Effect of Chorioamnionitis on Early Childhood Asthma

Objective  To examine the association between chorioamnionitis and childhood asthma based on gestational age at birth and race/ethnicity.

Design  A retrospective cohort study using the Kaiser Permanente Southern California (KPSC) Matched Perinatal records.

Setting  Kaiser Permanente Southern California, Pasadena, California.

Participants  All singleton children born in KPSC hospitals between 1991 and 2007 (N = 510 216).

Main Exposure  Clinically diagnosed chorioamnionitis.

Main Outcome Measures  Physician-diagnosed asthma in children aged 8 years or younger.

Results  The incidence rates of asthma among preterm- and full term–born children of pregnancies complicated by chorioamnionitis were 100.7 and 39.6 per 1000 person-years, respectively (incidence rate ratio, 2.9; 95% confidence interval [CI], 2.6-3.3). Children aged 8 years or younger with asthma were more likely to be born to women who were aged 35 years or older, African American, had 13 or more years of education, had maternal asthma, used antibiotics, had chorioamnionitis during the pregnancy, and had a male child. Multivariable Cox regression analysis revealed that children born at 23 to 28, 29 to 33, and 34 to 36 weeks' gestation after pregnancies complicated by chorioamnionitis had a 1.23-fold (95% CI, 1.02-1.49), 1.51-fold (95% CI, 1.26-1.80), and 1.20-fold (95% CI, 1.03-1.47), respectively, increased risk of asthma compared with children of similar gestational age born after pregnancies not complicated by chorioamnionitis. A preterm pregnancy complicated by chorioamnionitis was associated with increased risk of asthma among white (hazard ratio [HR], 1.66; 95% CI, 1.32-2.07), African American (HR, 1.98; 95% CI, 1.60-2.44), and Hispanic (HR, 1.70; 95% CI, 1.45-2.00), but not Asian/Pacific Islander (HR, 1.15; 95% CI, 0.83-1.58) women.

Conclusion  Findings suggest that chorioamnionitis at preterm gestation is independently associated with increased risk of childhood asthma.

Asthma is a heterogeneous disease of the airways that is characterized by increased responsiveness of the tracheobronchial tree to a variety of stimuli and airway obstruction that is partially or completely reversible.^1,2 The underlying mechanism for the increased airway reactivity is unknown; however, there is a consensus that airway inflammation plays an important role. According to the US Department of Health and Human Services, in 2006, 9.9 million (14%) children younger than 18 years have been diagnosed with asthma, of which 6.8 million (9%) had current asthma,³ making it the most common chronic childhood disease. The prevalence is about 25% higher among American Indian, Alaskan Native, and African American children than among their white counterparts.⁴ The high rates of emergency department visits and hospitalizations,⁵ learning problems, school days missed, and disruption of daily activities⁴ clearly attest to childhood asthma's importance as a public health problem. In 1994, the annual costs of illness related to asthma in those younger than 18 years and of work lost for caretakers were estimated to be $3.2 billion and $956.7 million, respectively.^6-8

Chorioamnionitis, an inflammation of the maternal-fetal interface, complicates approximately 8% of pregnancies. Ascending bacterial infections at the maternal-fetal interface^9,10 as well as inflammatory processes at sites remote from the female genital tract^11-14 are potential pathoetiologic mechanisms of the condition. Chorioamnionitis is a major cause of maternal, fetal, and neonatal morbidity such as stillbirth,¹⁵ premature rupture of membranes,¹⁶ preterm birth,^17-19 bronchopulmonary dysplasia,^20,21 and neonatal intensive care unit admissions.²² More than half of preterm births are thought to be associated with histological chorioamnionitis.^17,18,23,24

There is persuasive evidence from studies based on animal models and human subjects that there is an association between chorioamnionitis and fetal lung injuries at an earlier gestational age.^25-29 Elevated umbilical vein and amniotic fluid interleukin 6 (IL-6) concentrations have been found in preterm neonates who were born with histological evidence of chorioamnionitis,³⁰ and elevated levels of IL-6, IL-8, IL-1β, and tumor necrosis factor α in the amniotic fluid of preterm delivery may predict risk of bronchopulmonary dysplasia.^28,30 While these important associations indicate lasting injury to the developing lung tissue, the relationship between chorioamnionitis and childhood asthma has not been examined in detail. Furthermore, whether gestational age at birth and race/ethnicity modify the magnitude of the association remains undetermined. Therefore, we undertook this study to test the hypotheses that (1) chorioamnionitis is associated with increased risk of asthma in children younger than 8 years and (2) gestational age at birth and race/ethnicity modify the magnitude of association.

We conducted a retrospective cohort study to examine the association between chorioamnionitis and asthma in singleton children born alive in a large health maintenance organization (Kaiser Permanente Southern California [KPSC]). The study was approved by the KPSC institutional review board.

The data for this study were derived from the 1991 through 2007 KPSC Matched Perinatal Records, which match birth certificate records for all KPSC births (Perinatal Services System) to maternal and infant hospitalizations in all KPSC hospitals, outpatient physician encounters, and pharmacy records. The Perinatal Services System records contain information on pregnancies and deliveries, maternal sociodemographic and behavioral characteristics, complications of pregnancy, labor, and delivery, and fetal and neonatal outcomes of all births in KPSC hospitals (more than 30 000 annual deliveries).

The cohort comprised singleton live and still births in all KPSC hospitals from 1991 through 2007 (N = 510 216). Of this group, we sequentially excluded stillbirths (n = 2023), spontaneous and induced abortions (n = 603), pregnancies delivered at fewer than 23 weeks' gestation (n = 2480), children with birth defect (n = 2676), and neonatal mortality cases (n = 956). The justification for restricting the analyses to pregnancies that lasted 23 or more weeks is to avoid errors in gestational age estimation and based on the fact that children who were born at this gestational age are less likely to survive. Furthermore, we excluded children who were born in the KPSC hospital but did not become a health plan member within 60 days of birth (19% of the total sample). This left us with a total of 397 852 singleton children born alive for analysis.

The exposure variable was clinically diagnosed chorioamnionitis, and the outcome variable was physician-diagnosed asthma in children younger than 8 years. International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes 762.7 and 658.4x were used to identify clinical diagnosis of chorioamnionitis, and ICD-9-CM code 493.x and presence of at least 2 prescriptions specific to asthma medication (β-agonists or asthma controller medications or their combination) were used to identify physician diagnosis of asthma.

Gestational age was defined as the time between the first day of the last normal menstrual period (LMP) and delivery of the fetus calculated in completed weeks of gestation. When data on last menstrual period were unavailable or when the last menstrual period–based gestational age estimate was not consistent with birthweight (<4% of records), a clinical estimate of gestational age was used as a substitute. Variables considered as potential confounders or mediators for the association between chorioamnionitis and childhood asthma included maternal race/ethnicity (grouped as non-Hispanic white, non-Hispanic African American, Hispanic, Asian/Pacific Islander, and other), maternal age (<20, 20-29, 30-34, and ≥35 years), maternal education (<12, 12, and ≥13 years of completed schooling), timing of prenatal care initiation (early or first trimester vs later/no prenatal care or second trimester), smoking during pregnancy (yes or no), antibiotic use during pregnancy (yes or no), maternal asthma (yes or no),³¹ and child's sex (male or female).

The distribution of maternal demographic and behavioral characteristics as well as child's characteristics were examined by the presence or absence of asthma diagnosis in children younger than 8 years. Categorical data were compared using the χ² test. Follow-up of children started from the date of delivery and ended with an outcome (asthma diagnosis) or, when censoring occurred, on the earliest of the following dates: health plan disenrollment, eighth birthday, non–asthma-related death, or end of study (December 31, 2007). A Cox proportional hazards model was fit to examine the association between exposure to chorioamnionitis in utero and asthma at a later age after controlling for potential confounders (maternal age, race/ethnicity, education, initiation of prenatal care, smoking and antibiotic use during pregnancy, maternal asthma, and child's sex).³² Because preterm birth lies on the causal pathway between chorioamnionitis and childhood asthma, we examined the associations without adjustment for gestational age at delivery in the model. However, we examined the independent effect of chorioamnionitis on childhood asthma after stratifying the data set by gestational age categories. The categorization of gestational age into extreme (23-28 weeks), very (29-33 weeks), and moderate (34-36 weeks) preterm birth as well as birth at term (37-42 weeks) was based on an a priori decision.

The diagnosis of asthma in children younger than 3 years is challenging, as objective measurement of lung function in young children is difficult and not usually performed clinically. Including this group of children may, therefore, have affected our outcomes. Therefore, to verify whether our findings still hold, we performed a sensitivity analysis after excluding children younger than 3 years.

In all analyses, children who were not exposed to chorioamnionitis in utero served as the reference group. Because there are racial and ethnic disparities in the prevalence of childhood asthma,⁴ and because the incidence of chorioamnionitis among African American women is much higher than among their white counterparts,³³ we repeated the analysis after stratifying the data by maternal race/ethnicity to assess whether race/ethnicity modifies the magnitude of association. Hazard ratios (HR) with 95% confidence intervals (CI) were used to quantify the associations. All analyses were performed using SAS 9.1 for Windows (SAS Institute, Cary, North Carolina).

The median follow-up times for children with and without exposure in utero to chorioamnionitis were 2.0 and 2.5 years, respectively. There was a statistically significant difference in asthma incidence rate among preterm children (60.2 per 1000 person-years) and those carried to term (40.0 per 1000 person-years), regardless of exposure status to chorioamnionitis (incidence rate ratio [IRR], 1.5; 95% CI, 1.4-1.6). However, the incidence rates of asthma in preterm children and those carried to term for pregnancies complicated by chorioamnionitis were 100.7 per 1000 person-years and 39.6 per 1000 person-years, respectively (IRR, 2.9; 95% CI, 2.6-3.3).

We present the distribution of maternal and infants characteristics among singleton children aged 8 years or younger with and without the diagnosis of asthma in Table 1. Compared with women who had a child without a history of asthma, women who had a child with a history of asthma were more likely to be African American, aged 35 years and older, have 13 years or more of formal education, and to have smoked during pregnancy. A history of chorioamnionitis, antibiotic use, and physician-diagnosed asthma during the pregnancy were more likely to occur among women whose children develop asthma later in life. There was a significant difference in the rate of asthma by gestational age at delivery.

Table 2 shows the association between exposure to chorioamnionitis in utero and physician-diagnosed asthma in children younger than 8 years stratified by gestational age categories at preterm and full-term gestations. After adjusting for important confounding variables, at preterm gestation, exposure to chorioamnionitis in utero was found to be associated with increased risk of asthma (HR, 1.68; 95% CI, 1.52-1.87) when compared with no exposure to chorioamnionitis. Further stratifying the analysis by gestational age categories has revealed an independent relationship between chorioamnionitis and childhood asthma. Compared with gestational age category–specific children who were not exposed to chorioamnionitis in utero, the risks of asthma were 1.23-fold (95% CI, 1.02-1.49), 1.51-fold (95% CI, 1.26-1.80), and 1.20-fold (95% CI, 1.03-1.47) among extremely (23-28 weeks), very (28-33 weeks), and moderately (34-36 weeks) preterm children who were exposed to chorioamnionitis in utero, respectively. The same analysis was repeated by maternal race/ethnicity groups. Of those born at less than 37 weeks' gestation, non-Hispanic white (HR, 1.66; 95% CI, 1.32-2.07), non-Hispanic African American (HR, 1.98; 95% CI, 1.60-2.44) and Hispanic (HR, 1.70; 95% CI, 1.45-2.00), but not Asian/Pacific Islander (HR, 1.15; 95% CI, 0.83-1.58) children aged 8 years or younger who were exposed to chorioamnionitis in utero were associated with increased risk of asthma compared with children of similar gestational age born after pregnancies not complicated by chorioamnionitis. Furthermore, compared with children of similar gestational age born after pregnancies not complicated by chorioamnionitis, the risk of childhood asthma seem to be elevated among non-Hispanic white, non-Hispanic African American, and Hispanic children born at 23 to 28, 29 to 33, and 34 to 36 weeks of gestation after pregnancies complicated by chorioamnionitis; however, for most part, these risks failed to reach statistical significance (data not shown).

One of the mechanisms through which preterm birth is presumably associated with respiratory problems in early childhood is bronchopulmonary dysplasia. Therefore, to assess whether our findings persist, we repeated the analysis after adjusting for the variables in the model. However, adjusting for the variable did not affect our results. Furthermore, we performed a sensitivity analysis after excluding children whose follow-up time was less than 3 years to evaluate whether our finding persist. Findings were virtually unchanged through these exclusions (data not shown).

Compared with children who were born between 37 and 38 weeks' gestation and who were not exposed to chorioamnionitis in utero, the risks of asthma in children who were born in all preterm subcategories and exposed to chorioamnionitis in utero were significantly higher (Figure), with a significant inverse gestational age at exposure–response relationship apparent in the magnitude of the association.

In this large-scale cohort study, we showed that fetal exposure to chorioamnionitis combined with preterm delivery is associated with increased risk of physician-diagnosed asthma at 8 years of age or younger, and that the effects of chorioamnionitis on childhood asthma differ by race/ethnicity. Furthermore, we showed a significant inverse relationship between the gestational age at which the child was born and asthma, suggesting that early gestational exposure to chorioamnionitis may be much more detrimental to the developing lung of the fetus and/or may lead to heightened immune response following each subsequent antigen encounter and may predispose the child to asthma later in life. This effect could also be present if conservative management is planned to delay delivery in the presence of chorioamnionitis, leading to a higher bacterial load and fetal lung injury.

Although atopy plays an important role in the pathogenesis of asthma, fewer than half of childhood asthma cases are attributable to atopy.³⁴ Other potential risk factors implicated in the association with childhood asthma include preterm birth,^35-37 low birth weight,^38-41 and cesarean delivery.^42,43 Most previous studies that examined the associations between preterm birth and low birth weight and childhood asthma were in the setting of the delivery and postpartum period and, unfortunately, findings are often inconsistent regarding the direction and magnitude of associations. There are few studies, however, that specifically examine associations between the fetal environment and childhood respiratory conditions, including asthma.^44-47 Fetal exposure to cigarette smoke has been shown to be associated with increased risk of childhood wheezing and asthma.^44,45 Nicotine readily passes across the placenta and can directly affect fetal lung development.⁴⁸ In utero exposure to antibiotics, when much of the immune system develops, may also alter fetal microbial exposure and predispose the child to a higher risk of developing asthma and other atopic conditions later in life.⁴⁶

The fetal lung is in constant contact with amniotic fluid. Consequently, in a pregnancy complicated by chorioamnionitis, the developing airway is exposed to microorganisms, toxic substances, and inflammation mediators released by stimulated cells at the maternal-fetal interface. Studies based on animal models clearly showed that exposure to chorioamnionitis in utero increases the risk of inflammation, injury, apoptosis, and remodeling of the fetal lung.^25,26 Correspondingly, studies based on human subjects showed association between chorioamnionitis and fetal lung injury.^27-29,49 Elevated levels of IL-6, IL-8, IL-1β, and tumor necrosis factor α in the amniotic fluid of infants delivered preterm are associated with increased risk of bronchopulmonary dysplasia,^28,30 suggesting a fetal origin of childhood respiratory disease.

Kumar et al⁴⁷ showed an association between preterm birth and chorioamnionitis and increased risk of asthma. Our results and those of Kumar et al suggest that there is probably a shared effect of chorioamnionitis and preterm birth in the risk of asthma. However, because their study was limited by its small sample, they were not able to show the effect of chorioamnionitis on childhood asthma in moderately preterm infants. One important contribution of the current study is that it shows an independent association between chorioamnionitis and childhood asthma at preterm gestation. Furthermore, unlike our study, Kumar et al⁴⁷ only presented data on African American children, once again owing to small sample size. Our findings indicate that, compared with the preterm children of women who did not develop chorioamnionitis during a pregnancy, preterm children of non-Hispanic white, non-Hispanic African American, and Hispanic, but not Asian/Pacific Islander women whose pregnancy was complicated by chorioamnionitis were at significantly increased risk of asthma later in life.

There are some potential limitations that need to be considered when interpreting our findings. For this study, subjects were selected through stringent exclusion criteria that resulted in a lower rate of chorioamnionitis than other studies. In this study, the diagnosis of chorioamnionitis was based on clinical examination of subjects (ICD-9-CM codes), which is less likely to be influenced by variation owing to self-reporting. However, the clinical diagnosis of chorioamnionitis is likely to miss most cases of asymptomatic chorioamnionitis. This subsequently may affect our estimates. Furthermore, subjects could have a wide range of exposure to microbials and/or inflammation mediators. Because we did not obtain data on the level of exposure to microbials and/or inflammation mediators, we could not investigate whether the level of exposure to microbial burden and/or inflammation mediators modulated the observed association between chorioamnionitis and childhood asthma. Thus, the association could be even stronger in children exposed to high levels of microbial and/or inflammation products in utero and in those whose infants mounted an exuberant inflammatory response.

Making a definite diagnosis of asthma in children younger than 3 years is challenging because respiratory tract infections are common in this age group and their symptoms closely resemble those of asthma. Furthermore, objective measurement of lung function is difficult and not usually performed clinically in those younger than 3 years. However, in the current study, limiting the analysis to children older than 3 years did not change the overall and race/ethnicity-specific results. Although there is clear evidence that both maternal and paternal history of atopy and asthma⁵⁰ and maternal smoking during pregnancy^51,52 are associated with increased risk of childhood asthma, the findings are inconsistent regarding the effects of paternal smoking on childhood asthma.^53-55 Because the data set used in this study lacks information on parental atopy and smoking, we were unable to adjust for these potential confounders. In the medical records used for this study, information on maternal smoking during pregnancy was self-reported and severely underreported, which may have influenced our estimates. However, Buka et al⁵⁶ reported that there is significant agreement between self-reported smoking and serum levels of the nicotine metabolite cotinine. Although we adjusted for several potential confounding factors, the potential for residual confounding remains.

Our findings may provide clues to the role of a heretofore unrecognized component of the complex etiology of childhood asthma. This risk factor, intrauterine exposure to chorioamnionitis, is well supported by animal models,^25,26 biologically plausible, and readily amenable to a variety of preventive interventions. Further research seems warranted to confirm the findings. If confirmed, they could lead to a better understanding of the pathoetiologic mechanism of the disease and its prevention.

In conclusion, the findings of this study on the association between chorioamnionitis and childhood asthma have important implications. Chorioamnionitis and preterm gestation may result in increased risk of childhood asthma, presumably in response to inflammation-mediated fetal lung injury and/or heightened immune response against subsequent encounters with pathogens.

Correspondence: Darios Getahun, MD, MPH, Department of Research and Evaluation, Kaiser Permanente Southern California Medical Group, 100 Los Robles Ave, 2nd Floor, Pasadena, CA 91101 (darios.t.getahun@kp.org).

Accepted for Publication: July 9, 2009.

Author Contributions:Study concept and design: Getahun. Acquisition of data: Getahun. Analysis and interpretation of data: Getahun, Strickland, Zeiger, Fassett, Rhoads, Chen, and Jacobsen. Drafting of the manuscript: Getahun. Critical revision of the manuscript for important intellectual content: Getahun, Strickland, Zeiger, Fassett, Chen, Rhoads, and Jacobsen. Statistical analysis: Getahun and Chen. Administrative, technical, and material support: Strickland and Zeiger. Study supervision: Getahun, Strickland, Rhoads, and Jacobsen.

Financial Disclosure: None reported.

Funding/Support: This study was supported by Kaiser Permanente Direct Community Benefit funds.

Disclaimer: The funding body did not influence the design or conduct of the study; analysis or interpretation of the data; or the preparation, review, or approval of the manuscript.