- Email: [email protected]
Amniotic fluid interleukin-6 and tumor necrosis factor-α at mid-trimester genetic amniocentesis: Relationship to intra-amniotic microbial invasion and preterm delivery
European Journal of Obstetrics & Gynecology and Reproductive Biology 148 (2010) 147–151
Contents lists available at ScienceDirect
European Journal of Obstetrics & Gynecology and Reproductive Biology journal homepage: www.elsevier.com/locate/ejogrb
Amniotic fluid interleukin-6 and tumor necrosis factor-a at mid-trimester genetic amniocentesis: Relationship to intra-amniotic microbial invasion and preterm delivery Nikolaos Thomakos a, George Daskalakis a,*, Angeliki Papapanagiotou b, Nikolaos Papantoniou a, Spyros Mesogitis a, Aris Antsaklis a a b
1st Department of Obstetrics and Gynecology, ‘‘Alexandra’’ Maternity Hospital, Athens University, 80 Vas. Sophias Av., Athens 115 28, Greece Department of Biological Chemistry, Athens University, 74 M.Asias Str., Athens, Greece
A R T I C L E I N F O
A B S T R A C T
Article history: Received 16 September 2008 Received in revised form 15 August 2009 Accepted 21 October 2009
Objective: To determine the value of amniotic fluid interleukin-6 (AF IL-6) and tumor necrosis factor-a (AF TNF-a) in the diagnosis of microbial invasion of the amniotic cavity and in the prediction of preterm delivery (PTD). Study design: Following amniocentesis, a sample of amniotic fluid was sent for aerobic and anaerobic bacterial cultures along with Ureaplasma urealyticum culture and it was also assessed for IL-6 and TNF-a. Results: Forty-eight women who delivered preterm (<37 weeks) were matched with 96 controls. The AF IL-6 and TNF-a concentrations of women with spontaneous PTD were significantly higher than those who delivered at term (IL-6: 176.3 pg/ml [130.6–208.6] vs. 52.3 pg/ml [37.2–92.3]; TNF-a: 8.8 pg/ml [7.2–10.7] vs. 5.5 pg/ml [5.0–6.3]). AF IL-6 and TNF-a concentrations of >99.3 pg/ml and of >6.6 pg/ml respectively, had a sensitivity of 89.6% and 81.3% and a specificity of 80.3% and 79.2% for the prediction of spontaneous PTD. Moreover, AF IL-6 and TNF-a concentrations of >99.3 pg/ml and of 6.3 pg/ml respectively, had a sensitivity of 91.9% and 78.4% and a specificity of 73.8% and 70.1% for the prediction of a positive AF culture. Conclusions: Elevated mid-trimester concentrations of AF IL-6, or/and of TNF-a can identify women at risk for intra-amniotic infection and for spontaneous PTD. ß 2009 Elsevier Ireland Ltd. All rights reserved.
Keywords: Interleukin-6 TNF-a Amniocentesis Inflammation Preterm delivery
1. Introduction Preterm delivery remains one of the major problems in modern obstetrics as it accounts for approximately 10% of all neonatal deaths and nearly one-half of the long term neurological disability [1]. Many studies have shown that there is a strong causal association between intrauterine infection and spontaneous preterm labor and delivery [2–4]. The frequency of intra-amniotic infection diagnosed by clinical signs and symptoms has been calculated between 6% and 17% in women delivering preterm and between 1% and 11% in women delivering at term [5]. However, in many cases the infection remains subclinical, therefore it is difficult for an accurate diagnosis to be made early. An occult infection of the normally sterile amniotic cavity may activate the cascade of inflammatory mediators which stimulates
* Corresponding author at: 5 Katsarou Street, 12351 St. Barbara, Athens, Greece. Tel.: +30 210 5618001/6945 235757; fax: +30 210 5317224. E-mail address: [email protected] (G. Daskalakis). 0301-2115/$ – see front matter ß 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejogrb.2009.10.027
prostaglandin synthesis and release, leading ultimately to uterine contractions and irreversible cervical changes [6,7]. The diagnosis of intra-amniotic infection becomes evident by the isolation of microorganisms in the amniotic fluid when this is obtained by amniocentesis. Unfortunately culture results take several days and therefore are not always available in time for immediate management decisions. The need to identify the microbial invasion of the amniotic cavity and thus manage pregnant women accordingly, has led to investigations where various cytokines have been used as markers for intrauterine infection, such as interleukin-6 (IL-6), IL-1b, IL-8, tumor necrosis factor (TNF-a) and granulocyte-colony stimulating factor. AF IL-6 is consistently related with intra-amniotic infection, histologic chorioamnionitis, and the presence of bacteria in chorioamnion [8]. TNF-a is not detected normally in the AF during the second and third trimester, but it is expressed under pathologic conditions such as intrauterine infection, or during physiologic parturition [9–11]. We know that a subclinical intrauterine inflammatory process may begin very early in pregnancy, in women who subsequently
N. Thomakos et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 148 (2010) 147–151
148
Table 1 Demographic and clinical characteristics of women according to the gestational age of delivery. Preterm delivery
Maternal age, mean SD Smoking, n (%) Previous preterm delivery, n (%) Birth weight, median (interquartile range) Gestational age at delivery, median (interquartile range) IL-6, median (interquartile range) (pg/ml) TNF, median (interquartile range) (pg/ml) a b
Yes (N = 48)
No (N = 96)
P (Mann–Whitney test)
38.4 2.8 13 (27.1) 4 (8.3) 2425 (2200–2625) 35.1 (33.7–35.9) 176.3 (130.6–208.6) 8.8 (7.2–10.7)
37.8 3.5 23 (24.0) 15 (15.6) 3300 (3100–3600) 38.5 (37.5–39.5) 52.3 (37.2–92.3) 5.5 (5.0–6.3)
0.252a 0.683b 0.223b <0.001 <0.001 <0.001 <0.001
Student’s t-test. Chi-square test.
deliver preterm. We hypothesized that AF IL-6 and TNF-a determination during the second trimester amniocentesis can potentially identify patients at risk for preterm delivery due to subclinical intra-amniotic infection. The aim of this study was to determine the value of AF IL-6 and TNF-a in the diagnosis of microbial invasion of the amniotic cavity and in the prediction of preterm delivery in asymptomatic women undergoing midtrimester genetic amniocentesis.
(ELISA) (Bender MedSystems, Vienna, Austria). The ELISA was validated for AF, and the samples were measured in duplicate. For AF IL-6 the sensitivity of the assay was 0.92 pg/ml, intra-assay and inter-assay coefficients of variation were 6.2% and 7% respectively. For TNF-a the sensitivity of the assay was determined to be 1.65 pg/ml, intra-assay and inter-assay coefficients were 6.0% and 9.3% respectively. 2.2. Statistical analysis
2. Materials and methods Patients who underwent mid-trimester genetic amniocentesis between February 2006 and September 2007 were asked to participate in the study. The study had received the approval of the ethics Committee of the Hospital. In all cases written informed consent was given prior to the procedure. All amniocentesis were performed between 16 and 19 weeks of gestation. Each amniocentesis was preceded by a detailed ultrasound scan with a 3.75 MHz curvilinear transducer, to assess fetal anatomy and to determine the location of the placenta. Following amniocentesis a sample of amniotic fluid was transported to the laboratory for aerobic and anaerobic bacterial cultures as well as for Ureaplasma urealyticum culture. Another sample was centrifuged at 200 g for 10 min and the supernatant was collected and stored at 70 8C in polypropylene tubes, within 1 h of the procedure. The gestational age was assessed either by the last menstrual period or by an early ultrasound scan if there was a discrepancy of more than a week. Inclusion criteria included singleton pregnancy, normal pregnancy course prior to the procedure, maternal age > 18 years, intact fetal membranes, no signs of preterm labor or cervical dilatation at the time of amniocentesis, no antibiotic treatment at/or shortly before amniocentesis. We excluded from the analysis cases with abnormal fetal karyotype, major fetal anomalies, and significant medical or obstetric complications such as preeclampsia, fetal growth restriction, gestational diabetes, poly- or oligohydramnios, placenta praevia leading to iatrogenic preterm delivery. Pregnancy losses within 4 weeks of the amniocentesis (thought to be possible procedure related) were also excluded. The results of the IL-6 and TNF-a assays and AF cultures were not available to the clinicians. Pregnancy outcomes were obtained by accessing labor and delivery records or by contacting the patient’s private physician if the delivery was not in our hospital. All women who delivered preterm (<37 weeks’ gestation) formed the study group. The control group consisted of the two subsequent women matched for maternal age, parity and indication for amniocentesis, who also underwent amniocentesis during the same time period and who delivered a neonate at term. 2.1. Amniotic fluid assay of IL-6 and TNF-a Levels of IL-6 and TNF-a in amniotic fluid were measured with commercially available enzyme-linked immunosorbent assays
Student’s t-tests were used and Mann–Whitney test otherwise. Chi-square and Fisher’s exact tests were used for the comparison of proportions. The diagnostic ability of IL-6 and TNF-a for the prediction of preterm deliveries and positive amniotic fluid cultures was evaluated using ROC curves (receiver operating characteristic curves). Sensitivity, specificity, negative and positive predictive values were calculated for optimal cut-offs. The area under the curve (AUC) was also calculated. Analyses were conducted using STATA statistical software (version 6.0). 3. Results During the study period, 685 women were enrolled in the study. Among them, 6 women had a spontaneous pregnancy loss within 4 weeks of the procedure and 20 were delivered <37 weeks for fetal or maternal indications. A chromosomal abnormality was found in 8 women, b-thalassemia major in 3 women and abnormal second trimester ultrasound in 3 women; all the above cases underwent pregnancy termination. Thirty-two women were lost to follow-up, therefore were excluded from analysis. The remaining 613 women met the inclusion criteria. The prevalence of spontaneous preterm delivery before 37 weeks was 7.8% (48/613). The preterm delivery group consisted, therefore, of 48 cases, which were matched with 96 controls who delivered at term. Table 1 presents the demographic and clinical characteristics of the two groups. Maternal age, smoking habits, the number of previous pregnancies, the gestational age at sampling and the number of previous preterm deliveries were not different between the two groups. As expected, birth weight and gestational age were significantly lower in the preterm delivery group compared to those who delivered at >37 weeks. The preterm delivery group had significantly higher concentrations of IL-6 and TNF-a. When ROC curves were constructed to evaluate the diagnostic performance of IL-6 (Fig. 1) and TNF-a (Fig. 2) in the detection of preterm delivery it was found that both had significant discriminant ability. The AUC for IL-6 was 0.91 (SE = 0.024, p < 0.001) and for TNF-a was 0.86 (SE = 0.033, p < 0.001). The optimal cut-off for the detection of preterm delivery was 99.3 pg/ml for IL-6 and 6.6 pg/ml for TNF-a. Sensitivities, specificities and other diagnostic indices for the aforementioned cut-offs are presented in Table 2. A comparison of the ROC curves demonstrated that the diagnostic performance of IL-6 and TNF-a was equivalent for the prediction of preterm
N. Thomakos et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 148 (2010) 147–151
149
Table 4 Diagnostic indices for interleukin-6 and TNF-a as predictors of a positive amniotic fluid culture.
Sensitivity (%) Specificity (%) Positive predictive value (%) Negative predictive value (%) RR (95% CI)a a
Fig. 1. ROC curve of IL-6 for the prediction of preterm delivery.
Fig. 2. ROC curve of TNF-a for the prediction of preterm delivery. Table 2 Diagnostic indices for interleukin-6 and TNF-a as predictors of delivery 37 weeks.
Sensitivity (%) Specificity (%) Positive predictive value (%) Negative predictive value (%) RR (95% CI)a a
IL-6 99.3
TNF-a 6.6
89.6 80.2 40.0 98.1 11.4 (4.8–27.0)
81.3 79.2 36.2 96.7 6.2 (3.3-11.9)
Relative risk (95% confidence interval).
delivery (p = 0.145). 89.6% of the preterm delivery group had an IL6 level over 99.3 pg/ml, compared to only 19.8% of the term delivery group. TNF-a level over 6.6 pg/ml was present in 81.3% of
IL-6 99.3
TNF-a 6.3
91.9 73.8 54.8 96.3 15.0 (4.8–45.5)
78.4 70.1 34.6 90.4 4.9 (2.4–10.0)
Relative risk (95% confidence interval).
the preterm delivery group and in 20.8% of the term delivery group. An inflammatory response, defined as IL-6 over 99.3 and/or TNF-a over 6.6 was present in 47 of the 48 preterm delivery cases (97%). The risk for preterm delivery was 11.4 times greater for subjects with IL-6 level above 99.3 pg/ml and 6.2 times greater for subjects with TNF-a level above 6.6 pg/ml. Positive amniotic fluid cultures were found in 37 cases (52.1% in preterm delivery group vs. 12.5% in term delivery group, p < 0.001). Ureaplasma urealyticum was present in 51.4% of positive amniotic fluid cultures, while Gardnerella vaginalis, Mycoplasma hominis and Escherichia coli were present in 29.7%, 24.3% and 18.9% of the positive amniotic fluid cultures respectively. The most common microorganism found in the cases of preterm delivery and positive amniotic fluid culture (n = 25) was Ureaplasma urealyticum (68%) following by G. vaginalis (36%) and M. hominis (32%). Table 3 describes the sample characteristics according to amniotic fluid results. Subjects with positive amniotic fluid cultures had significantly lower birth weight and gestational age. IL-6 and TNF-a concentrations were significantly higher in the cases with positive amniotic fluid cultures. Using ROC curve the optimal cut-off for the detection of a positive amniotic fluid culture was 99.3 for IL-6 with AUC equal to 0.85 (SE = 0.032, p < 0.001). The corresponding cut-off for TNF-a was 6.3 with AUC equal to 0.73 (SE = 0.049, p < 0.001). The diagnostic indices for the aforementioned cut-offs are presented in Table 4. IL-6 had a superior predictive ability compared to TNF-a for the detection of a positive amniotic fluid culture, as defined by the comparison of the AUCs (p = 0.005). Approximately 47.9% of preterm deliveries vs. 11.5% of term ones, had a positive culture and IL-6 99.3 (p < 0.001). Positive amniotic fluid culture with TNF-a 6.3 was present at 45.8% of preterm deliveries and 7.3% of term deliveries (p < 0.001). The proportions of preterm deliveries with positive amniotic fluid culture and IL-6 < 99.3; and positive amniotic fluid culture and TNF-a < 6.3 were 4.2% and 6.3% respectively. Furthermore, a negative amniotic fluid culture with IL-6 99.3 was present in 41.7% of preterm deliveries, while a negative amniotic fluid culture with TNF-a 6.3 was present in 35.4% of preterm deliveries.
Table 3 Demographic and clinical characteristics of women according to amniotic fluid culture results. Positive amniotic fluid culture
Maternal age, mean SD Smoking, n (%) Previous preterm delivery, n (%) Birth weight, median (interquartile range) Gestational age at delivery, median (interquartile range) IL-6, median (interquartile range) TNF-a, median (interquartile range) a b c
Student’s t-test. Chi-square test. Fisher’s exact test.
Yes (N = 37)
No (N = 107)
P (Mann–Whitney test)
38.4 2.3 11 (29.7) 4 (10.8) 2520 (2300–2690) 35.4 (32.8–35.9) 163.4 (126.8–206.6) 7.8 (6.7–9.8)
37.9 3.5 25 (23.4) 11 (10.3) 3250 (2900–3550) 38.7 (37.3–38.9) 62.3 (38.8–100.0) 5.6 (5.1–7.5)
0.504a 0.441b 1.000c <0.001 0.001 <0.001 <0.001
150
N. Thomakos et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 148 (2010) 147–151
4. Discussion The results of this study demonstrate that a proportion of preterm deliveries are the consequence of a chronic inflammatory process that begins early in gestation. It is obvious that this asymptomatic intra-amniotic inflammation which finally leads to preterm delivery might begin as early as, or even before, early second trimester. In the AF retrieved at a mean of 17.5 weeks’ gestation during amniocentesis for standard genetic indications, we found that IL-6 and TNF-a levels were significantly higher in women who subsequently delivered preterm compared to those who delivered at term. The Relative Risk for preterm delivery in the presence of elevated AF concentrations of these inflammatory markers obtained at the time of genetic amniocentesis was 11.4 for IL-6 and 6.2 for TNF-a. In our study population approximately 90% of the women who delivered preterm had IL-6 levels above the threshold which was determined by the ROC curve, while 80% of them had elevated TNF-a levels. This is another evidence that among the various etiologies of preterm labor, such as trauma, abruption, polyhydramnios, uterine anomalies, drug abuse, genetic/racial predisposition, or cervical insufficiency, intraamniotic inflammation is the main cause of spontaneous preterm delivery. The observation that women who delivered preterm had higher concentration of AF IL-6 and TNF-a, is consistent with similar findings reported by other investigators [12–14]. The majority of these studies have been conducted in women in preterm labor, or with preterm premature rupture of membranes. However, in recent studies the association between inflammatory cytokine response in AF obtained at the time of mid-trimester amniocentesis in asymptomatic women and the risk for subsequent preterm delivery has been investigated. The concentrations of MMP-8, IL-6, TNF-a, and angiogenin were significantly higher in patients who subsequently delivered preterm than in patients who delivered at term [12,14–17]. In contrast to previous data [9], where TNF-a was not detected in the AF of uncomplicated pregnancies but was expressed under either pathologic conditions or during the physiologic process of parturition, we found that approximately 20% of women who delivered at term had detectable immunoreactive TNF-a and IL-6. This finding may represent the laboratory evidence of another pathologic process, such as ischemic-hemorrhagic placental lesions, which also presents asymptomatically early in the second trimester of pregnancy and probably resolves spontaneously without ending in preterm delivery or any other clinical consequences [18,19]. Another significant observation of this study was that women who delivered preterm had a significantly higher rate of positive AF cultures than those who delivered at term (52.1% vs. 12.5%). Moreover, there was a strong association between increased AF IL6 and TNF-a levels and the microbial invasion of the amniotic cavity. However, 41.7% of the women who delivered preterm had an elevated AF IL-6 but negative AF cultures. Similarly, 35.4% of the women with preterm delivery had negative AF cultures but detectable levels of AF TNF-a. Several explanations for this finding may be considered. First, it can be suspected that in some women with inflammatory response and cytokine secretion the pathogens are located primarily in the decidua or the space between chorion and amnion (extra-amniotic infection) [9,20,21]. Elevated AF IL-6 concentration was found to correlate well with histologic evidence of chorioamnionitis and amniochorial cultures [7,22,23]. We should also take into account that intra-amniotic viral infection, which cannot be diagnosed by standard microbiologic techniques, might increase AF inflammatory markers [24]. An alternative explanation is that the bacteria responsible for the intrauterine infection which will become clinically evident
later, cannot be demonstrated and escape detection with the use of classic microbiologic techniques [9,25]. It seems that the use of PCR has a higher detection rate for intra-amniotic infections [26,27]. Also, potential antibacterial factors which are present in the AF may inhibit bacterial growth that could colonize the chorioamnion [28]. 4.2% of the women who delivered preterm had microbial invasion of the amniotic cavity and positive AF cultures, but did not show elevated AF IL-6 above the cut-off value selected in our study (IL-6 < 99.3 pg/ml). A similar finding was observed in 3 women with preterm delivery 6.3% and positive AF cultures but with TNF-a < 6.3 pg/ml. Low levels of AF IL-6 or TNF-a and positive AF cultures might indicate earlier stages of an intrauterine infection process where there is not enough time for cytokine response. Also, hyporesponder mothers may present with microbial invasion of the amniotic cavity and positive AF cultures but with low concentrations of AF cytokines. Functional polymorphism in several cytokines, including IL-6 and TNF-a, has also been reported [29]. Finally, the variable capacity of different microorganisms to stimulate the production of cytokines may also explain this finding. Indeed, bacterial endotoxins are not equally potent in stimulating the production of different pro-inflammatory markers [29]. The present study supports the hypothesis that a pathologic process such as intrauterine inflammation in the mid-trimester of pregnancy or even earlier, is a risk factor for preterm delivery. Our results demonstrated that measurement of AF IL-6 and TNFa can be used as markers for preterm delivery and as surrogate tests for the identification of intra-amniotic infection. A simple immunoassay for AF IL-6, TNF-a, or both, could be converted for use as a dipstick during mid-trimester genetic amniocentesis. Immediately after the procedure the sample would be examined for IL-6, TNF-a, or both, and if their concentrations are above the cut-off value, then we could proceed for AF cultures and possible antibiotic treatment. However, the inter-relationship between antibiotic use and preterm labor is complex and at the moment there is no evidence that antibiotics alone can be of any help. It is possible that the introduction of cytokines inhibitors in combination with antibiotics may delay delivery, in this group of patients.
References [1] Hack M, Fanaroff AA. Outcomes of children of extremely low birthweight and gestational age in the 1990s. Early Hum Dev 1999;53:193–218. [2] Guzick DS, Winn K. The association of chorioamnionitis with preterm delivery. Obstet Gynecol 1985;65:11–6. [3] Dodson MG, Fortunato SJ. Microorganisms and premature labor. J Reprod Med 1988;33:87–96. [4] Romero R, Oyarzun E, Mazor M, Sirtori M, Hobbins JC, Bracken M. Metaanalysis of the relationship between asymptomatic bacteriuria and preterm delivery/low birth weight. Obstet Gynecol 1989;73:576–82. [5] Yoder PR, Gibbs RS, Blanco JD, Castaneda YS, St Clair PJ. A prospective, controlled study of maternal and perinatal outcome after intra-amniotic infection at term. Am J Obstet Gynecol 1983;145:695–701. [6] Andrews WW, Goldenberg RL, Hauth JC. Preterm labor: emerging role of genital tract infections. Infect Agents Dis 1995;4:196–211. [7] Goldenberg RL, Hauth JC, Andrews WW. Intrauterine infection and preterm delivery. N Engl J Med 2000;342:1500–7. [8] El-Bastawissi AY, Williams MA, Riley DE, Hitti J, Krieger JN. Amniotic fluid interleukin-6 and preterm delivery: a review. Obstet Gynecol 2000;95: 1056–64. [9] Romero R, Mazor M, Sepulveda W, Avila C, Copeland D, Williams J. Tumor necrosis factor in preterm and term labor. Am J Obstet Gynecol 1992; 166:1576–87. [10] Fortunato SJ, Menon R, Lombardi SJ. Role of tumor necrosis factor-alpha in the premature rupture of membranes and preterm labor pathways. Am J Obstet Gynecol 2002;187:1159–62. [11] Romero R, Chaiworapongsa T, Kuivaniemi H, Tromp G. Bacterial vaginosis, the inflammatory response and the risk of preterm birth: a role for genetic epidemiology in the prevention of preterm birth. Am J Obstet Gynecol 2004;190:1509–19.
N. Thomakos et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 148 (2010) 147–151 [12] Spong CY, Ghindini A, Sherer DM, Pezzullo JC, Ossandon M, Eglinton GS. Angiogenin: a marker for preterm delivery in midtrimester amniotic fluid. Am J Obstet Gynecol 1997;176:415–8. [13] Romero R, Munoz H, Gomez R, et al. Two thirds of spontaneous abortion/fetal deaths after amniocentesis are the results of a pre-existing subclinical inflammatory process of the amniotic cavity. Am J Obstet Gynecol 1995;172:261 [abstract No. 24]. [14] Wenstrom KD, Andrews WW, Hauth JC, Goldenberg RL, DuBard MD, Cliver SP. Elevated second-trimester amniotic fluid interleukin-6 levels predict preterm delivery. Am J Obstet Gynecol 1998;178:546–50. [15] Yoon B, Oh S, Romero R, et al. An elevated amniotic fluid matrix metalloproteinase-8 level at the time of genetic amniocentesis is a risk factor for preterm delivery. Am J Obstet Gynecol 2001;185:1162–7. [16] Ghidini A, Eglinton GS, Spong CY, et al. Elevated midtrimester amniotic fluid tumor necrosis alpha levels: a predictor of preterm delivery. Am J Obstet Gynecol 1996;174:307 [abstract]. [17] Ghidini A, Jenkins CB, Spong CY, Pezzullo JC, Salafia CM, Eglinton GS. Elevated amniotic fluid interleukin-6 levels during the early second trimester are associated with greater risk of subsequent preterm delivery. Am J Reprod Immunol 1997;37:227–31. [18] Ramsey PS, Shinetugs B, Goldenberg RL, Cliver SP, Wenstrom KD. Midtrimester amniotic fluid collagen metabolism and the subsequent development of preterm premature rupture of membranes. J Soc Gynecol Investig 2002;9:85A [abstract]. [19] Arias F, Rodriguez L, Rayne SC, Kraus FT. Maternal placental vasculopathy and infection: two distinct subgroups among patients with preterm labor and preterm ruptured membranes. Am J Obstet Gynecol 1993;168(2):585–91. [20] Andrews WW, Hauth JC, Goldenberg RL, Gomez R, Romero R, Cassell GH. Amniotic fluid interleukin-6: correlation with upper genital tract microbial
[21]
[22] [23]
[24]
[25] [26]
[27]
[28] [29]
151
colonization and gestational age in women delivered following spontaneous labour versus indicated delivery. Am J Obstet Gynecol 1995;173(2):606–12. Cassell G, Andrews W, Hauth J, Cutter G, et al. Isolation of microorganisms from the chorioamnion is twice that from amniotic fluid at cesarean delivery in women intact membranes. Am J Obstet Gynecol 1993; 168:424 [abstract]. Steinborn A, Kuhnert M, Halberstadt E. Immunomodulating cytokines induce term and preterm parturition. J Perinat Med 1996;24:381–90. Saito S, Kasahara T, Kato Y, Ishihara Y, Ichijo M. Elevation of amniotic fluid interleukin 6 (IL-6), IL-8 and granulocyte colony stimulating factor (G-CSF) in term and preterm parturition. Cytokine 1993;5(1):81–8. Wenstrom KD, Andrews WW, Bowles NE, et al. Intrauterine viral infection at the time of second trimester genetic amniocentesis. Obstet Gynecol 1998; 92(September (3)):420–4. Romero R, Avila C, Anthanam U, Sehgal PB. Amniotic fluid interleukin-6 in preterm labor: association with infection. J Clin Invest 1990;85:1392–400. Hitti J, Riley DE, Krohn MA, et al. Broad-spectrum bacterial rDNA polymerase chain reaction assay for detecting amniotic fluid infection among women in premature labor. Clin Infect Dis 1997;24:1228–32. Markenson GR, Martin RK, TilLotson-Criss M, Foley KS, Stewart Jr RS, Yancey M. The use of the polymerase chain reaction to detect bacteria in amniotic fluid in pregnancies complicated by preterm labor. Am J Obstet Gynecol 1997; 177:1471–7. Schlievert P, Johnson W, Galask RP. Amniotic fluid antibacterial mechanisms: newer concepts. Semin Perinatol 1977;1:59–70. Yoon BH, Romero R, Moon BJ, et al. Clinical significance of intra-amniotic inflammation in patients with preterm labor and intact membranes. Am J Obstet Gynecol 2001;185:1130–6.
Contents lists available at ScienceDirect
European Journal of Obstetrics & Gynecology and Reproductive Biology journal homepage: www.elsevier.com/locate/ejogrb
Amniotic fluid interleukin-6 and tumor necrosis factor-a at mid-trimester genetic amniocentesis: Relationship to intra-amniotic microbial invasion and preterm delivery Nikolaos Thomakos a, George Daskalakis a,*, Angeliki Papapanagiotou b, Nikolaos Papantoniou a, Spyros Mesogitis a, Aris Antsaklis a a b
1st Department of Obstetrics and Gynecology, ‘‘Alexandra’’ Maternity Hospital, Athens University, 80 Vas. Sophias Av., Athens 115 28, Greece Department of Biological Chemistry, Athens University, 74 M.Asias Str., Athens, Greece
A R T I C L E I N F O
A B S T R A C T
Article history: Received 16 September 2008 Received in revised form 15 August 2009 Accepted 21 October 2009
Objective: To determine the value of amniotic fluid interleukin-6 (AF IL-6) and tumor necrosis factor-a (AF TNF-a) in the diagnosis of microbial invasion of the amniotic cavity and in the prediction of preterm delivery (PTD). Study design: Following amniocentesis, a sample of amniotic fluid was sent for aerobic and anaerobic bacterial cultures along with Ureaplasma urealyticum culture and it was also assessed for IL-6 and TNF-a. Results: Forty-eight women who delivered preterm (<37 weeks) were matched with 96 controls. The AF IL-6 and TNF-a concentrations of women with spontaneous PTD were significantly higher than those who delivered at term (IL-6: 176.3 pg/ml [130.6–208.6] vs. 52.3 pg/ml [37.2–92.3]; TNF-a: 8.8 pg/ml [7.2–10.7] vs. 5.5 pg/ml [5.0–6.3]). AF IL-6 and TNF-a concentrations of >99.3 pg/ml and of >6.6 pg/ml respectively, had a sensitivity of 89.6% and 81.3% and a specificity of 80.3% and 79.2% for the prediction of spontaneous PTD. Moreover, AF IL-6 and TNF-a concentrations of >99.3 pg/ml and of 6.3 pg/ml respectively, had a sensitivity of 91.9% and 78.4% and a specificity of 73.8% and 70.1% for the prediction of a positive AF culture. Conclusions: Elevated mid-trimester concentrations of AF IL-6, or/and of TNF-a can identify women at risk for intra-amniotic infection and for spontaneous PTD. ß 2009 Elsevier Ireland Ltd. All rights reserved.
Keywords: Interleukin-6 TNF-a Amniocentesis Inflammation Preterm delivery
1. Introduction Preterm delivery remains one of the major problems in modern obstetrics as it accounts for approximately 10% of all neonatal deaths and nearly one-half of the long term neurological disability [1]. Many studies have shown that there is a strong causal association between intrauterine infection and spontaneous preterm labor and delivery [2–4]. The frequency of intra-amniotic infection diagnosed by clinical signs and symptoms has been calculated between 6% and 17% in women delivering preterm and between 1% and 11% in women delivering at term [5]. However, in many cases the infection remains subclinical, therefore it is difficult for an accurate diagnosis to be made early. An occult infection of the normally sterile amniotic cavity may activate the cascade of inflammatory mediators which stimulates
* Corresponding author at: 5 Katsarou Street, 12351 St. Barbara, Athens, Greece. Tel.: +30 210 5618001/6945 235757; fax: +30 210 5317224. E-mail address: [email protected] (G. Daskalakis). 0301-2115/$ – see front matter ß 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ejogrb.2009.10.027
prostaglandin synthesis and release, leading ultimately to uterine contractions and irreversible cervical changes [6,7]. The diagnosis of intra-amniotic infection becomes evident by the isolation of microorganisms in the amniotic fluid when this is obtained by amniocentesis. Unfortunately culture results take several days and therefore are not always available in time for immediate management decisions. The need to identify the microbial invasion of the amniotic cavity and thus manage pregnant women accordingly, has led to investigations where various cytokines have been used as markers for intrauterine infection, such as interleukin-6 (IL-6), IL-1b, IL-8, tumor necrosis factor (TNF-a) and granulocyte-colony stimulating factor. AF IL-6 is consistently related with intra-amniotic infection, histologic chorioamnionitis, and the presence of bacteria in chorioamnion [8]. TNF-a is not detected normally in the AF during the second and third trimester, but it is expressed under pathologic conditions such as intrauterine infection, or during physiologic parturition [9–11]. We know that a subclinical intrauterine inflammatory process may begin very early in pregnancy, in women who subsequently
N. Thomakos et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 148 (2010) 147–151
148
Table 1 Demographic and clinical characteristics of women according to the gestational age of delivery. Preterm delivery
Maternal age, mean SD Smoking, n (%) Previous preterm delivery, n (%) Birth weight, median (interquartile range) Gestational age at delivery, median (interquartile range) IL-6, median (interquartile range) (pg/ml) TNF, median (interquartile range) (pg/ml) a b
Yes (N = 48)
No (N = 96)
P (Mann–Whitney test)
38.4 2.8 13 (27.1) 4 (8.3) 2425 (2200–2625) 35.1 (33.7–35.9) 176.3 (130.6–208.6) 8.8 (7.2–10.7)
37.8 3.5 23 (24.0) 15 (15.6) 3300 (3100–3600) 38.5 (37.5–39.5) 52.3 (37.2–92.3) 5.5 (5.0–6.3)
0.252a 0.683b 0.223b <0.001 <0.001 <0.001 <0.001
Student’s t-test. Chi-square test.
deliver preterm. We hypothesized that AF IL-6 and TNF-a determination during the second trimester amniocentesis can potentially identify patients at risk for preterm delivery due to subclinical intra-amniotic infection. The aim of this study was to determine the value of AF IL-6 and TNF-a in the diagnosis of microbial invasion of the amniotic cavity and in the prediction of preterm delivery in asymptomatic women undergoing midtrimester genetic amniocentesis.
(ELISA) (Bender MedSystems, Vienna, Austria). The ELISA was validated for AF, and the samples were measured in duplicate. For AF IL-6 the sensitivity of the assay was 0.92 pg/ml, intra-assay and inter-assay coefficients of variation were 6.2% and 7% respectively. For TNF-a the sensitivity of the assay was determined to be 1.65 pg/ml, intra-assay and inter-assay coefficients were 6.0% and 9.3% respectively. 2.2. Statistical analysis
2. Materials and methods Patients who underwent mid-trimester genetic amniocentesis between February 2006 and September 2007 were asked to participate in the study. The study had received the approval of the ethics Committee of the Hospital. In all cases written informed consent was given prior to the procedure. All amniocentesis were performed between 16 and 19 weeks of gestation. Each amniocentesis was preceded by a detailed ultrasound scan with a 3.75 MHz curvilinear transducer, to assess fetal anatomy and to determine the location of the placenta. Following amniocentesis a sample of amniotic fluid was transported to the laboratory for aerobic and anaerobic bacterial cultures as well as for Ureaplasma urealyticum culture. Another sample was centrifuged at 200 g for 10 min and the supernatant was collected and stored at 70 8C in polypropylene tubes, within 1 h of the procedure. The gestational age was assessed either by the last menstrual period or by an early ultrasound scan if there was a discrepancy of more than a week. Inclusion criteria included singleton pregnancy, normal pregnancy course prior to the procedure, maternal age > 18 years, intact fetal membranes, no signs of preterm labor or cervical dilatation at the time of amniocentesis, no antibiotic treatment at/or shortly before amniocentesis. We excluded from the analysis cases with abnormal fetal karyotype, major fetal anomalies, and significant medical or obstetric complications such as preeclampsia, fetal growth restriction, gestational diabetes, poly- or oligohydramnios, placenta praevia leading to iatrogenic preterm delivery. Pregnancy losses within 4 weeks of the amniocentesis (thought to be possible procedure related) were also excluded. The results of the IL-6 and TNF-a assays and AF cultures were not available to the clinicians. Pregnancy outcomes were obtained by accessing labor and delivery records or by contacting the patient’s private physician if the delivery was not in our hospital. All women who delivered preterm (<37 weeks’ gestation) formed the study group. The control group consisted of the two subsequent women matched for maternal age, parity and indication for amniocentesis, who also underwent amniocentesis during the same time period and who delivered a neonate at term. 2.1. Amniotic fluid assay of IL-6 and TNF-a Levels of IL-6 and TNF-a in amniotic fluid were measured with commercially available enzyme-linked immunosorbent assays
Student’s t-tests were used and Mann–Whitney test otherwise. Chi-square and Fisher’s exact tests were used for the comparison of proportions. The diagnostic ability of IL-6 and TNF-a for the prediction of preterm deliveries and positive amniotic fluid cultures was evaluated using ROC curves (receiver operating characteristic curves). Sensitivity, specificity, negative and positive predictive values were calculated for optimal cut-offs. The area under the curve (AUC) was also calculated. Analyses were conducted using STATA statistical software (version 6.0). 3. Results During the study period, 685 women were enrolled in the study. Among them, 6 women had a spontaneous pregnancy loss within 4 weeks of the procedure and 20 were delivered <37 weeks for fetal or maternal indications. A chromosomal abnormality was found in 8 women, b-thalassemia major in 3 women and abnormal second trimester ultrasound in 3 women; all the above cases underwent pregnancy termination. Thirty-two women were lost to follow-up, therefore were excluded from analysis. The remaining 613 women met the inclusion criteria. The prevalence of spontaneous preterm delivery before 37 weeks was 7.8% (48/613). The preterm delivery group consisted, therefore, of 48 cases, which were matched with 96 controls who delivered at term. Table 1 presents the demographic and clinical characteristics of the two groups. Maternal age, smoking habits, the number of previous pregnancies, the gestational age at sampling and the number of previous preterm deliveries were not different between the two groups. As expected, birth weight and gestational age were significantly lower in the preterm delivery group compared to those who delivered at >37 weeks. The preterm delivery group had significantly higher concentrations of IL-6 and TNF-a. When ROC curves were constructed to evaluate the diagnostic performance of IL-6 (Fig. 1) and TNF-a (Fig. 2) in the detection of preterm delivery it was found that both had significant discriminant ability. The AUC for IL-6 was 0.91 (SE = 0.024, p < 0.001) and for TNF-a was 0.86 (SE = 0.033, p < 0.001). The optimal cut-off for the detection of preterm delivery was 99.3 pg/ml for IL-6 and 6.6 pg/ml for TNF-a. Sensitivities, specificities and other diagnostic indices for the aforementioned cut-offs are presented in Table 2. A comparison of the ROC curves demonstrated that the diagnostic performance of IL-6 and TNF-a was equivalent for the prediction of preterm
N. Thomakos et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 148 (2010) 147–151
149
Table 4 Diagnostic indices for interleukin-6 and TNF-a as predictors of a positive amniotic fluid culture.
Sensitivity (%) Specificity (%) Positive predictive value (%) Negative predictive value (%) RR (95% CI)a a
Fig. 1. ROC curve of IL-6 for the prediction of preterm delivery.
Fig. 2. ROC curve of TNF-a for the prediction of preterm delivery. Table 2 Diagnostic indices for interleukin-6 and TNF-a as predictors of delivery 37 weeks.
Sensitivity (%) Specificity (%) Positive predictive value (%) Negative predictive value (%) RR (95% CI)a a
IL-6 99.3
TNF-a 6.6
89.6 80.2 40.0 98.1 11.4 (4.8–27.0)
81.3 79.2 36.2 96.7 6.2 (3.3-11.9)
Relative risk (95% confidence interval).
delivery (p = 0.145). 89.6% of the preterm delivery group had an IL6 level over 99.3 pg/ml, compared to only 19.8% of the term delivery group. TNF-a level over 6.6 pg/ml was present in 81.3% of
IL-6 99.3
TNF-a 6.3
91.9 73.8 54.8 96.3 15.0 (4.8–45.5)
78.4 70.1 34.6 90.4 4.9 (2.4–10.0)
Relative risk (95% confidence interval).
the preterm delivery group and in 20.8% of the term delivery group. An inflammatory response, defined as IL-6 over 99.3 and/or TNF-a over 6.6 was present in 47 of the 48 preterm delivery cases (97%). The risk for preterm delivery was 11.4 times greater for subjects with IL-6 level above 99.3 pg/ml and 6.2 times greater for subjects with TNF-a level above 6.6 pg/ml. Positive amniotic fluid cultures were found in 37 cases (52.1% in preterm delivery group vs. 12.5% in term delivery group, p < 0.001). Ureaplasma urealyticum was present in 51.4% of positive amniotic fluid cultures, while Gardnerella vaginalis, Mycoplasma hominis and Escherichia coli were present in 29.7%, 24.3% and 18.9% of the positive amniotic fluid cultures respectively. The most common microorganism found in the cases of preterm delivery and positive amniotic fluid culture (n = 25) was Ureaplasma urealyticum (68%) following by G. vaginalis (36%) and M. hominis (32%). Table 3 describes the sample characteristics according to amniotic fluid results. Subjects with positive amniotic fluid cultures had significantly lower birth weight and gestational age. IL-6 and TNF-a concentrations were significantly higher in the cases with positive amniotic fluid cultures. Using ROC curve the optimal cut-off for the detection of a positive amniotic fluid culture was 99.3 for IL-6 with AUC equal to 0.85 (SE = 0.032, p < 0.001). The corresponding cut-off for TNF-a was 6.3 with AUC equal to 0.73 (SE = 0.049, p < 0.001). The diagnostic indices for the aforementioned cut-offs are presented in Table 4. IL-6 had a superior predictive ability compared to TNF-a for the detection of a positive amniotic fluid culture, as defined by the comparison of the AUCs (p = 0.005). Approximately 47.9% of preterm deliveries vs. 11.5% of term ones, had a positive culture and IL-6 99.3 (p < 0.001). Positive amniotic fluid culture with TNF-a 6.3 was present at 45.8% of preterm deliveries and 7.3% of term deliveries (p < 0.001). The proportions of preterm deliveries with positive amniotic fluid culture and IL-6 < 99.3; and positive amniotic fluid culture and TNF-a < 6.3 were 4.2% and 6.3% respectively. Furthermore, a negative amniotic fluid culture with IL-6 99.3 was present in 41.7% of preterm deliveries, while a negative amniotic fluid culture with TNF-a 6.3 was present in 35.4% of preterm deliveries.
Table 3 Demographic and clinical characteristics of women according to amniotic fluid culture results. Positive amniotic fluid culture
Maternal age, mean SD Smoking, n (%) Previous preterm delivery, n (%) Birth weight, median (interquartile range) Gestational age at delivery, median (interquartile range) IL-6, median (interquartile range) TNF-a, median (interquartile range) a b c
Student’s t-test. Chi-square test. Fisher’s exact test.
Yes (N = 37)
No (N = 107)
P (Mann–Whitney test)
38.4 2.3 11 (29.7) 4 (10.8) 2520 (2300–2690) 35.4 (32.8–35.9) 163.4 (126.8–206.6) 7.8 (6.7–9.8)
37.9 3.5 25 (23.4) 11 (10.3) 3250 (2900–3550) 38.7 (37.3–38.9) 62.3 (38.8–100.0) 5.6 (5.1–7.5)
0.504a 0.441b 1.000c <0.001 0.001 <0.001 <0.001
150
N. Thomakos et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 148 (2010) 147–151
4. Discussion The results of this study demonstrate that a proportion of preterm deliveries are the consequence of a chronic inflammatory process that begins early in gestation. It is obvious that this asymptomatic intra-amniotic inflammation which finally leads to preterm delivery might begin as early as, or even before, early second trimester. In the AF retrieved at a mean of 17.5 weeks’ gestation during amniocentesis for standard genetic indications, we found that IL-6 and TNF-a levels were significantly higher in women who subsequently delivered preterm compared to those who delivered at term. The Relative Risk for preterm delivery in the presence of elevated AF concentrations of these inflammatory markers obtained at the time of genetic amniocentesis was 11.4 for IL-6 and 6.2 for TNF-a. In our study population approximately 90% of the women who delivered preterm had IL-6 levels above the threshold which was determined by the ROC curve, while 80% of them had elevated TNF-a levels. This is another evidence that among the various etiologies of preterm labor, such as trauma, abruption, polyhydramnios, uterine anomalies, drug abuse, genetic/racial predisposition, or cervical insufficiency, intraamniotic inflammation is the main cause of spontaneous preterm delivery. The observation that women who delivered preterm had higher concentration of AF IL-6 and TNF-a, is consistent with similar findings reported by other investigators [12–14]. The majority of these studies have been conducted in women in preterm labor, or with preterm premature rupture of membranes. However, in recent studies the association between inflammatory cytokine response in AF obtained at the time of mid-trimester amniocentesis in asymptomatic women and the risk for subsequent preterm delivery has been investigated. The concentrations of MMP-8, IL-6, TNF-a, and angiogenin were significantly higher in patients who subsequently delivered preterm than in patients who delivered at term [12,14–17]. In contrast to previous data [9], where TNF-a was not detected in the AF of uncomplicated pregnancies but was expressed under either pathologic conditions or during the physiologic process of parturition, we found that approximately 20% of women who delivered at term had detectable immunoreactive TNF-a and IL-6. This finding may represent the laboratory evidence of another pathologic process, such as ischemic-hemorrhagic placental lesions, which also presents asymptomatically early in the second trimester of pregnancy and probably resolves spontaneously without ending in preterm delivery or any other clinical consequences [18,19]. Another significant observation of this study was that women who delivered preterm had a significantly higher rate of positive AF cultures than those who delivered at term (52.1% vs. 12.5%). Moreover, there was a strong association between increased AF IL6 and TNF-a levels and the microbial invasion of the amniotic cavity. However, 41.7% of the women who delivered preterm had an elevated AF IL-6 but negative AF cultures. Similarly, 35.4% of the women with preterm delivery had negative AF cultures but detectable levels of AF TNF-a. Several explanations for this finding may be considered. First, it can be suspected that in some women with inflammatory response and cytokine secretion the pathogens are located primarily in the decidua or the space between chorion and amnion (extra-amniotic infection) [9,20,21]. Elevated AF IL-6 concentration was found to correlate well with histologic evidence of chorioamnionitis and amniochorial cultures [7,22,23]. We should also take into account that intra-amniotic viral infection, which cannot be diagnosed by standard microbiologic techniques, might increase AF inflammatory markers [24]. An alternative explanation is that the bacteria responsible for the intrauterine infection which will become clinically evident
later, cannot be demonstrated and escape detection with the use of classic microbiologic techniques [9,25]. It seems that the use of PCR has a higher detection rate for intra-amniotic infections [26,27]. Also, potential antibacterial factors which are present in the AF may inhibit bacterial growth that could colonize the chorioamnion [28]. 4.2% of the women who delivered preterm had microbial invasion of the amniotic cavity and positive AF cultures, but did not show elevated AF IL-6 above the cut-off value selected in our study (IL-6 < 99.3 pg/ml). A similar finding was observed in 3 women with preterm delivery 6.3% and positive AF cultures but with TNF-a < 6.3 pg/ml. Low levels of AF IL-6 or TNF-a and positive AF cultures might indicate earlier stages of an intrauterine infection process where there is not enough time for cytokine response. Also, hyporesponder mothers may present with microbial invasion of the amniotic cavity and positive AF cultures but with low concentrations of AF cytokines. Functional polymorphism in several cytokines, including IL-6 and TNF-a, has also been reported [29]. Finally, the variable capacity of different microorganisms to stimulate the production of cytokines may also explain this finding. Indeed, bacterial endotoxins are not equally potent in stimulating the production of different pro-inflammatory markers [29]. The present study supports the hypothesis that a pathologic process such as intrauterine inflammation in the mid-trimester of pregnancy or even earlier, is a risk factor for preterm delivery. Our results demonstrated that measurement of AF IL-6 and TNFa can be used as markers for preterm delivery and as surrogate tests for the identification of intra-amniotic infection. A simple immunoassay for AF IL-6, TNF-a, or both, could be converted for use as a dipstick during mid-trimester genetic amniocentesis. Immediately after the procedure the sample would be examined for IL-6, TNF-a, or both, and if their concentrations are above the cut-off value, then we could proceed for AF cultures and possible antibiotic treatment. However, the inter-relationship between antibiotic use and preterm labor is complex and at the moment there is no evidence that antibiotics alone can be of any help. It is possible that the introduction of cytokines inhibitors in combination with antibiotics may delay delivery, in this group of patients.
References [1] Hack M, Fanaroff AA. Outcomes of children of extremely low birthweight and gestational age in the 1990s. Early Hum Dev 1999;53:193–218. [2] Guzick DS, Winn K. The association of chorioamnionitis with preterm delivery. Obstet Gynecol 1985;65:11–6. [3] Dodson MG, Fortunato SJ. Microorganisms and premature labor. J Reprod Med 1988;33:87–96. [4] Romero R, Oyarzun E, Mazor M, Sirtori M, Hobbins JC, Bracken M. Metaanalysis of the relationship between asymptomatic bacteriuria and preterm delivery/low birth weight. Obstet Gynecol 1989;73:576–82. [5] Yoder PR, Gibbs RS, Blanco JD, Castaneda YS, St Clair PJ. A prospective, controlled study of maternal and perinatal outcome after intra-amniotic infection at term. Am J Obstet Gynecol 1983;145:695–701. [6] Andrews WW, Goldenberg RL, Hauth JC. Preterm labor: emerging role of genital tract infections. Infect Agents Dis 1995;4:196–211. [7] Goldenberg RL, Hauth JC, Andrews WW. Intrauterine infection and preterm delivery. N Engl J Med 2000;342:1500–7. [8] El-Bastawissi AY, Williams MA, Riley DE, Hitti J, Krieger JN. Amniotic fluid interleukin-6 and preterm delivery: a review. Obstet Gynecol 2000;95: 1056–64. [9] Romero R, Mazor M, Sepulveda W, Avila C, Copeland D, Williams J. Tumor necrosis factor in preterm and term labor. Am J Obstet Gynecol 1992; 166:1576–87. [10] Fortunato SJ, Menon R, Lombardi SJ. Role of tumor necrosis factor-alpha in the premature rupture of membranes and preterm labor pathways. Am J Obstet Gynecol 2002;187:1159–62. [11] Romero R, Chaiworapongsa T, Kuivaniemi H, Tromp G. Bacterial vaginosis, the inflammatory response and the risk of preterm birth: a role for genetic epidemiology in the prevention of preterm birth. Am J Obstet Gynecol 2004;190:1509–19.
N. Thomakos et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 148 (2010) 147–151 [12] Spong CY, Ghindini A, Sherer DM, Pezzullo JC, Ossandon M, Eglinton GS. Angiogenin: a marker for preterm delivery in midtrimester amniotic fluid. Am J Obstet Gynecol 1997;176:415–8. [13] Romero R, Munoz H, Gomez R, et al. Two thirds of spontaneous abortion/fetal deaths after amniocentesis are the results of a pre-existing subclinical inflammatory process of the amniotic cavity. Am J Obstet Gynecol 1995;172:261 [abstract No. 24]. [14] Wenstrom KD, Andrews WW, Hauth JC, Goldenberg RL, DuBard MD, Cliver SP. Elevated second-trimester amniotic fluid interleukin-6 levels predict preterm delivery. Am J Obstet Gynecol 1998;178:546–50. [15] Yoon B, Oh S, Romero R, et al. An elevated amniotic fluid matrix metalloproteinase-8 level at the time of genetic amniocentesis is a risk factor for preterm delivery. Am J Obstet Gynecol 2001;185:1162–7. [16] Ghidini A, Eglinton GS, Spong CY, et al. Elevated midtrimester amniotic fluid tumor necrosis alpha levels: a predictor of preterm delivery. Am J Obstet Gynecol 1996;174:307 [abstract]. [17] Ghidini A, Jenkins CB, Spong CY, Pezzullo JC, Salafia CM, Eglinton GS. Elevated amniotic fluid interleukin-6 levels during the early second trimester are associated with greater risk of subsequent preterm delivery. Am J Reprod Immunol 1997;37:227–31. [18] Ramsey PS, Shinetugs B, Goldenberg RL, Cliver SP, Wenstrom KD. Midtrimester amniotic fluid collagen metabolism and the subsequent development of preterm premature rupture of membranes. J Soc Gynecol Investig 2002;9:85A [abstract]. [19] Arias F, Rodriguez L, Rayne SC, Kraus FT. Maternal placental vasculopathy and infection: two distinct subgroups among patients with preterm labor and preterm ruptured membranes. Am J Obstet Gynecol 1993;168(2):585–91. [20] Andrews WW, Hauth JC, Goldenberg RL, Gomez R, Romero R, Cassell GH. Amniotic fluid interleukin-6: correlation with upper genital tract microbial
[21]
[22] [23]
[24]
[25] [26]
[27]
[28] [29]
151
colonization and gestational age in women delivered following spontaneous labour versus indicated delivery. Am J Obstet Gynecol 1995;173(2):606–12. Cassell G, Andrews W, Hauth J, Cutter G, et al. Isolation of microorganisms from the chorioamnion is twice that from amniotic fluid at cesarean delivery in women intact membranes. Am J Obstet Gynecol 1993; 168:424 [abstract]. Steinborn A, Kuhnert M, Halberstadt E. Immunomodulating cytokines induce term and preterm parturition. J Perinat Med 1996;24:381–90. Saito S, Kasahara T, Kato Y, Ishihara Y, Ichijo M. Elevation of amniotic fluid interleukin 6 (IL-6), IL-8 and granulocyte colony stimulating factor (G-CSF) in term and preterm parturition. Cytokine 1993;5(1):81–8. Wenstrom KD, Andrews WW, Bowles NE, et al. Intrauterine viral infection at the time of second trimester genetic amniocentesis. Obstet Gynecol 1998; 92(September (3)):420–4. Romero R, Avila C, Anthanam U, Sehgal PB. Amniotic fluid interleukin-6 in preterm labor: association with infection. J Clin Invest 1990;85:1392–400. Hitti J, Riley DE, Krohn MA, et al. Broad-spectrum bacterial rDNA polymerase chain reaction assay for detecting amniotic fluid infection among women in premature labor. Clin Infect Dis 1997;24:1228–32. Markenson GR, Martin RK, TilLotson-Criss M, Foley KS, Stewart Jr RS, Yancey M. The use of the polymerase chain reaction to detect bacteria in amniotic fluid in pregnancies complicated by preterm labor. Am J Obstet Gynecol 1997; 177:1471–7. Schlievert P, Johnson W, Galask RP. Amniotic fluid antibacterial mechanisms: newer concepts. Semin Perinatol 1977;1:59–70. Yoon BH, Romero R, Moon BJ, et al. Clinical significance of intra-amniotic inflammation in patients with preterm labor and intact membranes. Am J Obstet Gynecol 2001;185:1130–6.