlation test was used to measure the relationship between the continuous variables that did not follow a normal distribution. All P values were two-sided, and P < .05 was considered statistically significant.