Intra-amniotic infection (IAI) is a complication of pregnancy that increases perinatal morbidity and may induce premature rupture of the membranes (PROM), preterm labor (PL), or puerperal infection. Early diagnosis of IAI allows the clinician to reduce other perinatal risks, such as fetal death, prematurity, or early-onset neonatal sepsis. However, early diagnosis is not easy because the clinical expression of IAI is frequently a late event or due to a subclinical course of IAI with premature uterine activity as the only manifestation. In order to establish an early identification of chorioamnionitis, several markers of infection have been tested in amniotic fluid (AF), such as gram stain, microbiologic cultures, glucose levels, leukocyte quantification and catalase activity, among others. Recently, interleukin-1 (IL-1), tumor necrosis factor- (TNF), interleukin-6 (IL-6) and IL-8 have been proposed as markers for chorioamnionitis and could also be related to neonatal sepsis. Among these, IL-6 has shown high sensitivity and specificity in the diagnosis of IAI.

IL-6 is a low molecular weight protein (26 kDa) that functions as an important mediator of immune and inflammatory responses. Interestingly, IL-6 may modulate the production of other pro-inflammatory cytokines, such as IL-1 and TNF at the level of transcription. The IL-6 present in AF seems to be produced by the decidua, with a possible fetal component. High concentrations of this cytokine have been demonstrated in AF from women with IAI, suggesting that augmentation of IL-6 in AF may be of great value for diagnosis and prognosis of PL induced by infection. In a recent study by Romero et al., IL-6 levels in AF were also related to a general increase in neonatal morbidity and mortality, whereas other authors and ourselves have reported the strong association of elevated IL-6 plasma levels in newborns with proven sepsis.

Early-onset neonatal sepsis is one of the most important complications in infants born to mothers with IAI. If IL-6 is a good marker for IAI, its increase in AF could also be associated with the development of early neonatal sepsis a useful indicator in predicting sepsis in preterm newborns due to the fact that early sepsis is a more frequent as in low gestational age and low birth weight infants.

The aim of this study was to determine the relationship between IL-6 concentrations in AF of mothers with preterm delivery without PROM and the development of early-onset neonatal sepsis in their infants.

A cohort study was carried out to examine the relationship between AF concentrations of IL-6 and risk for early-onset neonatal sepsis. This study consisted of women who were admitted to the National Institute of Perinatology Hospital in Mexico City between March 1, 1995 and October 30, 1996, with PL without PROM, and who met the following criteria: (a) preterm singleton pregnancy (26–36 weeks); (b) intact membranes at time of enrollment; and (c) written informed consent for the use of AF for research purposes. Women with complications of pregnancy such as preeclampsia, immune diseases previously diagnosed non-bacterial infectious diseases, such as rubella, toxoplasmosis or HIV infection, and women who were having antimicrobial treatment at the time of admission were excluded. The study project was approved by the research review board of the Institute.

Participants were at gestational ages of 28–36 weeks, determined by menstrual dating or ultrasonographic examination. Preterm labor was defined as regular uterine contractions at a frequency of <10 min and documented cervical dilatation. All women had intact membranes at the time of enrollment. Newborn anthropometric status was characterized by the neonatal staff based on standard charts of expected weight, length, and head circumference for Mexican preterm neonates, using a one-tenth percentile of birth weight for gestational age as a cut-off point. Newborn sepsis during the first 72 h was defined as early-onset sepsis. Diagnosis of sepsis was made by the attending neonatologist, who was unaware of the AF IL-6 values according to the criteria previously described.

The AF sample was immediately aliquoted for determination of IL-6 and microbiologic evaluation including cultures for aerobic and anaerobic bacteria and for genital mycoplasma, according to methods previously described. The AF samples were centrifuged at 10,000 × g for 5 min and stored at ?70°C in pyrogenic-free containers until assayed for IL-6, which was determined by commercial enzyme-linked immunoassays (Sanofi-Pasteur, Marnes, France) that had been validated for plasma, serum, and AF. Standard positive controls of IL-6 were run simultaneously with the study specimens. The lower limit of detection was 27 pg/mL. Amniotic fluid IL-6 levels above 1250 pg/mL were considered elevated; this value was the median for the entire distribution of IL-6 concentrations, and was chosen as the cut-off point. Similar values have been reported and used by others to determine associations between elevated cytokines in AF, infection and preterm delivery.

The associations between potential risk factors and the occurrence of early-onset neonatal sepsis were examined first by unadjusted odds ratio (OR) and confidence intervals. Either Fisher’s two-tailed test or ?² test was used for categoric data, and Student’s t test was applied to continuous variables. In order to control potentially confounding variables, a stratified analysis was performed by means of the Mantel-Haenszel chi-square procedure. Multivariable logistic regression models were constructed to assess the association between AF IL-6 levels and neonatal sepsis, considering adjustment for birth weight, gestational age, clinical chorioamnionitis, pathogenic bacteria in the AF, prenatal care at the Institute and use of prophylactic antibiotics during delivery.

Ninety-three women met the study criteria, and 31 (33%) of their newborns had early-onset neonatal sepsis. The study population was stratified according to their clinical characteristics and the presence or absence of neonatal sepsis. No significant differences were observed between groups for maternal age, positive AF cultures, mode of delivery, prenatal care, gestational age, and birth weight. Women whose infants developed neonatal sepsis had a higher frequency of clinical chorioamnionitis (p = 0.02) and also delivered their babies at earlier gestational ages (p <0.01).

Positive AF cultures were obtained from 18 women; Escherichia coli being the most common bacteria isolated from women whose infants had early sepsis, whereas Mycoplasma hominis and Ureaplasma urealyticum were commonly isolated from the women whose newborns did not have sepsis. Although clinical sepsis was proven in 31 neonates, only 10 had positive cultures. Escherichia coli and Staphylococcus coagulase negative were the most frequent isolations. The same microorganism was cultured from both neonatal blood and AF in seven cases: E. coli in four cases, Enterobacter cloacae, Streptococcus agalactie, and Citrobacter freundii in one case each, respectively.

Mothers of newborns who had early-onset sepsis showed significantly higher mean concentrations of IL-6 in their AF samples. The mean of AF IL-6 in the mothers of septic newborns was 5779 ± 2804 pg/mL, compared to 729 ± 382 pg/mL in mothers of non-infected neonates (p <0.001). Amniotic fluid IL-6 levels were higher than 1250 pg/mL in 26 of the 31 (83.8%) mothers whose newborns experienced early sepsis.