*Correspondence: Dr B. Jacobsson, Perinatal Center, Department of Obstetrics and Gynaecology, Institute for the Health of Women and Children, Sahlgrenska University Hospital/East, 416 85 Göteborg, Sweden.
Objective To evaluate the relationship between interleukin (IL)-18 in cervical mucus and amniotic fluid and microbial invasion of amniotic fluid, preterm delivery and intra-amniotic inflammation in women in preterm labour, with preterm prelabour rupture of membranes and at term.
Design A prospective follow up study.
Setting Sahlgrenska University Hospital, Göteborg, Sweden.
Sample Women with singleton pregnancies (<34 weeks) presenting with preterm labour (n= 87) or preterm prelabour rupture of membranes (n= 47) and women, not in labour, at term (n= 28).
Methods Amniotic fluid was retrieved transabdominally. Cervical mucus was taken from the uterine cervix of women in preterm labour and at term. IL-18 was analysed with enzyme-linked immunosorbent assay.
Main outcome measures IL-18 in relation to microbial invasion of the amniotic fluid, delivery within seven days or <34 weeks of gestation and intra-amniotic inflammation.
Results The levels of IL-18 in cervical mucus and amniotic fluid were higher in women with preterm labour than in those not in labour at term. In the preterm labour group, significant associations were found between elevated IL-18 in amniotic fluid and microbial invasion of the amniotic fluid, as well as between delivery within seven days or <34 weeks of gestation and intra-amniotic inflammation. Delivery was delayed longer in the preterm prelabour rupture of membranes subgroup with IL-18 ≥1.0 ng/mL than in that with IL-18 <1.0 ng/mL.
Conclusions In the preterm labour group, high IL-18 in amniotic fluid (but not in the cervix) was associated with microbial invasion of the amniotic fluid, intra-amniotic inflammation and prompt delivery. On the other hand, elevated IL-18 in preterm prelabour rupture of the membranes group correlated with a longer interval to delivery.
Interleukin (IL)-18 is a recently identified cytokine1. It is considered to have pleiotrophic qualities that regulate both the innate and acquired immune responses and it can stimulate both a T helper 1 and a T helper 2 response depending on the local cytokine environment2,3. IL-18 is synthesised as a pro-form and is activated through cleavage by caspase-1 (IL-1β converting enzyme)4. IL-18 is important in the host defence against severe infections via induction of other cytokines and effector cells and molecules2. It is mainly synthesised by macrophages, monocytes and keratinocytes, but can also be produced by epithelial cells4–8. IL-18 enhances the inflammatory process by stimulating the production of interferon γ (INF-γ), tumour necrosis factor (TNF)-α and IL-1β2. IL-12 can act synergistically with IL-18 to provoke a T helper 1 response4,9. The IL-18-receptor in the cell membrane activates the same intracellular signal pathway as the IL-1-receptor, through activation of nuclear factor κβ2,10. Several human epithelial cell lines express pro-IL-18 and Chlamydia trachomatis infection has been shown to cause epithelial cells to produce mature IL-18 through caspase-1 activation5. IL-18 can also activate apoptosis by enhancing Fas ligand and Fas expression11.
IL-18 is present in the amniotic fluid as well as in maternal and fetal plasma 12–14. The levels of IL-18 appear to increase with advancing gestational age12. Menon et al.13 detected IL-18 mRNA in the chorion and in the interface between the decidua and the chorion, whereas the amniotic epithelium was devoid of mRNA and protein for IL-18. Higher levels of IL-18 were found in women with preterm prelabour rupture of membranes, no microbial invasion of the amniotic fluid or contractions, in comparison with women with preterm labour. On the other hand, Pacora et al.12 found that the concentration of IL-18 in the amniotic fluid increases with microbial invasion of the amniotic fluid in preterm labour and in women in labour at term. The elevated level of IL-18 in preterm prelabour rupture of membranes was suggested13 to initiate apoptosis in the fetal membranes through the Fas–Fas ligand pathway but this has not been confirmed. Thus, the role of IL-18 in preterm labour and preterm prelabour rupture of membranes remains uncertain, and no data are available on the levels of IL-18 in cervical mucus.
Considering its critical role in host defence, IL-18 may participate in providing a cervical/decidual barrier against microbial invasion of the amniotic fluid and might, conceivably, serve as a cervical marker for intra-amniotic infection. Therefore, our primary aim was to investigate the relationship between IL-18 in cervical mucus and amniotic fluid in preterm labour and in amniotic fluid in preterm prelabour rupture of membranes and microbial invasion of the amniotic fluid, preterm delivery (<34 weeks, ≤7 days) and intra-amniotic inflammation. Secondly, the intra-amniotic IL-18 level in preterm labour was compared with that in preterm prelabour rupture of membranes, as IL-18-induced apoptosis has been suggested to be important in the rupture of membranes13. Third, cervical mucus and amniotic fluid levels in women in preterm labour and in non-labouring women at term were compared.
The study population consisted of women with singleton pregnancies in preterm labour (n= 87) or with preterm prelabour rupture of membranes (n= 47) who presented at two delivery wards in Göteborg [Sahlgrenska Hospital (1996–1997) and Sahlgrenska University Hospital/East (1997–2001)] at a gestational age of less than 34 weeks. Preterm labour was defined as regular uterine contractions (at least two uterine contractions/10 minutes during ≥30 minutes) in combination with cervical changes at admittance to the clinic: 1. ≤2 cm length +≥1 cm dilatation or 2. ≤2 cm length + cervical softening or 3. ≥1 cm dilatation + cervical softening or 4. cervical length <30 mm at endovaginal ultrasound. Preterm prelabour rupture of membranes was defined as amniorrhexis (visible amniotic fluid in the vagina) before the onset of spontaneous labour. Contractions began prior to amniocentesis in seven women with preterm prelabour rupture of membranes. Women with known uterine abnormalities, fetal malformations, significant vaginal bleeding, imminent delivery or fetal distress were not included. Twenty-eight women at term (≥37 weeks) were included. These women were scheduled for an elective caesarean section with the following indications: psychosocial, breech presentation or two previous caesarean section. None of the women at term had contractions or rupture of membranes prior to surgery.
Gestational age was determined in all women except three by routine ultrasound in the second trimester (16th to 19th weeks of gestation). Tocolytic therapy (intravenous terbutaline and/or indomethacin, the latter if the pregnancy was <28 weeks of gestation) was administered according to the local protocol at the department.
Cervical mucus was obtained with a Cytobrush (Cytobrush Plus GT, Medscan Medical, Malmö, Sweden) from the external os of the cervix in all women in preterm labour (n= 87) and in all women at term (n= 28). The cervical mucus was weighed and kept in a refrigerator (+4°C) until processed within 5 hours. The Cytobrush with the cervical mucus was submerged in 1.0 mL NaCl, shaken for 30 minutes at +4°C, followed by centrifugation at 855 ×g at +4°C for 10 minutes and storage at −80°C until analysis.
Ultrasound-guided transabdominal amniocentesis, aspirating 30–50 mL of amniotic fluid, was performed in 59 women in preterm labour and in 47 women with preterm prelabour rupture of membranes under antiseptic conditions within 12 hours after admittance. In women at term, amniotic fluid was retrieved during caesarean section. A catheter was introduced into the amniotic cavity and 50 mL of amniotic fluid was aspirated prior to opening the membranes. After retrieval, the amniotic fluid was immediately placed in a refrigerator (+4°C) and was centrifuged within 5 hours of sampling at 855 ×g in +4°C for 10 minutes. The supernatant was stored at −80°C until analysis.
A sample of uncentrifuged amniotic fluid was immediately transported to the microbiological laboratory for polymerase chain reaction analysis of Ureaplasma urealyticum and Mycoplasma hominis and for aerobic and anaerobic culture. Microbial invasion was defined as positive polymerase chain reaction and/or growth of any bacteria in the amniotic fluid except Coagulase negative Staphylococcus that was considered to be skin contamination.
IL-18 in amniotic fluid was analysed with enzyme-linked immunosorbent assay (ELISA) (paired antibodies from R&D Systems, Minneapolis, Minnesota, USA). Amniotic fluid samples were diluted 1:5, 1:20 and 1:100 and run in duplicates. The inter-assay variation was calculated to <25%, based on analysis of several samples on three separate occasions. Low values (≤400 pg/mL) showed the highest variation coefficient. The detection limit of the ELISA test was 30 pg/mL, but because the samples were run at 1:5 dilution, the actual lower limit of detection was 150 pg/mL.
We have analysed IL-6 and IL-8 in amniotic fluid in the same population in previous studies 15,16 and calculated the diagnostic indices for delivery within seven days. A receiver–operator characteristic curve was used to identify the best cutoff levels for IL-6 and IL-8. Intra-amniotic inflammation was defined as IL-6 ≥1.5 ng/mL and/or IL-8 ≥1.3 ng/mL for women in preterm labour and as IL-6 ≥0.80 ng/mL and/or IL-8 ≥0.42 ng/mL for women with preterm prelabour rupture of membranes.
Two investigators (BJ, RMH) scrutinised the medical records and entered maternal and perinatal data into a database.
Ethical approval for the study was obtained from the local ethics committee in Göteborg. The women gave informed consent before enrolment in the study.
Calculations were made using the computer programmes StatView 5.01 (SAS Institute, Cary, North Carolina, USA) and InStat 2.01 (Graph Pad Software, San Diego, California, USA). Continuous variables were analysed with the Mann–Whitney U test and proportions with Fisher's exact test. A P value <0.05 was considered statistically significant, as was a confidence interval not including 1.00.
Demographics, presence of bacteria in the amniotic fluid, and pregnancy outcome are presented in Table 1. IL-18 was detectable in the cervix in 60% (52/87) of women with preterm labour and in 11% (3/28) of the women at term. IL-18 was detectable in the amniotic samples in 73% (43/59) of the women in preterm labour and in 70% (33/47) of women with preterm prelabour rupture of membranes. In the term group, IL-18 was detected in amniotic fluid in 11% (3/28) of the cases. Micro-organisms isolated from the amniotic cavity were U. urealyticum (n= 8), M. hominis (n= 1), Fusobacterium species (n= 1), Corynebacterium (no intra-amniotic inflammation) (n= 1), Eubacterium species (n= 1), Actinomyces odontolyticus (n= 1), Snethia sanguinegens (n= 1), Listeria monocytogenes (n= 1), Streptococcus mitis (no intra-amniotic inflammation) (n= 1), Haemophilus influenzae (n= 1), S. agalactiae (n= 1), Bacteroides fragilis (no intra-amniotic inflammation) (n= 1), Bifidobacterium adolecentis (no intra-amniotic inflammation) (n= 1), Difteriodic rods (n= 1), Anaerob gram-negative rods (n= 1) and coagulase negative Staphylococcus (no intra-amniotic inflammation, n= 3) (n= 5)15,16. None of these women developed clinical chorioamnionitis.
Table 1. Demographics and pregnancy outcome in women undergoing caesarean section at term, women in preterm labour and with preterm prelabour rupture of membranes. Data are presented as median [range] except for nulliparity, births <34 weeks and microbial invasion of the amniotic fluid for which the number and the percentage [n (%)] of the women in preterm and term population are given, respectively.
Preterm labour (CF)
P [term/preterm labour (CF)]
Preterm labour (AF)
Preterm prelabour rupture of membranes
P [preterm labour (AF)/preterm prelabour rupture of membranes]
AF = amniotic fluid; CF = cervical fluid; NA = not applicable.
No. of patients (n)
Maternal age (years)
Gestational age at sampling (days)
Gestational age at delivery (days)
Patients giving birth at <34 weeks (n)
Microbial invasion of the amniotic fluid (n)
There were higher levels of IL-18 in the cervical mucus in women in preterm labour compared with non-labouring women at term (median 0.51 ng/mL vs <0.15 ng/mL; P < 0.001) (Fig. 1a). In women with preterm labour, no significant associations were detected between cervical IL-18 and microbial invasion in the amniotic fluid, time of delivery (≤7 days, <34 weeks) or intra-amniotic inflammation (Table 2). A low correlation was found between IL-18 in cervical mucus and amniotic fluid (r= 0.30 and P= 0.028).
Table 2. IL-18 in cervical fluid in women in preterm labour related to different outcome variables.
MIAC = microbial invasion of the amniotic cavity; IAI = intra-amniotic inflammation.
*Cervical sample missing in five women.
MIAC (n= 7)/no MIAC (n= 47)*
Delivery ≤7 days (n= 26)/>7 days (n= 61)
Delivery <34 weeks (n= 37)/ ≥34 weeks (n= 50)
IAI (n= 25)/no IAI (n= 29)*
There were higher levels of IL-18 in the amniotic fluid of women in preterm labour compared with non-labouring women at term (median 0.59 vs <0.15 ng/mL; P= 0.004) (Fig. 1b).
Women in preterm labour with microbial invasion of the amniotic fluid had significantly higher levels of IL-18 in the amniotic fluid than those without microbial invasion (Fig. 2a). In women with preterm prelabour rupture of membranes, this association was not significant. In the term group, IL-18 was not detectable in the amniotic fluid from the woman with positive polymerase chain reaction for neither U. urealyticum nor the woman with coagulase negative Staphylococcus.
Women in preterm labour who delivered within seven days after amniocentesis had significantly higher levels of IL-18 than those who delivered after seven days (median 1.01 vs 0.57 ng/mL; P= 0.046) (Fig. 2b). In the preterm prelabour rupture of membranes group, the levels of IL-18 in women that delivered ≤7 days did not differ from those with an interval >7 days (median 0.24 vs median 0.38 ng/mL; P= 0.19). However, women with preterm prelabour rupture of membranes and low IL-18 in the amniotic fluid had shorter interval to delivery than those with high IL-18 levels (<1.0 ng/mL: median interval 4 days; ≥1.0 ng/mL: median latency 36 days; P= 0.03).
Women in preterm labour who delivered before 34 completed weeks had significantly higher levels of IL-18 than those who gave birth ≥34 weeks of gestation (median 1.0 vs 0.47 ng/mL; P= 0.02) (Fig. 2c). In women with preterm prelabour rupture of membranes, there was no association between amniotic fluid IL-18 and delivery <34 weeks (median 0.42 vs 0.23 ng/mL; P= 0.22).
Intra-amniotic IL-18 correlated to general intra-amniotic inflammation in women in preterm labour (median 0.81 vs 0.39 ng/mL; P= 0.02) (Fig. 2d), but no such relationship was found in the preterm prelabour rupture of membranes group (median 0.46 vs 0.28 ng/mL; P= 0.13).
No difference was found between IL-18 levels in the preterm labour and preterm prelabour rupture of membranes groups (median 0.59 vs 0.35 ng/mL; P= 0.068). Women with preterm prelabour rupture of membranes in the absence of both contractions and microbial invasion of the amniotic fluid (n= 31) did not have higher concentrations of IL-18 than those women with preterm prelabour rupture of membranes that had contractions and/or microbial amniotic fluid invasion (n= 16) (median 0.39 vs 0.35 ng/mL; P= 0.75).
This is the first study to document IL-18 in cervical mucus and its relationship to preterm labour. The major finding of this study is that high levels of amniotic fluid IL-18 correlated to intra-amniotic inflammation and prompt delivery in women presenting with preterm labour, whereas IL-18 levels ≥1.0 ng/mL were associated with a longer interval to delivery in the preterm prelabour rupture of membranes group. In agreement with a previously published study12, we found that IL-18 in amniotic fluid is associated with microbial invasion of the amniotic fluid and preterm delivery in preterm labour women12.
The level of IL-18 in cervical mucus and amniotic fluid was higher in women in preterm labour than in non-labouring women at term. A poor correlation was observed between the levels of IL-18 in cervical mucus and amniotic fluid. Pacora et al.12 have shown that the levels of IL-18 in amniotic fluid increase with gestational age. Our interpretation of the differences between women in preterm labour and non-labouring at term is that the higher levels of IL-18 are related to the general inflammatory response in preterm labour; this is reflected both in the cervical mucus and in the amniotic fluid.
There was no association between cervical IL-18 and microbial invasion of the amniotic fluid, intra-amniotic inflammation and delivery within seven days or <34 weeks, suggesting that this cytokine is unlikely to be of value as a cervical marker of preterm birth or infection in women in preterm labour.
IL-18 plays a well-documented role in the development of inflammatory disease (e.g. rheumatoid arthritis and inflammatory bowel disease)3,6. In this study, it is shown that intra-amniotic IL-18 correlates to the general intra-amniotic inflammation in women in preterm labour. The IL-18 level was also higher in women who delivered shortly after admission, indicating that IL-18 is part of the inflammatory response in preterm labour. This is interesting as IL-18 could also be involved in the fetal inflammatory response syndrome, which might have pathophysiological consequences, as this cytokine could exert adverse effects in the fetus. A recently published study shows that preterm infants who later developed periventricular leucomalacia or cerebral palsy had elevated concentrations of IL-18 in cord blood17. Experimental studies also show that IL-18 mediates liver cell injury in response to bacteria18 and, like IL-1β, induces injury in the central nervous system19.
IL-18 plays a role in the defence against both extra- and intracellular bacterial infection20. Initially, IL-18 was called INF-γ-inducing factor and INF-γ is known to play a central role in the immune response against infectious agents4,9. This is also supported by data from several animal studies21–23. IL-18 has the capacity to stimulate cytotoxic natural killer cells and stimulates T cells to produce IFN-γ and granulocyte/macrophage colony-stimulating factor8. Bacteria found in the amniotic fluid in women in preterm labour (e.g. Corynebacteria, Pseudomonas aeruginosa, H. influenzae and C. tracomatis) have been shown to provoke an IL-18 response in vivo in other compartments of the human body20. It is therefore interesting that we find, in agreement with Pacora et al.12, higher IL-18 levels in cases of preterm labour with positive cultures. It is a tentative hypothesis that IL-18 is part of the immune response to bacteria within the amniotic cavity of women in preterm labour.
Women in preterm labour who gave birth within seven days (or ≤34 weeks) had higher levels of IL-18 in the amniotic fluid than those who delivered after several days (or >34 weeks). Such a relationship between IL-18 and interval to delivery was not found by Pacora et al.12. This discrepancy could be due to the difference in gestational age (<34 vs <37 weeks, respectively) in our study compared with Pacora et al.'s study12, as inflammation might be more important at low gestational ages24. In women with preterm prelabour rupture of membranes, the opposite relationship was found (i.e. high IL-18 was associated with longer interval to delivery). It is also important to note that the amniotic concentrations of IL-18 in women with preterm prelabour rupture of membranes and microbial invasion of the amniotic fluid were not higher than in cases of preterm prelabour rupture of membranes with sterile amniotic fluid. The response is different for IL-6 and IL-8, levels of which were higher in women with microbial invasion both in the preterm labour and preterm prelabour rupture of membranes groups15,16. These data suggest that IL-18 plays different roles in preterm labour and preterm prelabour rupture of membranes. A similar dichotomy between preterm labour and preterm prelabour rupture of membranes has previously been observed regarding TNF-α and IL-1α12,25. The mechanisms involved are unclear, but IL-18, TNF-α and IL-1α all possess cytotoxic properties, in contrast to IL-6 and IL-8 that seem to be part of the general activation of the inflammatory response26.
Neither our study nor Pacora et al.'s12 could confirm the findings by Menon et al.13 of higher levels of IL-18 in preterm prelabour rupture of membranes. On the contrary, our data indicated a tendency towards lower values in the preterm prelabour rupture of the membranes group.
In conclusion, IL-18 is involved in intra-amniotic inflammation related to preterm labour and host defence. IL-18 in cervical mucus cannot serve as a marker of microbial invasion of the amniotic fluid or preterm delivery. Our data support the idea that IL-18 plays a different role in preterm prelabour rupture of membranes than in preterm labour.
The assistance of colleagues at the clinics recruiting the patients and the technical assistance of Ellen Samuelsson and Jolanta Bonislawska are very much appreciated. The study was supported by the Swedish Medical Research Council (09455), The Göteborg Medical Society, The Frimurare Barnhus Foundation and by Swedish government grants to researchers in public health service (ALF).
Dr Jacobsson recruited patients, scrutinised the records and was primarily responsible for analysis and writing the manuscript.
Dr Holst recruited patients, scrutinised the records and contributed to the writing of the manuscript.
Dr Inger Mattsby-Baltzer and Dr Nikolaitchouk were mainly responsible for laboratory analysis of the material.
Dr Wennerholm designed the study, recruited patients and contributed to the writing of the manuscript.
Dr Hagberg designed the study, recruited patients and supervised and contributed to the analysis of the study and writing of the manuscript.