OBJECTIVE: To evaluate amniotic fluid arachidonic acid metabolites using enzymatic and nonenzymatic (lipid peroxidation) pathways in spontaneous preterm birth and term births, and to estimate whether prostanoid concentrations correlate with risk factors (race, cigarette smoking, and microbial invasion of amniotic cavity) associated with preterm birth.
METHODS: In a case-control study, amniotic fluid was collected at the time of labor or during cesarean delivery. Amniotic fluid samples were subjected to gas chromatography, negative ion chemical ionization, and mass spectrometry for prostaglandin (PG) E2, PGF2α, and PGD2 and for 6-keto-PGF1α (thromboxane 2 and F2-isoprostane). Primary analysis examined differences between prostanoid concentrations in preterm birth (n=133) compared with term births (n=189). Secondary stratified analyses (by race, cigarette smoking, and microbial invasion of amniotic cavity) compared eicosanoid concentrations in three epidemiological risk factors.
RESULTS: Amniotic fluid F2-isoprostane, PGE2, and PGD2 were significantly higher at term than in preterm birth, whereas PGF2α was higher in preterm birth 6-keto-PGF1α and thromboxane 2 concentrations were not different. Data stratified by race (African American or white) showed no significant disparity among prostanoid concentrations. Regardless of gestational age status, F2-isoprostane was threefold higher in smokers, and other eicosanoids were also higher in smokers compared with nonsmokers. Preterm birth with microbial invasion of amniotic cavity had significantly higher F2-isoprostane compared with preterm birth without microbial invasion of amniotic cavity.
CONCLUSION: Most amniotic fluid eicosanoid concentrations (F2-isoprostane, PGE2, and PGD2), are higher at term than in preterm births. The only amniotic fluid eicosanoid that is not higher at term is PGF2α.
LEVEL OF EVIDENCE: III
Amniotic fluid eicosanoids vary in preterm and term births, and oxidative stress is associated with cigarette smoking and term parturition.
From the Departments of Epidemiology and Biostatistics, Rollins School of Public Health, and the Department of Gynecology and Obstetrics, Emory University, Atlanta, Georgia; The Perinatal Research Center, Nashville, Tennessee; the Division of Clinical Pharmacology, Eicosanoid Core Laboratory, Vanderbilt University, Nashville, Tennessee; and the Department of Obstetrics and Gynaecology, University of Melbourne, Victoria, Australia.
Supported in part by PHS grant UL1 RR025008 from the Clinical and Translational Science Award program, National Institutes of Health, National Center for Research Resources, and University Research Committee award (2009163) to Ramkumar Menon.
Corresponding author: Ramkumar Menon, PhD, Departments of Epidemiology and Gynecology and Obstetrics, Emory University, Room 4053, 1518 Clifton Road NE, Atlanta, GA 30322; e-mail: email@example.com.
Financial Disclosure The authors did not report any potential conflicts of interest.
Despite remarkable advances in general medical care, the preterm birth rate has increased in the United States, increasing by as much as 30% during the past 25 years.1–3 Racial disparity in preterm birth, a long-recognized risk factor, also appears to be widening, further complicating our understanding of its pathophysiologic manifestations.1,2,4 The current management of spontaneous preterm labor resulting in preterm birth is based on a universal strategy to inhibit uterine contractions in a large subset of women (especially those at less than 34 weeks of gestation) without proper identification of risk factors, understanding of specific initiators and effectors, or knowledge of responsible pathophysiologic pathways in individuals who typically have multiple exposures.4
Preterm birth dogma invokes a linear mechanistic pathway, with several putative etiologic factors (eg, infection, cigarette smoking, preterm premature rupture of membranes, coagulation disorders, uterine distension, and behavioral or psychosocial stress) converging on an inflammatory reaction mediated by cytokines and matrix metalloproteinases, ultimately precipitating prostaglandin [PG] generation, from cell membrane-derived arachidonic acid metabolism. PGF2α and PGE2 are the predominant and most studied PGs associated with preterm birth because they can cause cervical ripening and dilatation, membrane weakening, and membrane rupture, and they can stimulate myometrial contractility, resulting in labor.5–13 Because of their uterotonic activities, PGs have been postulated as the final effector molecules of labor, although the initiating signals of these events may differ in preterm and term deliveries. Inflammation has been documented as a manifestation of both preterm and term labor, but targeting PG synthesis as a first-line intervention has not been successful in reducing the rate of preterm birth. The limited success of PG inhibitors to prevent preterm birth may reflect a failure to precisely identify the underlying causal or effector pathways involved. It is possible that an alternate pathways of labor exists, possibly mediated by less well-characterized eicosanoids.
Oxidative stress has been suggested as a mediator of labor initiation in response to certain risk exposures.14,15 Oxidative stress in pregnancy arises when the production of reactive oxygen species exceeds the capacity of antioxidant stores. Cigarette smoking, nutrient deficiencies, high energy demands of the fetoplacental unit, intrinsic microvascular disease, anaerobic infections, and placental apoptosis can result in an imbalanced redox state. Although inflammation and oxidative stress may share some risk factors and mechanistic pathways, we postulate that different mediators are involved. Nonenzymatic oxidation of arachidonic acid can result in the production of isoprostanes, whereas cytokine activation secondary to inflammation yields enzymatically produced PGs (PGF2α, PGE2, and PGD2), thromboxane B2 (TXB2), and the prostacyclin metabolite 6-keto-PGF1 α.16–18 The best-studied isoprostane form, 8-epiprostane or F2-isoprostane, contains F-type prostane rings isomeric to PGF2α.16–18 Biological functions of isoprostanes are tissue-, cell-, and concentration-dependent16–19 and include smooth muscle constriction, macrophage activation, vascular cell proliferation, and induction of endothelin-1 release.16–19 The latter is a potent uterotonin known to be produced by fetal membranes.20–22
The roles of eicosanoids besides PGF2α and PGE2 have not been well-characterized or proposed as alternate mediators of preterm birth or term birth. In fact, even the role of PGE2 has been questioned in a recent report showing that amnion production of PGE2 was not associated with preterm birth.23 To obtain eicosanoid signatures in preterm and term deliveries, we measured F2-isoprostane, PGF2α, PGE2, PGD2, 6-keto-PGF1α, and TXB2 in amniotic fluid from preterm birth (case group) and normal term births (control group) using a specific and sensitive method using gas chromatography (GC), negative ion chemical ionization, and mass spectrometry. Secondary analyses were performed on stratified data to document the effect of three risk factors of preterm birth: race, cigarette smoking, and intra-amniotic infection or microbial invasion of the amniotic cavity (microbial invasion of the amniotic cavity), which are reported to modify pregnancy outcome (preterm birth compared with normal term birth). Our findings indicate that F2-isoprostane represents a biomarker for an alternate pathway mediating a subgroup of preterm birth cases.
MATERIALS AND METHODS
This study was conducted at Centennial Women's Hospital, Nashville, Tennessee, and Emory University, Atlanta, Georgia, and was approved by the TriStar Nashville Institutional Review Board (Centennial Medical Center), Western Institutional Review Board (Seattle, WA), and the Emory University Institutional Review Board (Atlanta, GA). All individuals were recruited at Centennial Women's Hospital between September 2003 and December 2008, assays were performed at Vanderbilt University Eicosanoid Core Laboratory, and study design, selection of parameters, and data analysis were conducted at Emory University.
In this cross-sectional study of a retrospective cohort, pregnant women between the ages of 18 and 40 years presenting to Centennial Women's Hospital, Nashville, Tennessee, for delivery were eligible to consent for this study. Individuals were enrolled after obtaining written consent. Amniotic fluids were collected and eicosanoids were measured on samples from each group based solely on the order of recruitment into the study.23–27 All individuals had spontaneous labor (defined as the presence of regular uterine contractions at a minimum frequency of two contractions every 10 minutes) that led to delivery. Race was identified by self-report and determined by the race of the mother and father of the fetus, their parents, and grandparents. If individuals reported any family members different from the rest, then they were excluded (eg, African American when the rest were white or vice versa).24–28 Gestational age was determined by last menstrual period dating, corroborated by ultrasonography. Patients with spontaneous onset of labor who delivered preterm (between 24 weeks and 36 weeks 6 days of gestation age) were considered as being in the case group. Control group individuals were chosen based on a normal pregnancy ending in term labor and delivery (37 weeks of gestation or more) who had intact membranes and no pregnancy-related complications or history of pregnancy complications, including preterm labor and preterm premature rupture of membranes. These criteria were used to exclude overlap between case group and control group individuals. Individuals with multiple gestations, preeclampsia, placenta previa, preterm prelabor rupture of the membranes, fetal anomalies, gestational diabetes mellitus, or other medical or surgical complications of pregnancy were excluded. Individuals who were treated for preterm labor or for suspected intra-amniotic infection and delivered at term were excluded from the control group; however, those who were treated and delivered preterm were included in the case group.
Demographic data were collected from interviews and clinical data were abstracted from individuals' medical records. Age, socioeconomic data (education, yearly income, insurance status, and marital status), behavioral status (smoking during pregnancy), body mass index, and a complete medical and obstetric history were collected. For vaginal deliveries, amniotic fluid samples were collected during labor (either preterm or term) immediately before artificial rupture of the membranes by transvaginal amniocentesis of intact membranes using a 22-gauge needle through the dilated cervical os. Samples were also collected by placement of an intrauterine pressure catheter in a few cases when indicated for clinical reasons. This procedure avoided contamination of amniotic fluid from vaginal and cervical fluids as verified by analysis for microbial invasion of the amniotic cavity. In case group individuals undergoing cesarean delivery, samples were collected by transabdominal amniocentesis. Amniotic fluid was centrifuged immediately for 10 minutes at 2,000 g to remove cellular and particulate matter and supernatant aliquots were processed rapidly and stored in the dark at −80°C in filled tubes to minimize artifactual oxidation until analysis. Because no preexisting data were available to estimate power to detect differences between races, we studied the first 150 samples from African American individuals and 177 samples from white individuals that were collected. In post hoc analysis, we determined that 133 women in the case group and 169 women in the control group provided more than 80% power to detect the differences in analyte concentrations between the groups at a 0.05 significance level for each analyte in their log-transformed form.
Microscopic examination of the placenta and umbilical cord was performed in all women in the case group. Histologic evidence of chorioamnionitis was defined as a dense polymorphonuclear leukocyte and neutrophil infiltration of the amniochorionic membrane (excluding decidua), and funisitis was defined as inflammation in one or more of the umbilical cord vessels with or without inflammation in the Wharton jelly. Clinical chorioamnionitis (modified from Duff et al30) was defined as any three of the following: increased white blood cell count more than 15,000/dL; increased C-reactive protein levels (0.8 units/mL or more); and fever (102.0°F or higher), abdominal pain, or uterine tenderness and foul smelling vaginal discharge as verified from case group records. Bacterial vaginosis was diagnosed using Nugent's criteria on Gram-stained vaginal swabs, although its specificity has been challenged. Microbial invasion of the amniotic cavity (microbial invasion of the amniotic cavity) was defined by the presence of bacteria in the amniotic fluid detected by amplification of microbial 16s ribosomal DNA by polymerase chain reaction (TaqMan Assay).31,32 Microbial invasion of the amniotic cavity was determined by either polymerase chain reaction or by microbial cultures in all samples from the case group and in a subgroup of (n=50) randomly selected control group samples collected by transvaginal amniocentesis (from both races) to rule out contamination of amniotic fluid during specimen collection.
All samples included in this study were assayed simultaneously to avoid any variability introduced during assay procedures. amniotic fluid samples (0.20 mL) were diluted to a volume of 10 mL with 0.01 N HCl and 1.0 ng of each of the following internal standards were added to the solution; [2H4]-15-F2-isoprostane ([2H4]-8-iso-PGF2α), [2H4]-PGD2, [2H4]-PGE2, [2H3]-11-dehydro-TXB2, and [2H4]-6KPGF1α (all purchased from Cayman Chemicals). The details of this method can be seen in our previous publications.33
GC, negative ion chemical ionization, and mass spectrometry were performed on an Agilent 5973 Inert Mass Selective Detector interfaced with a computerized Agilent 6890n Network GC system. The GC phase was performed using a 15-m 0.25-mm film thickness, DB-1701-fused, silica capillary column (J and W Scientific). The column temperature was programmed from 190°C to 300°C at 20°C increments per minute. Levels of endogenous eicosanoids in the biological samples were calculated from the ratio of intensities of the [2H0] ions and [2H4] ions. GC, negative ion chemical ionization, and mass spectrometry allowed us to measure six different analytes in the same amniotic fluid sample in a single run and provided a more sensitive and accurate measurement of these analytes than enzyme-linked immunosorbent assay.
The summary statistics for amniotic fluid F2-isoprostane and PG concentrations between the case and control groups stratified by race are displayed in Table 1. [F2]-isoprostane and PGF2α were measurable in 99% (133 of 134) of women in the case group and 98% (189 of 193) of women in the control group. PGE2 was measurable in 89% (119 of 134) of women in the case group and 98% (189 of 193) of women in the control group. 6-keto-PGF1α was measured in 69% (92 of 134) of women in the case group and 89% (171 of 194) of women in the control group. TXB2 was detected in 85% (114 of 134) of women in the case group and 93% of women in the control group. PGD2 was obtained only in 63% (84 of 134) women in the case group and 86% (166 of 194) of women in the control group. The limit of detection for all analytes was 2 pg/mL and (nonmeasurable) and indicates that no peaks were observed in some samples for some of the analytes.
Data are presented as mean±standard deviation. For continuous and categorical variables, respectively, Student t tests and χ2 tests (Fisher exact test for bacterial vaginosis) were performed to determine the statistical significance of any differences in the distribution of baseline characteristics between preterm births in the case and control groups, overall and stratified by race. Normality was tested using Kolmogorov-Smirnov test before performing statistical analysis. Because the analyte concentrations were not normally distributed, log-transformation was performed to justify the use of Student t test to compare concentrations among the case and control groups and also between races. Two-tailed P=.05 was considered the threshold for statistical significance. Primary analysis of differences in prostanoid concentrations between the case and control groups documented their distributions during preterm and term births. Secondary analyses were utilized to compare differences in analytes between races, between smokers compared with nonsmokers, and in cases complicated by microbial invasion of the amniotic cavity.
Of the 327 participating women (150 African Americans and 177 whites), 134 individuals had preterm birth (case group) and 193 had normal term deliveries (control group). Table 2 shows selected maternal characteristics and pregnancy outcomes in the case and control groups, stratified by race. Demographic and clinical data are based on available information because all end points were not available from every participant. No significant differences between the case and control groups were observed for neonatal sex, marital status, smoking status, annual income, body mass index, and gravidity for both races. Among whites, the mean age of women delivering preterm was 2 years younger than that of women with normal term births (case group 26.8±6.1 years; control group 28.7±5.8 years; P=.04), and the proportion of women with high school education or more was higher in the case group (55.8%) than in the control group (32.6%; P=.002). Among the case group, no significant differences between the two racial groups were seen in maternal age (P=.08), gestational age (P=.55), birth weight (P=.56), Apgar score at 1 minute (P=.38), or latency (P=.10). In the control group, maternal age (25.2±5.5 compared with 28.7±5.8 years), birth weight (3,247±399 compared with 3,386±470 g) was significantly lower in African Americans (P<.007). In both races, the proportion of women with bacterial vaginosis (based on Nugent scores) was significantly greater among women in the case group than women in the control group (P=.001; Table 2). The prevalence of microbial invasion of the amniotic cavity with histologic chorioamnionitis (20% in African Americans and 26% in whites) and of funisitis (6% in African Americans and 1% in whites) in preterm birth were not significantly different between the two races.
Data presented here are log-transformed mean±standard deviation. In combined data analysis, preterm birth had significantly lower mean concentrations of F2-isoprostane (0.34±0.26 ng/mL) than those in the control group (1.30±0.69 ng/mL; P=.003). Racial disparity was not evident in our stratified analysis. African American women (0.32±0.20 ng/mL) and white women (0.36±0.29 ng/mL) in the case group had significantly lower F2-isoprostane than women in the control group (African American women in control group: 0.46±.38 ng/mL and P=.01; white women in control group: 0.45±0.31 ng/mL and P=.008; Table 1). PGF2α concentrations were higher in women in the case group than in the control group in combined data analysis (11.34±13.74 ng/mL compared with 6.63±8.04 ng/mL; P=.004), among African Americans (11.39±14.13 ng/mL compared with 6.93±8.19 ng/mL; P=.034), and in whites (11.31±13.55 ng/mL compared with 6.35±7.92 ng/mL in control group; P=.004). Similar to F2-isoprostane, PGE2 concentrations were significantly lower in women in the case group overall (4.36±6.38 compared with 7.20±8.94 ng/mL; P<.001), and racial disparity also was not evident with PGE2, although the statistical significance in African Americans were marginal. In whites, PGE2 concentrations were 3.04±4.02 in the case group and 7.18±8.07 ng/mL in the control group (P<.001) and in African Americans (6.00±8.18 compared with 7.22±9.83 ng/mL; P=.08). Similarly, overall PGD2 concentrations were significantly lower in the case group than in the control group (1.53±1.72 ng/mL compared with 2.89±4.03 ng/mL; P<.001), regardless of race (African American women in case group 1.72±1.72 ng/mL, African American women in control group 3.22±4.74 ng/mL, P=.01; white women in case group 1.31±1.71 ng/mL, white women in control group 2.63±3.37 ng/mL, P<.001). Concentrations of 6-keto-PGF1α and TXB2 between the case and control groups were not different in combined or stratified analysis.
The secondary objective of this study was to address three major risk factors associated with preterm birth and their effects on eicosanoid concentrations. Race, cigarette smoking, and microbial invasion of the amniotic cavity are factors previously documented to be associated with early delivery and adverse pregnancy outcomes. Data were stratified based on these factors in the following analyses.
Among women in the case group, comparisons of the concentrations of eicosanoid analytes between African Americans and whites showed no significant differences (Table 3), with the exception of PGE2, which doubled in African American women (6.00±8.19 ng/mL) compared with white women (3.04±4.02 ng/mL), with a marginal level of significance (P=.07) in women in the case group. No differences among any of the analytes were noted between races in controls. In general, amniotic fluid prostanoid concentrations in preterm birth or term births did not reveal evidence of racial disparity.
To document the effects of cigarette smoking, a known inducer of oxidative stress, we analyzed amniotic fluid eicosanoids after stratification for this behavior (Table 4). Although all eicosanoids were increased, F2-isoprostane was increased by approximately threefold in smokers compared with nonsmokers regardless of pregnancy status (preterm or term). F2-isoprostane was significantly higher in smokers with preterm birth (smokers −0.60±0.28 ng/mL compared with nonsmokers 0.26±0.19 ng/mL; P<.001) and also those with term births (smokers 0.92±0.39 ng/mL compared with nonsmokers 0.37±0.26 ng/mL; P<.001). Data further stratified by race showed no racial disparity in F2-isoprostane concentrations in women in either the case group or the control group (Table 5). In a combined or stratified analysis, PGF2α concentration was not significantly different among normal term births regardless of smoking status (P=.213); however, more than twofold more PGF2α was seen in smokers compared with nonsmokers (P<.001) in the preterm birth group. PGE2, PGD2, and 6-keto-PGF1α were also higher among smokers compared with nonsmokers regardless of pregnancy status (case or control group). Data stratified by race showed increased PGE2 in African American smokers compared with nonsmokers in both case (P<.001) and control (P=.04) groups. This was also evident in white smokers in the normal term birth group compared with nonsmokers (P=.002), but not in the preterm birth group (P=.48). Similar trends were evident with PGD2 and 6-keto-PGF1a. TXB2 was higher in smokers with normal term births compared with nonsmokers with normal term births (P=.001), but not in those in the preterm birth group. Only smokers in the white normal term birth group had higher TXB2 compared with nonsmokers, but it was not different in African Americans. Finally, we examined the role of microbial invasion of the amniotic cavity in amniotic fluid prostanoid concentrations. This aspect of the study was limited to preterm birth only, because infection or antibiotic treatments during pregnancy were exclusion criteria for our controls. Twenty-nine women in the case group with microbial invasion of the amniotic cavity were identified (Table 6). Overall, F2-isoprostane was marginally higher among women in the case group with infection (0.45±0.35 ng/mL) compared with women in the control group (0.31±0.22 ng/mL; P=.02). Data stratified by race showed that white women in the case group with microbial invasion of the amniotic cavity had higher F2-isoprostane compared with women in the case group without microbial invasion of the amniotic cavity (P=.008; Table 7). None of the other eicosanoids showed any differences between women in the case group with and without microbial invasion of the amniotic cavity. Although higher concentrations of PGF2α were seen in both African American and white women in the case group with microbial invasion of the amniotic cavity compared with those with no microbial invasion of the amniotic cavity, none of these data reached statistical significances in either races, likely because of sample size and variability (P=.88 and P=0.65, respectively).
Using a very sensitive and specific mass spectrometry assay, we profiled six different eicosanoids in amniotic fluid from preterm and term births. Our primary objective was to document the differential accumulation of variably bioactive eicosanoids and related arachidonic acid metabolites under well-characterized clinical conditions. Based on our primary analysis, we discovered: measurable quantities of six different eicosanoids in the amniotic fluid samples of both preterm and term births; PGF2α is the only amniotic fluid eicosanoid that is not higher at term; in support of recent reports from Lee et al,23 we also observed that PGE2 and PGD2 concentrations are approximately twofold higher at term than preterm, implying that these eicosanoids are more likely to play a role in physiological parturition than in preterm birth; and F2-isoprostane can be detected in preterm and term amniotic fluid in pregnancies with intact membranes. This is the first report of the latter observation. Higher F2-isoprostane concentrations in amniotic fluid at term suggest that oxidative stress is a contributor to normal parturition, and 6-keto-PGF1α and TXB2 concentrations do not appear to differ between preterm and term births.
Secondary analyses were performed to examine amniotic fluid eicosanoid signatures based on three classical risk factors for preterm birth: race, cigarette smoking, and microbial invasion of the amniotic cavity. Data stratified by race showed no racial disparity in the pattern of eicosanoids, with the exception that higher PGE2 concentrations were noted in African Americans with preterm birth. Cigarette smoking, a known inducer of oxidative stress, was found to be associated with high concentrations of F2-isoprostane. Regardless of the pregnancy status and race, smokers had approximately threefold higher F2-isoprostane levels, confirming its role as a biomarker of oxidative stress. Cigarette smoking shows a generalized increase in other eicosanoids. High F2-isoprostane in normal term deliveries suggests that oxidative stress is a physiological precursor or consequence of labor. Infection was also associated with higher F2-isoprostane and PGF2α in the case group. The data on PGF2α confirm those already reported by other groups; however, the F2-isoprostane findings are novel with respect to intra-amniotic infection and suggest oxidative stress as a mechanism in microbial invasion of the amniotic cavity-induced preterm birth. Contrary to previous reports, we did not see any difference in PGE2 concentrations and infection. Amniotic fluid, 6-keto-PGF1α, TXB2, and PGD2 also showed no relationship to infection status.
One of the major findings of this study is the documentation of substantially increased oxidative stress during normal labor and delivery at term. Oxidative stress previously has been hypothesized as a mediator of preterm birth, but biomarkers of oxidative stress have not been reported in preterm or term amniotic fluid. Placental and nutrient antioxidants like superoxide dismutase, catalase, and glutathione peroxidase maintain amniotic fluid redox status during pregnancy. Although there is an overall increase in antioxidant over pro-oxidant activity as gestation advances, antioxidant production is not sufficient to balance oxidative stress once gestation reaches term.34,35 Therefore, we suggest that oxidative stress exceeds a threshold triggering the onset of labor when gestation and fetal growth are complete. Oxidative stress at term and during labor could result from multiple mechanisms, such as increased maternal and fetal physiological stress, increased mitochondrial activity attributable to high energy demand associated with labor, and increased apoptosis among placental, fetal membrane, and decidual cells. Apoptosis and oxidative stress should be considered as natural physiological responses at term because of aging of the fetoplacental unit. In addition, maternal-fetal signals favoring lipid peroxidation may promote production of inflammatory or uterotonic eicosanoids (eg, PGF2α) and induce labor. Other arachidonic acid metabolites, such as F2-isoprostane, which are not known to have intrinsic uterotonic properties, may have indirect effects on labor, eg, by increasing endothelin production, a known inducer of myometrial contractility.36–39 To confirm that oxidative stress is a labor-associated change, we compared amniotic fluid F2-isoprostane from individuals not in labor undergoing cesarean deliveries to laboring women at term having normal vaginal deliveries. The latter had significantly higher amniotic fluid F2-isoprostane concentrations (0.52±0.37 ng/mL) than the former individuals (0.36±0.30; P=.004). Higher F2-isoprostane levels also were noted in cigarette smokers (regardless of race or pregnancy status), confirming the intrauterine response to this well-documented provocation of oxidative stress. These data from cigarette smokers are supported by our in vitro data in which cigarette smoke extract-stimulated normal term fetal membrane explants secrete more F2-isoprostane than vehicle-stimulated membranes.33
Because it is well-known that many cases of preterm birth have an inflammatory basis, we were not surprised by the elevated amniotic fluid PGF2α levels noted in this subgroup. However, we were impressed by the lack of oxidative stress-associated biomarker (F2-isoprostane) in preterm birth with intact membranes relative to births at term. However, increased F2-isoprostane in preterm birth was observed in pregnancies complicated by cigarette smoking or infection. These results prompt us to postulate that oxidative stress, as manifested by F2-isoprostane, is a risk-specific response to smoking and infection but not necessarily a general effector of labor process. Although considerable overlap and interaction exist between the inflammatory and oxidative stress pathways, our results suggest that these pathways may uniquely be operative in different pregnancy complications and their activation is likely dependent on the type of risk exposure.
Our study also rules out PGE2 and PGD2 as eicosanoids significantly associated with preterm labor.23,40–46 Further risk modifiers including race, cigarette smoking, and infection also were unrelated to PGE2 and PGD2 concentrations in preterm birth. Differential effects on labor by PGF2α and PGE2 have been reported previously, based on different PG receptors and modes of signal transduction.47,48 Our findings confirm those of Romero et al42 that PGF2α is the dominant uterotonic PG. Also, 6-keto-PGF1α and TXB2, markers of vasodilation and vasoconstriction, respectively, have been invoked in pregnancy complications like preeclampsia, but our data indicate that they are unlikely to be involved in preterm birth.49–51
The use of mass spectrometric methods allowed us to avoid the cross-reactivity and lack of specificity associated with classical immunological methods to detect these structurally similar compounds used in previous publications.52 In summary, coordinated (but still unclear) interactions among eicosanoid mediators of inflammation and oxidative stress are physiological signals of maturation and labor and appear to be pathophysiologically induced by some risk exposures in preterm birth. We report that term birth is associated with oxidative stress, as evidenced by increased amniotic fluid F2-isoprostane, and that certain risk categories of preterm birth also are associated with oxidative stress. These chemically related effectors are unlikely independent in their action. This hypothesis partly explains the failure of a single intervention (eg, COX inhibition) to be effective in all individuals with preterm labor. Our secondary analyses document the influence of specific risk factors of spontaneous preterm birth on amniotic fluid prostanoid levels and support oxidative stress as an underlying mechanistic factor in cases involving cigarette smoking and microbial invasion of the amniotic cavity. Depending on an individual's own risk exposures that can cause inflammation or oxidative stress (infection, behavioral, nutritional deficiencies, body mass index, psychosocial and socioeconomic stressors, genetic, race, and others), pathophysiologic manifestations can vary from individual to individual. In summary, biomarkers and mechanistic effectors of preterm and term labor are not universal; hence, clinical interventions in preterm birth cannot be generalized. Depending on an individual's unique risk factors, pathophysiologic manifestations can vary from individual to individual. This report supports the hypothesis that alternate pathways and biomarkers need to be considered in the etiology and individualized interventions tailored for effective treatment of preterm labor.
1. March of Dimes PeriStat, 2009. White Plains (NY): March of Dimes; 2010.
2. Preterm birth: causes, consequences and prevention. Washington, DC: Institute of Medicine; 2006.
3. Beck S, Say L, Betran AP, Merialdi M, Rubens C, Menon R, et al. The Worldwide incidence of preterm birth: a systematic review of maternal mortality and morbidity. Bull World Health Organ 2010;88:31–8.
4. Menon R. Spontaneous preterm birth. Race and genetics in understanding the complexities of preterm birth. Expert Rev Obstet Gynecol 2009;4:695–704.
5. Mitchell MD, Romero RJ, Edwin SS, Trautman MS. Prostaglandins and parturition. Reprod Fertil Dev 1995;7:623–32.
6. Romero R, Emamian M, Quintero R, Wan M, Hobbins JC, Mitchell MD. Amniotic fluid prostaglandin levels and intra-amniotic infections. Lancet 198614;1:1380.
7. Keirse MJ, Thiery M, Parewijck W, Mitchell MD. Chronic stimulation of uterine prostaglandin synthesis during cervical ripening before the onset of labor. Prostaglandins 1983;25:671–82.
8. McLaren J, Taylor DJ, Bell SC. Prostaglandin E(2)-dependent production of latent matrix metalloproteinase-9 in cultures of human fetal membranes. Mol Hum Reprod 2000;6:1033–40.
9. Keelan J, Helliwell R, Nijmeijer B, Berry E, Sato T, Marvin K, et al. 15-deoxy-delta12,14-prostaglandin J2-induced apoptosis in amnion-like WISH cells. Prostaglandins Other Lipid Mediat 2001;66:265–82.
10. Challis JR, Sloboda DM, Alfaidy N, Lye SJ, Gibb W, Patel FA, et al. Prostaglandins and mechanisms of preterm birth. Reproduction 2002;124:1–17.
11. King J, Flenady V, Cole S, Thornton S. Cyclo-oxygenase (COX) inhibitors for treating preterm labour. Cochrane Database Syst Rev 2005;CD001992.
12. Haas DM, Imperiale TF, Kirkpatrick PR, Klein RW, Zollinger TW, Golichowski AM. Tocolytic therapy: a meta-analysis and decision analysis. Obstet Gynecol 2009;113:585–94.
13. Romero R, Gotsch F, Pineles B, Kusanovic JP. Inflammation in pregnancy: its roles in reproductive physiology, obstetrical complications, and fetal injury. Nutr Rev 2007;65:S194–202.
14. Myatt L, Cui X. Oxidative stress in the placenta. Histochem Cell Biol 2004;22:369–82.
15. Wisdom SJ, Wilson R, McKillop JH, Walker JJ. Antioxidant systems in normal pregnancy and in pregnancy-induced hypertension. Am J Obstet Gynecol 1991;165:1701–4.
16. Morrow JD, Roberts LJ. The isoprostanes: unique bioactive products of lipid peroxidation. Prog Lipid Res 1997;36:1–21.
17. Roberts LJ II, Milne GL. Isoprostanes. J Lipid Res 2009;50(Suppl):S219–23.
18. Milne GL, Yin H, Morrow JD. Human biochemistry of the isoprostane pathway. J Biol Chem 2008;283:15533–7.
19. Morrow JD, Roberts LJ II. Mass spectrometric quantification of F2-isoprostanes as indicators of oxidant stress. Methods Mol Biol 2002;186:57–66.
20. Hertelendy F, Zakar T. Regulation of myometrial smooth muscle functions. Curr Pharm Des 2004;10:2499–517.
21. Sagawa N, Hasegawa M, Itoh H, Nanno H, Mori T, Yano J, et al. Current topic: the role of amniotic endothelin in human pregnancy. Placenta 1994;15:565–75.
22. Casey ML, Word RA, MacDonald PC. Endothelin-1 gene expression and regulation of endothelin mRNA and protein biosynthesis in avascular human amnion. Potential source of amniotic fluid endothelin. J Biol Chem 1991;266:5762–8.
23. Lee DC, Romero R, Kim JS, Yoo W, Lee J, Mittal P, et al. Evidence for a spatial and temporal regulation of prostaglandin-endoperoxide synthase 2 expression in human amnion in term and preterm parturition. J Clin Endocrinol Metab 2010;95:E86–91.
24. Menon R, Velez DV, Simhan H, Ryckman K, Jiang L, Thorsen P, et al. Multilocus interactions at maternal tumor necrosis factor-alpha, tumor necrosis factor receptors, interleukin-6 and interleukin-6 receptor genes predict spontaneous preterm labor in European-American women. Am J Obstet Gynecol 2006;194:1616–24.
25. Menon R, Thorsen P, Vogel I, Jacobson B, Williams SM, Fortunato SJ. Increased bioavailability of TNF-alpha in African Americans during in vitro infection: predisposing evidence for immune imbalance. Placenta 2007;200:28:946–50.
26. Menon R, Williams SM, Fortunato SJ. Amniotic fluid interleukin (IL)-1β and IL-8 concentrations: Ethnic disparity in preterm birth. Reprod Sci 2007;14:253–9.
27. Velez DR, Menon R, Thorsen P, Jiang L, Simhan H, Morgan N, et al. Ethnic differences in interleukin 6 (IL-6) and IL6 receptor genes in spontaneous preterm birth and effects on amniotic fluid protein levels. Ann Hum Genet 2007;71:586–600.
28. Velez DR, Fortunato SJ, Williams SM, Menon R. Preterm birth in caucasians is associated with coagulation pathway gene variants. PLoS ONE 2008;3:e3283.
29. Menon R, Taylor RN, Fortunato SJ. Chorioamnionitis—a complex pathophysiologic syndrome. Placenta 2010;31:113–20.
30. Duff P, Sanders R, Gibbs RS. The course of labor in term patients with chorioamnionitis. Am J Obstet Gynecol 1983;147:391–5.
31. Gardella C, Riley DE, Hitti J, Agnew K, Krieger JN, Eschenbach D. Identification and sequencing of bacterial rDNAs in culture-negative amniotic fluid from women in premature labor. Am J Perinatol 2004;21:319–23.
32. Hitti J, Riley DE, Krohn MA, Hillier SL, Agnew KJ, Krieger JN, et al. Broad-spectrum bacterial rDNA polymerase chain reaction assay for detecting amniotic fluid infection among women in preterm labor. Clin Infect Dis 1997;24:1228–32.
33. Menon R, Fortunato SJ, Yu J, Milne GL, Sanchez S, Drobek CO, et al. Cigarette smoke induces oxidative stress and apoptosis in normal term fetal membranes. Placenta 2011;32:317–22.
34. Gitto E, Reiter RJ, Karbownik M, Tan DX, Gitto P, Barberi S, et al. Causes of oxidative stress in the pre- and perinatal period. Bio Neonate 2002;81:146–57.
35. Watson AL, Palmer ME, Jauniaux E, Burton GJ. Variations in expression of copper/zinc superoxide dismutase in villous trophoblast of the human placenta with gestational age. Placenta 1997;18:295–9.
36. Jankovic SM, Jankovic SV, Lukic G, Radonjic V, Cupara S, Stefanovic S. Contractile effects of endothelins on isolated ampullar segment of human oviduct in luteal phase of menstrual cycle. Pharmacol Res 2009;59:69–73.
37. Janssen LJ. Are endothelium-derived hyperpolarizing and contracting factors isoprostanes? Trends Pharmacol Sci 2002;23:59–62.
38. Mitchell MD, Lundin-Schiller S, Edwin SS. Endothelin production by amnion and its regulation by cytokines. Am J Obstet Gynecol 1991;165:120–4.
39. Mitchell MD. Endothelins in perinatal biology. Semin Perinatol 1991;15:79–85.
40. Lee SE, Romero R, Park IS, Seong HS, Park CW, Yoon BH. Amniotic fluid prostaglandin concentrations increase before the onset of spontaneous labor at term. J Matern Fetal Neonatal Med 2008;21:89–94.
41. Challis JR, Lye SJ, Gibb W, Whittle W, Patel F, Alfaidy N. Understanding preterm labor. Ann N Y Acad Sci 2001;943:225–34.
42. Romero R, Munoz H, Gomez R, Parra M, Polanco M, Valverde V, et al. Increase in prostaglandin bioavailability precedes the onset of human parturition. Prostaglandins Leukot Essent Fatty Acids 1996;54:187–91.
43. Dowling DD, Romero RJ, Mitchell MD, Lundin-Schiller S. Isolation of multiple substances in amniotic fluid that regulate amnion prostaglandin E2 production: the effects of gestational age and labor. Prostaglandins Leukot Essent Fatty Acids 1991;44:253–5.
44. Berryman GK, Strickland DM, Hankins GD, Mitchell MD. Amniotic fluid prostaglandin D2 in spontaneous and augmented labor. Life Sci 1987;41:1611–4.
45. Sato TA, Mitchell MD. Preferential production of prostaglandin D2 by lipopolysaccharide stimulated human choriodecidual explants. Prostaglandins Leukot Essent Fatty Acids 2006;74:87–92.
46. Keelan JA, Blumenstein M, Helliwell RJ, Sato TA, Marvin KW, Mitchell MD. Cytokines, prostaglandins and parturition—a review. Placenta 2003;24:S33–46.
47. Madsen G, Zakar T, Ku CY, Sanborn BM, Smith R, Mesiano S. Prostaglandins differentially modulate progesterone receptor-A and -B expression in human myometrial cells: evidence for prostaglandin-induced functional progesterone withdrawal. J Clin Endocrinol Metab 2004;89:1010–3.
48. Hertelendy F, Zakár T. Prostaglandins and the myometrium and cervix. Prostaglandins Leukot Essent Fatty Acids 2004;70:207–22.
49. Mäkäräinen L, Ylikorkala O. Amniotic fluid 6-keto-prostaglandin F1 alpha and thromboxane B2 during labor. Am J Obstet Gynecol 1984;150:765–8.
50. Liu HS, Chu TY, Yu MH, Chang YK, Ko CS, Chao CF. Thromboxane and prostacyclin in maternal and fetal circulation in pre-eclampsia. Int J Gynaecol Obstet 1998;63:1–6.
51. Zhao S, Gu Y, Lewis DF, Wang Y. Predominant basal directional release of thromboxane, but not prostacyclin, by placental trophoblasts from normal and preeclamptic pregnancies. Placenta 2008;29:81–8.
52. Il'yasova D, Morrow JD, Ivanova A, Wagenknecht LE. Epidemiological marker for oxidant status: comparison of the ELISA and the gas chromatography/mass spectrometry assay for urine 2,3-dinor-5,6-dihydro-15-F2t-isoprostane. Ann Epidemiol 2004;14:793–7.
© 2011 The American College of Obstetricians and Gynecologists
Figure. No caption available.