|Home | About | Journals | Submit | Contact Us | Français|
Spontaneous term delivery involves activation of placental corticotropin-releasing hormone and the fetal adrenal axis, but the basis for extreme preterm delivery is unknown. Our objective was to determine whether placental corticotropin-releasing hormone is activated in extreme spontaneous preterm delivery.
1506 mothers delivering at less than 28 weeks gestation were enrolled. Each mother/infant pair was assigned to the category that described the primary reason for hospitalization. Observers who had no knowledge of patient categorization assessed placenta microbiology, histology, and CRH expression. These were correlated with the primary reason for hospitalization.
Among infants delivered at <28 weeks gestation, spontaneous (versus induced) delivery was associated with less placental corticotropin-releasing hormone expression and more frequent signs of placental inflammation and infection.
Inflammation and infection, rather than premature activation of the fetal adrenal axis, should be the major focus of research to prevent extremely preterm human birth.
In normal parturition, prostaglandins initiate uterine contraction, aided by a fall in progesterone’s effect to maintain uterine quiescence.1,2 As in other mammals, activation of the human fetal adrenal axis plays an important role in normal term parturition, with anthropoid primates uniquely employing placental corticotropin-releasing hormone (CRH) to drive this process.3–7 However, the mechanisms that lead to normal and very preterm delivery likely differ in ways that should guide the development of better diagnostic and therapeutic interventions.8
The ELGAN study was designed to identify pregnancy and neonatal characteristics and exposures that increase the risk of neurologic disorders in ELGANs (the acronym for Extremely Low Gestational Age Newborns)9. During the years 2002–2004, women delivering between 22 weeks, 0 days and 27 weeks, 6 days gestation at one of 14 participating institutions in 11 cities in 5 states were asked to enroll in the study. The individual institutional review boards approved the enrollment and consent processes. Mothers were approached for consent either upon antenatal admission or shortly after delivery, depending on clinical circumstance and institutional preference.
Of 1506 infants enrolled in the ELGAN study, placentas were collected from 1411, RNA was prepared from 1370, and 1219 contained non-degraded RNA. These 1219 placentas constitute the sample for all the tables. The Northern blot data (Figure 2) are based on measurements from 955 placentas with detectable CRH expression. The three samples (1506, 1411, 1219) are comparable in their proportion of children delivered for maternal and fetal reasons, and distribution of gestational ages and birth weights.
The clinical circumstances that led to preterm delivery were operationally defined using both data from the maternal interview and data abstracted from the medical record.10 Each mother/infant pair was assigned to the category that described the primary reason for hospitalization: Preterm labor was defined as progressive cervical dilation with regular contractions and intact membranes. The diagnosis of preterm premature rupture of fetal membranes (pPROM) was defined as the presence of vaginal pooling with either documented nitrazine positive testing or ferning prior to regular uterine activity. Preeclampsia was defined as new onset hypertension and proteinuria of sufficient severity to warrant delivery for either a maternal or fetal indication. For a diagnosis of cervical insufficiency, a woman had to present with cervical dilation of greater than two centimeters, in the absence of membrane rupture and detected or perceived uterine activity. Placental abruption was defined as presentation with significant amount of vaginal bleeding (either documented in the medical record or a post-partum hematocrit <24%) and a clinical diagnosis of placental abruption in the absence of cervical change. Although painful uterine contractions were not required, most women given this diagnosis tended to present with vaginal hemorrhage often accompanied or very soon followed by labor. In addition, placenta abruption, as defined, tended to be much more like the other spontaneous disorders of labor, pre-labor rupture of membranes, and cervical insufficiency.10,11 Presentations under the category of fetal indication/IUGR included severe intrauterine growth restriction based on antepartum ultrasound examination, non-reassuring fetal testing, oligohydramnious, and Doppler abnormalities of umbilical cord blood flow.
Preterm labor, pre-labor PROM, cervical insufficiency, and placental abruption were then grouped into “spontaneous” deliveries because they were initiated without medical assistance. “Induced for maternal or fetal reasons” is applied to deliveries that were not initiated by the mother or fetus, but by the physician to preserve the health of mother or fetus.
Delivered placentas were placed in a sterile exam basin and transported to a sampling room. Eighty-two percent of the samples were obtained within 1 hour of delivery.
In placentas from which RNA was prepared, CRH and ACTB mRNA (Actin B, a control mRNA ubiquitously expressed in all cells) expression was analyzed by quantitative reverse transcription-PCR and RNA blot. Biopsies of the fetal side of the placenta to a depth of approximately 1 cm (~500 mg) were obtained shortly after birth, frozen in liquid nitrogen, and stored at −80°C. Samples (~100 mg) were crushed to a fine powder in Dry Ice, and total RNA was extracted using TRI reagent (Sigma) according to the manufacturer’s instructions.
Control RNA was isolated from 7 normal term placentas delivered by Cesarean section and pooled. This RNA was used to create CRH and ACTB cDNA to make standard curves for qPCR analysis (see following). Serial dilutions of this RNA were also directly analyzed for CRH mRNA expression to establish a linear range for RNA blot quantification (see following).
cDNA was prepared in 10 microliter reactions using iScript cDNA Synthesis (Biorad) and 0.5 mcg of placental RNA. Subsequently, 2 microliters of this reaction was amplified by quantitative PCR using IQ SYBR Green Supermix (Biorad), a Biorad iQ5 qPCR apparatus, and the following primers: CRH forward primer: AGA GAA AGC CCC CGG AGA, CRH reverse primer: ATG TTA GGG GCA CTC GCT TC), ACTB forward primer: CGC GAG AAG ATG ACC CAG AT, ACTB reverse primer: GTA CAT GGC TGG GGT GTT G). qPCR conditions consisted of denaturation: 95.0°C for 3 minutes and amplification: (95.0°C for 10 seconds, 60.0°C for 40 seconds, repeated for 45 cycles). The standards were prepared by PCR amplification using these same primers and control placenta RNA (see preceding). ACTB results were used to confirm the quality of the mRNA and correct for total RNA amount. Of the 1370 samples, 151 contained degraded RNA (as judged by lack of ACTB and CRH signals by qRT-PCR), so that final analyses were performed on 1219 samples. To normalize the results among the different experimental runs, all CRH standards across all 17 qPCR runs were averaged and each of the 1219 experimental samples was evaluated against that average CRH standard curve. Similarly, all ACTB standards across all 17 qPCR runs were averaged and each of the 1219 experimental ACTB values was evaluated against that average ACTB standard curve. Each CRH value was then divided by its corresponding ACTB value to obtain the relative CRH mRNA concentrations, which are displayed in Figure 1.
We initially analyzed the above RNA by blot hybridization12 to examine CRH mRNA expression in different placental regions, and to assess CRH mRNA integrity and stability during the collection process. We subsequently found that samples from all four placental quadrants had equivalent expression of CRH mRNA, and that it remained intact for up to 10 h at room temperature prior to sample retrieval and storage at −80°C. The control RNA was used to normalize the relative amount of CRH mRNA in the experimental samples among the 68 blots, which differed slightly due to variations in probe labeling, hybridization, washing, and exposure. To correct for the amount of total RNA loaded within each gel, each filter was stripped and rehybridized to a 32P-UTP-labeled antisense human beta-actin ACTB (Ambion) riboprobe. Of the 1370 samples, 415 contained undetectable ACTB mRNA and were considered to contain degraded RNA, so that final analyses were performed on 955 samples. Each CRH mRNA value was then divided by its corresponding ACTB mRNA value to obtain relative CRH mRNA concentrations, which are displayed in Figure 2. RNA blot and qRT-PCR methods revealed similar relationships between CRH mRNA expression and placental and pregnancy characteristics, except that CRH mRNA measurement by RNA blot was less sensitive and more variable (Figures 1 and and2,2, and data not shown).
The microbiologic procedures are described in detail elsewhere13. Briefly, the frozen samples were allowed to thaw at room temperature, a portion approximately 1 cubic cm was removed and weighed, then diluted 1:10 with sterile phosphate buffered saline (PBS), homogenized and plated on selective and non-selective media, including pre-reduced brucella-base agar with 5% sheep blood enriched with hemin and vitamin K1, tryptic soy agar with 5% sheep blood, chocolate agar, and A-7 agar.
The ELGAN pathologists participated in the creation of the manual of procedures as well as in defining the histologic characteristics and designing the data collection form14. In addition, they participated in exercises to reduce observer variability.
In keeping with the guidelines of the 1991 CAP Conference15, representative sections were taken from all abnormal areas as well as routine sections of the umbilical cord and a membrane roll, and full thickness sections from the center and a paracentral zone of the placental disc.
Study pathologists examined the slides for histologic characteristics listed on a standardized data form. For multiple births, separate forms were filled for each newborn. Twins with fused placentas also had multiple forms filled.
The presence or absence of infarcts and inter-villous fibrin, fetal stem vessel thrombosis, and decidual hemorrhage and fibrin deposition consistent with abruption were coded as present or absent. Chorionic villi were scored for syncytial knots (0=no, 1= occasional, 2= increased).
Inflammation of the membranes was described in detail. Level 3 severity of chorionic plate inflammation required more than 20 neutrophils/20x. Grade 3 inflammation of the amnion required numerous large or confluent foci, while grade 4 required necrosis. Inflammation in the chorion/decidua was similarly, but separately graded.
Inflammation in the umbilical cord was graded from 0–5. Grade 3 required neutrophils in perivascular Wharton’s jelly, grade 4 required panvasculitis and umbilical cord vasculitis extending deep into Wharton’s jelly, and grade 5 required a ‘Halo lesion’ (ring of precipitate in Wharton’s jelly encircling each vessel). Neutrophilic infiltration into fetal stem vessels in the chorionic plate required that neutrophils appeared to have migrated towards the amnionic cavity.
The collection of newborn whole blood, preparation of dry blood spots and their processing for protein analysis is described elsewhere16. Briefly, after the newborn’s blood was drawn on postnatal day-1 for clinical indications, the tip of syringe containing the remainder was blotted with filter paper and stored at −80°C. Protein elution from 12mm punched biopsies of the frozen blood spots was performed as described16,17. The following 25 proteins were measured by the Meso Scale Discovery multiplex platform as described16: CRP, Serum Amyloid A (SAA), Myeloperoxidase (MPO), Interleukin-1β, IL-6, IL-6R, TNFα, TNF-R1, TNF-R2, IL-8, MCP-1, MCP-4, MIP-1β, RANTES, Interferon-inducible T cell Alpha-Chemoattractant (I-TAC), ICAM-1, ICAM-3, VCAM-1, E-selectin, MMP-1, MMP-9, VEGF, VEGF-R1, VEGF-R2, and IGFBP-1. In light of the observations that protein concentrations varied with gestational age at delivery18, the concentrations of most proteins did not follow a normal distribution, and because we were interested in the contribution of high concentrations, we dichotomized the distribution of each protein’s concentration at the 75th centile among children in each gestational age category (i.e., 23–24, 25–26, 27 weeks).
We evaluated the generalized null hypotheses that CRH mRNA expression in the placenta is NOT associated with:
CRH mRNA quantitation was performed without the knowledge of any other data associated with the study subjects. In the entire ELGAN Study sample, the pregnancy disorder that preceded preterm delivery10, organism recovery from the placenta13, histologic characteristics of the placenta14,19, concentration of inflammation-related proteins in the newborn’s circulation17,18, and CRH content of the placenta (Table 1) varied with gestational age at delivery. We adjusted for gestational age in groups of weeks (23–24, 25–26, 27).
We summarize some of our data with box and whiskers displays of the central tendency and dispersion of CRH mRNA content in infants grouped by the pregnancy disorder that led to their delivery so early before term (Figures 1 and and2).2). The central tendency is indicated by the line close to the middle of the box, which is the median, and by the top and bottom of each box, which indicate the 25th and 75th centiles. The dispersion of CRH mRNA content is indicated by the length of the vertical lines that emanate from the box, as well as by the block dots, which identify outliers. We did not see a “route of delivery” effect (vaginal vs. Cesarean) on CRH mRNA content in any of our analyses. Thus, we did not adjust for route of delivery in this study.
We created multinomial (also called polytomous or polychotomous) logistic regression models to evaluate the risk of a CRH mRNA concentration in the lowest and highest quartiles with the middle half as the referent. The contributions of antecedents/correlates are presented as risk ratios with 95% confidence intervals.
Because gestational age, pregnancy disorder, placenta histology, placenta microbiology, and CRH mRNA content were highly interrelated, we conducted additional analyses in two subgroups, those with spontaneous deliveries, and those delivered for maternal or fetal reasons. Odds ratios for lowest and highest CRH quartile associated with inflammation-related proteins and placental microrganisms were calculated using the middle two quartiles as the referent group.
To assess the contribution of a partial or complete course of antenatal corticosteroid to the CRH mRNA content, we created multivariable models of the risk of a CRH mRNA concentration in the highest quartile that included each pregnancy disorder, gestational age groups and the antenatal corticosteroid course. Multiple separate courses were not evaluated.
We measured placental CRH expression and other characteristics in 1219 pregnancies of extremely low gestational age (< 28 weeks) newborns10, and related these to the pregnancy disorders that lead to very preterm delivery (Table 1). Surprisingly, we found that spontaneous preterm deliveries were associated with lower CRH expression and higher frequency of markers of placental inflammation and infection than deliveries induced for maternal or fetal reasons.
In our cohort, the placentas of women who were induced to deliver because of preeclampsia tended to have the highest CRH expression, those who were induced to deliver for a fetal indication had intermediate CRH expression, and those from women who delivered spontaneously had the lowest CRH expression (Figure 1, Tables 2 and and3).3). As gestational age at birth increased, so did the percent of placentas that had CRH expression in the highest quartile (Table 2). However, these differences in CRH among the various pregnancy disorders are not explained by disparities in gestational age or antenatal corticosteroid use (Table 3). The association of induced delivery with the highest quartile of CRH RNA quartile was thus not confounded by corticosteroid receipt.
Placentas that had an infarct or an increased number of syncytial knots were significantly more likely than those without these characteristics to have a CRH concentration in the highest quartile, even after adjusting for gestational age (Table 4). In contrast, placentas that had inflammation were significantly less likely than placentas without inflammation to have a CRH concentration in the highest quartile, and at increased risk of having a CRH concentration in the lowest quartile. Histologic characteristics less clearly associated with inflammation (thrombosis of the fetal stem vessels, decidual hemorrhage and fibrin deposition) were also associated with an increased risk of a CRH concentration in the lowest quartile. Placenta infarcts were also associated with increased odds of being in the lowest CRH quartile (Table 4).
As further evidence for an association between inflammation and low placental CRH expression, those newborns whose placental CRH concentration was in the lowest quartile were more likely to have an elevated concentration in day-1 blood of IL-1β, IL-6R, TNF-R2, IL-8, MIP-1β, and VEGF-R2 (Table 5). In contrast, those with CRH in the highest quartile were significantly less likely to have an elevated concentration in day-1 blood of SAA, MPO, IL-6R, TNFα, TNF-R1, TNF-R2, ICAM-1, ICAM-3, VCAM-1, E-SEL, MMP-1, MMP-9, VEGF, and VEGF-R2, and more likely to have an elevated concentration of IGFBP-1.
One potential cause of inflammation during pregnancy is infection. Compared to placentas that did not yield any organism, those that harbored any of the group of organisms evaluated (i.e., pure cultures, mixed cultures, aerobe, anaerobes, Mycoplasma species, normal skin flora or normal vaginal flora) were at significantly reduced risk of having a CRH concentration in the highest quartile (Table 6).
All of the preceding data document that preeclampsia and inflammatory pregnancy disorders have opposite relationships with CRH expression. To evaluate if inflammation had an independent effect (separate from the one reflecting merely the absence of preeclampsia), we limited one set of analyses to placentas of infants delivered spontaneously, and excluded placentas of infants induced to deliver for maternal or fetal indications, including preeclampsia and fetal growth restriction (Table 7). In a model with gestational age variables only (left data column), low gestational age was associated with the lowest CRH mRNA concentrations. Histologic inflammation variables by themselves were also strongly associated with low CRH expression (middle data column). When variables for both gestational age and histologic inflammation were evaluated in the same model, the strong relationship between inflammation and low CRH expression persists, as does the strong relationship between low gestational age and low CRH expression (right data column).
In this large cohort of prospectively enrolled infants delivered prior to 28 weeks, the expression of CRH was lowest in the placentas of infants who delivered spontaneously, and highest in those of infants delivered for maternal or fetal reasons. The claim that CRH mRNA expression is higher in preterm placentas that are inflamed, which is contrary to what we found, was based on only six inflamed placentas (vs. 481 in our cohort) and did not compare these placentas to any delivered for fetal or maternal reasons.20 Supporting our results, Struwe et al.21 reported elevated CRH expression in placentas of infant with intrauterine growth restriction. They also found increased expression of placental IGFBP1,21 consistent with our findings in newborn blood. We found that spontaneous deliveries were most likely associated with infection and inflammation of the placenta, and with systemic inflammation. These findings were not due to differences in gestational length, antenatal steroid use, or whether preeclampsia was present.
Our study has several strengths. First, we included a large number of infants, making it unlikely that we have missed important associations due to lack of statistical power. Second, we selected infants based on gestational age, not birth weight, in order to minimize confounding due to factors related to fetal growth restriction. Third, we collected all of our data prospectively. Fourth, examiners were not aware of the medical histories of the placentas they examined, thereby minimizing bias. Fifth, we measured CRH mRNA expression using two completely different methods that yielded similar results. Sixth, our protein data are of high quality, and have high content validity.11,16 The weaknesses of our study are those of all observational studies. We are unable to distinguish between causation and association as explanations for what we found.
Although normal delivery near term depends upon linking the timing of fetal maturation to parturition via the fetal adrenal-placental unit driven by placental CRH,2,4,5,7 extreme preterm delivery is likely driven by different mechanisms, conceptualized as the preterm parturition syndrome.22 For example, a single measurement of maternal blood CRH at 16–20 weeks gestation predicted the timing of spontaneous labor in an unselected population of pregnant women7 but not in women at high risk for preterm delivery.23 Progesterone supplementation is only half as effective preventing extreme preterm delivery as compared to later recurrent preterm delivery.24 In addition, early preterm labor is more likely than term labor to be associated with intrauterine infection and cytokine production.25 Cytokines, via the prostaglandins that they stimulate,26 increase uterine contraction,27 and might directly initiate labor, independent of fetal signals and relatively unresponsive to progesterone’s quiescent effects.
In summary, among infants born before the 28th week of gestation, spontaneous deliveries were not associated with premature activation of the fetal adrenal axis, but were associated with inflammation. This suggests that the latter should be the major focus of research to prevent extremely preterm human birth.
Financial Support: This work was supported by a cooperative agreement with the National Institute of Neurological Disorders and Stroke (5U01NS040069-05), by a center grant award from the National Institute of Child Health and Human Development (NIH-P30-HD-18655), and by the Timothy Murphy Fund.
The authors gratefully acknowledge the contributions of their subjects, and their subjects’ families, as well as those of their colleagues at all the ELGAN study institutions.
DISCLOSURE: None of the authors have a conflict of interest.
Author ContributionsStudy concept, supervision, or design, including technical design: Majzoub, Leviton, Allred, Trivedi, Joachim, Fichorova, Kliman, Onderdonk.
Acquisition of data: Trivedi, Joachim, McElrath, Fichorova, Onderdonk, Heitor, Chaychi.
Analysis and interpretation of data: Majzoub, Leviton, McElrath, Kliman, Allred, Fichorova, Onderdonk.
Statistical analysis: Allred, Leviton.
Drafting of the manuscript: Majzoub, Leviton.
Critical revision of the manuscript for important intellectual content: Majzoub, Leviton, Trivedi, Joachim, McElrath, Kliman, Allred, Fichorova, Onderdonk, Heitor, Chaychi.
Majzoub and Leviton had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.