|Home | About | Journals | Submit | Contact Us | Français|
Prematurity (<37 weeks) has been inconsistently associated with asthma and wheezing. Chorioamnionitis may promote both prematurity and inflammatory pathways in infants’ airways.
To investigate the relationship of prematurity and chorioamnionitis with the development of early childhood recurrent wheezing.
The Boston Birth Cohort (n=1096) were followed prospectively from birth to a mean age of 2.2±2 years. Perinatal and postnatal clinical data and placental pathology were collected. The primary outcome was recurrent wheezing (≥2 physician documented episodes). Secondary outcomes included physician diagnosed asthma, food allergy and eczema. Preterm children were grouped by gestational age into moderately (33-36.9 weeks) and very preterm (<33 weeks) with and without chorioamnionitis, and compared to term children without chorioamnionitis (reference group). Chorioamnionitis was diagnosed either by intrapartum fever or by placental histology findings. Logistic regression models were preformed to investigate the independent and joint associations of degree of prematurity and chorioamnionitis.
Prematurity was associated with recurrent wheezing (OR:1.7, 95%CI:1.2-2.6). However, when subjects were grouped by degree of prematurity with or without chorioamnionitis, the highest risk of wheezing (OR:4.0, 95%CI:2.0-8.0) and physician diagnosed asthma (OR:4.4 95%CI:2.2-8.7) was present in the very preterm children with chorioamnionitis. The effect on both wheezing (OR:5.4, 95%CI:2.4-12.0) and asthma (OR:5.2, 95%CI:2.3-11.9) was greater in African Americans. Neither prematurity nor chorioamnionitis were associated with food allergy or eczema.
We found a strong joint effect of prematurity and chorioamnionitis on early childhood wheezing. This effect was stronger in African Americans.
Chorioamnionitis may increase the risk of recurrent wheezing in very low birth weight infants.
A limited number of studies have explored the role of prenatal factors on the development of respiratory morbidity or later asthma.1-5 At least one study suggests that either programming in-utero or immunological changes determined in-utero affect the risk for respiratory diseases.6 Given that a significant proportion of those who develop asthma in childhood wheezed in early life, 7 investigation of prenatal and early life factors is important to better understand etiology of recurrent wheezing and possibly later development of asthma.
In this context, we chose to evaluate the role of prematurity and associated chorioamnionitis on recurrent wheezing in early life. Preterm births (<37 weeks of gestation) account for 12.5% of all births in the US and prematurity has been identified as a risk factor for asthma in many8-11 but not all studies.12, 13 A major limitation of previous studies, which may have led to inconsistent findings, is that they did not consider the underlying pathogenesis of preterm birth. A great proportion of high risk preterm births are associated with chorioamnionitis.14-16 In this regard, febrile episodes in pregnancy (especially the third trimester) have also been associated with subsequent development of asthma in offspring.17, 18 Chorioamnionitis as a potential risk factor for wheezing illness becomes even more important since like asthma,19, 20 prematurity and chorioamnionitis disproportionately affect the African American community. (IOM report 2006) This co-morbid epidemic has not been evaluated as a contributor to asthma health disparities.
The purpose of this study was to evaluate the association between prematurity and wheezing accounting for the presence of perinatal chorioamnionitis utilizing subjects enrolled in the Boston Birth Cohort, a large-scale, multi-ethnic, inner city population21 based study, which includes both term and preterm children.
The Boston Birth Cohort was initiated in 1998 and recruitment is ongoing at the Boston University Medical Center (BUMC). It consists of mother-infant pairs enrolled from a range of socio-economic strata that includes inner city poor up to middle class subjects. It is a multi-ethnic population, and the three major racial groups are black (56%), white (11%), and Hispanic (20%).
The Boston Birth Cohort was originally designed to study adverse birth outcomes, particularly preterm birth and intra-uterine fetal growth restriction. The inclusion and exclusion criteria of the parent study are as follows: Any woman admitted to the Labor and Delivery floor at BMC who delivered a singleton live infant and met our case (gestational age <37 weeks or birthweight <2,500 grams) or control (gestational age >= 37 weeks with birthweight >=2,500 grams) criteria was eligible.21 For the purposes of this follow-up study, preterm status was solely determined by gestational age not by “case” status as defined in the parent study. A small percentage of women (<1%) with pregnancies that involved in vitro fertilization, multiple gestation, chromosomal abnormalities or major birth defects, or women with congenital or acquired uterine lesions, or incompetent cervix were excluded. The Children’s Memorial Hospital (CMH) Institutional Review Board (IRB), the Boston University Medical Center (BUMC) IRB, and the Massachusetts Department of Public Health approved the parent study protocol. Under a separate CMH and BUMC IRB approved study protocol and consent process. all infants enrolled in the Boston Birth Cohort are eligible for the postnatal follow-up study.
For the parent study, mother-infant pairs were recruited 24 to 48 hours post-delivery. After obtaining signed written informed consent, we interviewed subjects using a standardized questionnaire to gather epidemiologic data. We also reviewed maternal and infant medical records using a standardized abstraction form to obtain clinical data including ultrasound findings, placental pathology reports, laboratory reports, pregnancy complications, labor and delivery course, and birth outcomes.
In the postnatal follow-up study, after written informed consent was obtained from the biological mother, visits were conducted at 6-12 mo, 2 yr, 4 yr, and 6 yr in alignment with the child’s pediatric primary care visit schedule. Mothers were interviewed using standardized postnatal health questionnaires to gather important postnatal epidemiologic data. We also reviewed the study child’s medical record using a standard medical record abstraction form to obtain clinical data including type and date of visits, clinical diagnosis including ICD 9 codes, and growth parameters.
Precise estimation of gestational age is essential for defining degree of prematurity. In this cohort, gestational age was assessed based on both the first day of the last menstrual period (LMP) as recorded in the maternal medical record and early (<20 weeks) prenatal ultrasound. This approach has been used in large hospital-based preterm studies and in our ongoing funded preterm studies.21
For this study, chorioamnionitis was considered present if the pregnancy was associated with either the presence of intrapartum fever (>38 degrees centigrade)22 based on review of medical record or by the presence of placental histological changes associated with chorioamnionitis. A detailed description of placental pathology criteria has been published by our group elsewhere.23 Briefly, all placentas sent for pathology were reviewed by the hospital’s perinatal pathologist, and a subset were independently reviewed by a second pathologist to confirm reliability. The presence and location of inflammation was reported using well-established algorithms, e.g., types of inflammatory cells were identified and reported to designate acute versus chronic inflammation, and the location of inflammatory infiltrate was denoted in these reports to specify decidual, chorionic, amnionic, chorionic plate, and umbilical cord involvement. These guidelines for histologic reporting were consistent with recommendations developed by the College of American Pathologists and have been used consistently in the literature to classify types of placental infection 24 25.
The primary outcome was recurrent wheezing (>=2 episodes of physician documented wheezing). Secondary outcomes included physician diagnosis of asthma, eczema and/or food allergy. These outcomes were determined by standardized abstraction of the medical record. The follow up of these children and determination of the outcomes were greatly facilitated by the fact that approximately 60% of subjects who are recruited in the Boston Birth Cohort continue pediatric primary care at BUMC and 38% continue at a BUMC affiliated neighborhood health center. The availability of a unified electronic medical record allows for abstraction of data with efficiency and accuracy.
We used a Chi square test to compare the baseline characteristics of term and preterm children in terms of demographic measures, SES measures, and maternal and in-utero risk factors and complications of pregnancy. We also carried out this analysis comparing children whose mothers did and did not meet our criteria for chorioamnionitis.
Potential confounding variables included infant and maternal factors. Infant factors included child’s current age and sex, breast feeding, infection during pregnancy, preterm delivery, postnatal passive smoking, and type of delivery. Maternal factors included maternal ethnicity, maternal smoking during pregnancy, maternal asthma and maternal atopy, stress during pregnancy, parity, and SES as measured by highest level of maternal education achieved. To determine the best fitted and most parsimonious model, we performed a two step process. Firstly, certain variables which have been strongly associated with early childhood wheezing in the literature were retained in the model regardless of significance. These included tobacco smoke exposure in utero, postnatal passive smoke exposure, breastfeeding, maternal asthma, and the child’s age. This resulted in three variables which were not statistically significant being kept in the model (intrapartum antibiotics, breastfeeding and postnatal passive smoking). Secondly, the other variables listed were retained in the model depending on the level of significance in the bivariate analyses. The use of a backwards stepwise regression model, forcing inclusion of the covariates felt to be important from the literature were used as a tool to assist in this selection process.
This process resulted in the inclusion of the following covariates: child’s age and sex, post-natal passive smoke exposure, maternal ethnicity, presence of maternal atopic diseases, maternal education, maternal smoking postnatally, intrapartum antibiotics, and breastfeeding status. Of note, all variables which differed between those with and without chorioamnionitis (delivery type, antibiotic use, and gestational age) were accounted for in our models. (Gestational age group was used to stratify the comparision groups and the other 2 variables were included in the model.)
We carried out an initial logistic regression analysis to evaluate the effects of prematurity overall (<37 weeks gestation) on the outcomes, with adjustment for chorioamnionitis and the other covariates listed above. We further evaluated how the variables of degree of prematurity and chorioamnionitis jointly affect the risk of early childhood wheezing. Using term pregnancies (> =37 weeks) without chorioamnionitis as the reference group, the moderate preterm (33 to 36.9 weeks) and very preterm (<33 weeks) were further stratified by the presence of chorioamnionitis. Finally, we performed race-specific analysis to determine if these associations were consistent in African American children. For this subgroup analysis we removed the variable of maternal ethnicity.
Two other secondary analyses were carried out using the same covariates. The first divided the deliveries into spontaneous and medically indicated deliveries to evaluate the role of other perinatal morbidities on wheezing. The other secondary analysis evaluated the joint effect of fever and placental inflammation on the odds of recurrent wheezing or asthma. This was carried out to evaluate the relative impact of these two variables which were used to construct the chorioamnionitis variable.
All analyses were carried out using statistical software SAS for Windows 8.20 (SAS Institute, NC).
This analysis included 771 term and 325 preterm infants enrolled from the Boston Birth Cohort who completed of at least one postnatal follow-up study visit. Recruitment from the Boston Birth Cohort to the follow up study has been approximately 90% of those approached. No differences in key variables including birthweight, gestation and prevalence of chorioamnionitis have been found from those infants yet to be enrolled in the follow up study. (data not shown).
The mean age of children at the point of last follow up was 2.2 ± 2.0 years. Baseline characteristics of the study children, including demographic measures, SES measures, and maternal/in-utero factors are displayed in Table I, stratified by term and preterm birth. As compared to the term children, preterm children had higher frequencies of maternal cigarette smoking during pregnancy, cesarean sections, prenatal antibiotic and steroid use, and chorioamnionitis as well as other pregnancy complications. Preterm children also had a higher frequency of wheezing (19.7 % vs. 10.2% in term infants) and a slightly higher prevalence of maternal asthma (15.7 vs. 11.2 in term infants). We also compared subjects with and without chorioamnionitis in Table E1. Subjects with chorioamnionitis were similar to those without chorioamnionitis in demographic variables, SES variables, and maternal in-utero factors; but had lower mean gestational age, more frequent delivery by c-section, and a higher frequency of antibiotic use.
Table E2 displays the results of a stepwise logistic regression to evaluate major covariates, in addition to prematurity and chorioamnionitis, for the outcomes of interest. We evaluated the association of prematurity overall (<37 weeks gestational age) with recurrent wheezing. We found that prematurity was associated with increased risk of recurrent wheezing (OR:1.7, 95%CI:1.2-2.6), even after adjusting for child’s age, gender, breastfeeding, postnatal exposure to maternal cigarette smoking, and chorioamnionitis as well as maternal ethnicity, education, prenatal smoke exposure and history of atopy. In contrast, prematurity did not seem to increase the risk of eczema (OR:0.6, 95%CI:0.5-0.9) or food allergy (OR:0.8, 95%CI:0.4-1.5).
We further evaluated the joint association of degree of prematurity and presence of chorioamnionitis with the outcomes of interest as shown in Table IIa. Our data suggest a joint effect when gestational age and chorioamnionitis are combined. With the exception of the moderately preterm group, children with chorioamnionitis had an increased risk of recurrent wheezing compared to term children without chorioamnionitis. Of note, the moderately preterm group with chorioamnionitis had smaller numbers of subjects. Children who were very preterm and had chorioamnionitis had the highest risk of recurrent wheezing (OR:4.0, 95%CI:2.0-8.0). Similarly, in this group the likelihood of physician diagnosed asthma was increased (OR:4.4 95%CI:2.2-8.7). Neither prematurity nor chorioamnionitis were associated with an increased risk of food allergy or eczema. There was no evidence of a statistically significant effect modification of chorioamnionitis on the associations of prematurity with the outcomes studied. Specifically, this is evidence of a joint effect of these variables, but not an interaction.
Finally, we evaluated if the above associations were consistently observed if restricted to African American children. As shown in Table IIb, we found a similar association which was greater in magnitude than that seen in the analysis of all the study subjects. For example, African American children who were very preterm with chorioamnionitis had a 5.4- fold increased risk of recurrent wheezing (OR:5.4, 95%CI:2.4-12.0) and a 5.2-fold increased risk of physician diagnosed asthma (OR:5.2, 95%CI:2.3-11.9). Again, neither prematurity nor chorioamnionitis were associated with food allergy or eczema.
As a secondary analysis, deliveries were divided into spontaneous births and medically indicated deliveries to evaluate whether other perinatal morbidities confounded the effect of chorioamnionitis on wheezing. As expected, both hypertensive disorders (including preeclampsia, eclampsia, HELLP), and other morbidities (including gestational diabetes, abruption placenta, placenta previa or incompetent cervix) were more common in medically indicated deliveries (Table E3). However, the relationships of wheezing and asthma with chorioamnionitis and prematurity persisted despite stratification by type of delivery (Table E4).
Finally, for this analysis a total of 36 children had chorioamnionitis determined by maternal fever alone, 106 children had chorioamnionitis defined by placental histopathology alone (sub-clinical chorioamnionitis), and 30 children had chorioamnionitis determined by both. We further examined this classification in two ways. Firstly, we evaluated the proportion of elevated white counts at delivery by these variables. All the groups which would have been identified as chorioamnionitis in our main analysis had a higher proportion of elevated white counts at delivery (>15000 by Gibbs Criteria)22 than individuals who do not meet our criteria of chorioamnionitis. (Table E5) Second, we carried out a subgroup analysis using groupings defined by maternal fever and placental pathology in preterm infants with available data on both placental pathology and peripartum fever. As noted in the Table E6, those with placental inflammation and no fever showed higher risk of asthma compared to those without inflammation and fever. A similar trend was observed for the outcome of recurrent wheezing, but was not significant due to the smaller sample size in the subgroup analysis. The group at highest risk were those with both inflammation and fever. Our data indicates that placental inflammation appears to be a more sensitive and robust risk factor for wheezing or asthma than fever alone.
In-utero events can affect early life immune responses,26, 27 the development of atopic diseases in the offspring,2-5 and even result in epigenetic effects on subsequent generations.6 To date, no prospective studies have examined the joint effect of prematurity and chorioamnionitis in relation to the development of respiratory morbidity and atopic diseases in early childhood. In this study, we found that while prematurity was associated with recurrent wheezing in this cohort of children in the first few years of life, this association was greatly augmented by the severity of prematurity and presence of chorioamnionitis. The children who were very preterm and had chorioamnionitis had the greatest risk of recurrent wheezing. Furthermore, this association was much stronger in African American children. Our findings provide new information to a long-standing clinical and public health challenge, that is, increased risk of respiratory morbidities in African American children.
There were a number of features of our cohort which allowed us to carry this study out. We carried out a thorough evaluation of prenatal factors including placental pathology which provided a very sensitive indicator of intrauterine infection or inflammation. Ours is one of the few prospective studies with large enough numbers to allow stratification by severity of prematurity and presence of chorioamnionitis. Our study also has significant minority representation, with a high frequency of African Americans. Like asthma,28, 29 prematurity and chorioamnionitis disproportionately affect the African American community (IOM report 2006). Our findings underscore the importance of examining the role of in-utero factors on the later development of respiratory morbidities. They also raise questions about whether the persistent and high prevalence of prematurity and chorioamnionitis, especially in African Americans, are linked in a causal fashion to the asthma epidemic and ethnic disparity in respiratory outcomes.
Caution is needed in interpreting our findings because of a number of potential limitations. Although we have shown associations with physician diagnosed asthma, we focused on recurrent wheezing given the average age of the children (around 2 years). It would be important to evaluate if the associations persist as we continue to follow these children and if increased rates of recurrent wheezing correlate with increased rates of asthma diagnoses at later ages. We were not able to follow all of the patients in the initial cohort as recruitment into the follow up cohort is ongoing. Subjects already enrolled in the follow up study showed no differences from those not yet recruited in key variables including birth-weight, gestation and percent of the cohort with chorioamnionitis.
This study focused on two prominent prenatal factors (prematurity and chorioamnionitis). While our analyses attempted to adjust for many potential covariates, we need to recognize that there are many other important prenatal factors that could affect the associations, such as intrapartum antibiotics and mode of delivery. Intrapartum antibiotics and mode of delivery can affect types of commensal organisms present in the mother30 and infant31 , and atopic disease prevalence.32-34 However, we do not feel that mode of delivery or intrapartum use of antibiotics influenced our findings; intrapartum antibiotics was included as a covariate in our analyses. Also, modification of the commensal flora by antibiotics or mode of delivery should affect allergic diseases in general, not just wheezing. Since a significant proportion of the diagnoses of chorioamnionitis were made by histological diagnosis alone and represent subclinical inflammation, these mothers would not have had a clinical indication to receive intera-partum antibiotics. Notably, subclinical chorioamnionitis, determined by placental pathology, makes up the majority of our cases, and our analyses suggest that this is a stronger determinant than maternal fever alone. Finally, our findings compare subjects of similar gestational age, thereby decreasing any differential distribution of delivery mode. While we have controlled for modifiable risk factors such as smoking and breastfeeding in the analysis, we have insufficient sample size to stratify by all of these variables or evaluate interactions between all of these variables in addition to our main variables of interest, prematurity and chorioamnionitis. Further studies with increased sample size may be able to address these potential interactions.
Other potential mechanisms which need to be considered include possibilities such as increased susceptibility to early life respiratory infections and pneumonias such as RSV and the impact of these events on our findings. Also, further studies are needed to determine if the effects associated with chorioamnionitis are due to the administration of antipyretics in the perinatal period.
Several important points merit further discussion with respect to the prior literature. Prematurity has been associated with later childhood asthma in a number of studies3, 9-11 While the effect of gestational age has not been consistent in all studies12, 13, a recent meta-analysis of 19 studies reported that preterm infants are more likely than term infants to develop asthma, with preterm infants having a 7% increased risk8. One potential reason for discrepant results in the literature is that these studies did not evaluate the causes of prematurity. As emphasized in the 2006 IOM (Institute of Medicine) report, chorioamnionitis is a major pathway leading to the development of preterm birth.14-16 The potential role of chorioamnionitis as a cause of respiratory morbidity has been evaluated in only a few studies to date. In these studies, febrile episodes in pregnancy (especially the third trimester) were associated with subsequent development of asthma in offspring.17, 18 In keeping with the fact that chorioamnionitis is much more common in very preterm infants,35 we found our greatest effect in this group of infants.
Secondly, we did not find an association of prematurity and chorioamnionitis with atopic dermatitis or food allergy. This is also consistent with other published studies. A number of studies of prematurity suggest either no association with atopic diseases36 or decreased risk of atopic outcomes including atopic dermatitis37, and allergic rhinitis.38 The differential associations for respiratory morbidities vs. non-respiratory atopic diseases suggest differential pathways of disease promotion. In our study, chorioamnionitis increased the risk of wheezing and physician diagnosed asthma, but not atopy. While this outcome is important in itself, it may or may not equate to asthma later in life. If it does predispose to persistent asthma, it would suggest that some prenatal factors may lead to asthma via non-atopic pathways such as fetal programming or epigenetic effects in-utero. However, the biological links between prematurity or chorioamnionitis and wheezing are not well established.
Finally, there is some biological rationale for our findings. Chorioamnionitis is associated with a strong pro-inflammatory response with increased levels of TNF-α, IL-6, and IL-8.39, 40 These inflammatory cytokines may be a response to infection and also a trigger for premature delivery. Specifically, human decidual cells increase the production of delivery promoting factors such as PGE2 and PGF2 in response to inflammatory cytokines including IL1a, IL1b, and TNF-α.41 In animal models, administration of a PDE4 inhibitor prevents preterm delivery.42 Some of the same cytokines which are a response to infection may be involved in chronic respiratory disease development. Of these inflammatory cytokines, TNF-alpha variants have been found to be associated with recurrent wheezing after RSV bronchiolitis.43 IL-8 levels were increased in children who developed wheezing illnesses in response to Rhinovirus,44 and in another cohort, individuals who wheezed with Rhinovirus were at additional risk of persistent wheezing till follow up at age 3 years.45 Thus, the same cytokines which are generated as part of the inflammatory response to chorioamnionitis may play a role in asthma and responses to respiratory viral illness.
In summary, we found a strong joint effect of prematurity and chorioamnionitis on early childhood recurrent wheezing. This effect was even stronger in African Americans. Further investigation is warranted to evaluate if the associations persist as we continue to follow these children to older age and if increased rates of recurrent wheezing correlate with increased rates of later asthma. Our findings remain to be confirmed by future studies in other populations.
Source(s) of support: The parent study is in part supported by the March of Dimes PERI grants (PI: Wang, 20-FY02-56), NIEHS (PI: Wang, R21 ES011666), and NICHD (PI: Wang, R01 HD041702). The follow-up study is in part supported by Food Allergy Project.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.