 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
,
,||
* Department of Ambulatory Care and Prevention, Harvard Medical School and Harvard Pilgrim Health Care, Boston, Massachusetts
Channing Laboratory, Department of Medicine, Brigham and Women’s Hospital
Department of Emergency Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts
|| Departments of Environmental Epidemiology
¶ Nutrition, Harvard School of Public Health, Boston, Massachusetts
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Design. We report findings from 1005 mothers and their infants participating in Project Viva, a US cohort study of pregnant women and their offspring. The main outcome measure was maternal report of a provider’s diagnosis of eczema or atopic dermatitis in the first 6 months of life. We used multiple logistic regression models to assess the associations between several simultaneous predictors and incidence of atopic dermatitis.
Results. Cumulative incidence of atopic dermatitis in the first 6 months of life was 17.1%. Compared with infants born to white mothers, the adjusted odds ratio (OR) for risk of atopic dermatitis among infants born to black mothers was 2.41 (95% confidence interval [CI]: 1.47, 3.94) and was 2.58 among infants born to Asian mothers (95% CI: 1.27, 5.24). Male infants had an OR of 1.76 (95% CI: 1.24, 2.51). Increased gestational age at birth was a predictor (OR: 1.14; 95% CI: 1.02, 1.27, for each 1-week increment), but birth weight for gestational age was not. Infants born to mothers with a history of eczema had an OR of 2.67 (95% CI: 1.74, 4.10); paternal history of eczema also was predictive, although maternal atopic history was more predictive than paternal history. Several other perinatal, social, feeding, and environmental variables were not related to risk of atopic dermatitis.
Conclusions. Black and Asian race/ethnicity, male gender, higher gestational age at birth, and family history of atopy, particularly maternal history of eczema, were associated with increased risk of atopic dermatitis in the first 6 months of life. These findings suggest that genetic and pre- and perinatal influences are important in the early presentation of this condition.
Key Words: atopic dermatitis • eczema • perinatal • infancy • gestational age • race/ethnicity • gender
Abbreviations: BMI, body mass index • OR, odds ratio • CI, confidence interval • IgE, immunoglobulin E
Atopic dermatitis is a chronic inflammatory skin disease that occurs with a peak onset in infancy and a large majority of cases presenting in the first few years of life. Its incidence in the developed world has increased dramatically over the past several decades.1 Atopic dermatitis in the first few months of life causes significant family stress, interference with infant sleep and feeding, physician visits, and health care expenditures. It also may be a risk factor for aeroallergen sensitization, asthma, allergic rhinitis, and urticaria later in childhood.2,3 Risk for atopy may begin very early in life, even before birth. Both genetic and environmental factors may predispose to allergic disease. In utero exposures may be modulated by postnatal factors to predispose the developing immune system to atopy.
However, there are few studies of the hereditary and environmental predictors of the early-life presentation of atopic dermatitis. Those that do exist have been hampered by a limited range of exposures,4 failure to control for potential confounders,5 retrospective data collection,6 or small sample sizes.7 Few have assessed early disease in US populations. The purpose of this study was to investigate pre- and perinatal predictors of atopic dermatitis in the first 6 months of life among members of Project Viva, a US cohort study of pregnant women and their children.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Exclusion criteria include multiple gestation (twins, triplets, etc), inability to answer questions in English, plan to move out of the area before delivery, and gestational age >22 completed weeks at initial prenatal clinical appointment. We enrolled 2611 pregnant women (63% of those eligible) between April 22, 1999 and March 1, 2002, of whom 318 subsequently became ineligible because of multiple gestation (n = 18), transferring obstetric care to a nonstudy site (n = 111), or they were no longer pregnant (n = 189). Of the 2293 remaining participants, 191 (8%) withdrew and 18 (<1%) were lost to follow-up before delivery. Of the 2084 women still enrolled as of March 1, 2002, 332 had not delivered yet, leaving 1752 who delivered a live infant. Of these, 407 had not yet reached the 6-month evaluation, and 52 participants disenrolled before reaching the 6-month evaluation, leaving 1293 women. The 6-month interview was completed by 1056 (82%) of these women. For this analysis, we excluded 51 participants with missing data on the covariates included in our model, resulting in a sample of 1005 mother-infant pairs for this analysis. Human subjects committees of Harvard Pilgrim Health Care, Brigham and Women’s Hospital, and Beth Israel Deaconess Medical Center approved the study protocols.
Measurements
We obtain data for Project Viva from multiple sources. We calculated gestational age by using the mother’s reported last menstrual period or ultrasound dating if the 2 measures differed by >10 days. At the first visit, we also ask about maternal age, country of birth, number of previous pregnancies, prepregnancy weight, height, history of atopic disorders (asthma, eczema, or allergic rhinitis), maternal education, and household income. We calculated maternal body mass index (BMI) as weight in kilograms divided by the square of height in meters. We took maternal race/ethnicity from maternal responses to the question: "Which of the following best describes your race or ethnicity? You may choose more than one." Response choices were Hispanic or Latina, white or Caucasian, black or African American, Asian or Pacific Islander, American Indian or Alaskan Native, or other (to be specified).
From the hospital vital statistics record we abstracted infant birth weight. We determined birth weight percentiles using as a reference the combined 1999–2000 US Natality data sets,8 which include birth weights at each completed week of gestation for all infants born in the United States. We then assigned a z value for each percentile to have a normally distributed measure of birth weight adjusted for gestational age.
At the 6-month visit we ask: "Have you ever been told by a health care professional, such as a doctor, physician assistant, or nurse practitioner, that your infant has eczema (atopic dermatitis)?" We also ask about infant feeding and household exposures at this visit. Given the broad scope of Project Viva and the extensive range of maternal and infant variables and outcomes assessed, participants were blind to the objectives of our particular analysis.
For those infants in our cohort for whom 1-year follow-up data were already available (n = 551), we assessed the degree of concordance between reported diagnoses of atopic dermatitis at 6 months and 1 year. We ask mothers at the 1-year visit whether they have ever been told by a health care professional that their infant had eczema (atopic dermatitis). We found that 79 of 129 (61%) infants who had been diagnosed with atopic dermatitis by 1 year of age had received that diagnosis within the first 6 months of life, and that 79 of 97 (81%) diagnosed within the first 6 months retained a maternal report of that diagnosis at 1 year.
Data Analysis
Our main outcome was maternal report of a provider’s diagnosis of atopic dermatitis in the first 6 months of life. To assess the associations between several simultaneous predictors and atopic dermatitis in the first 6 months of life, we used logistic regression models. We assessed the confounding effect of covariates by examining the association of predictors with atopic dermatitis before and after adding the covariates to the model. If the log odds ratio (OR) associated with a predictor changed by >10%, we concluded that the covariate was a confounder. From the models, we present ORs and 95% confidence intervals (CIs). We kept in our final model covariates that independently predicted outcome, acted as confounders of other predictors, or for which we had strong a priori hypotheses of association with atopic dermatitis. Using stratified analyses, we also studied the data for evidence of effect modification by each of the variables included in our model.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Cumulative incidence of a maternal report of provider-diagnosed atopic dermatitis in the study infants was 172 of 1005 (17.1%) in this cohort (Table 1). Infants with atopic dermatitis were born at later gestational ages and were heavier at birth than infants without atopic dermatitis. Atopic dermatitis was more common in male than in female infants. Infants of black and Asian mothers were nearly twice as likely to have received a diagnosis of atopic dermatitis as those born to white or Hispanic mothers. The diagnosis was more than twice as likely in infants born to mothers with a history of eczema as in those born to mothers without this history.
|
2.5 times that of white mothers (Table 2). For infants born to Hispanic mothers, no increased risk was apparent. Figure 1 shows the multivariate-adjusted ORs for atopic dermatitis by maternal race/ethnicity. Male gender was a predictor of atopic dermatitis; the odds of developing atopic dermatitis among male infants was 1.8 times that of female infants.
|
|
Family history of atopy was associated with an increased risk of atopic dermatitis. In adjusted analyses, maternal history of atopy conferred greater risk than paternal history. The strongest parental variable seemed to be maternal history of eczema, for which the adjusted OR was 2.67 (95% CI: 1.74, 4.10). Maternal history of asthma, hay fever, or any atopy also predicted increased risk of atopic dermatitis (Table 2). Paternal eczema was a predictor of infantile atopic dermatitis, but other paternal atopic variables were not. The reported presence of cockroaches within the home also appeared to predict increased risk of atopic dermatitis in the first 6 months of life, with an adjusted OR of 1.86 (95% CI: 0.87, 3.96), but the wide CI precluded a strong inference.
After simultaneous control for all the other variables depicted in Table 2, maternal age, prepregnancy BMI, and education were not associated with atopic dermatitis. Adjustment for infant feeding type at 3 and 6 months, number of previous pregnancies, maternal country of birth, household income, furry pets at home, and household mold exposure did not materially alter any of the risk estimates, and none of these variables were associated with atopic dermatitis in multivariate models (data not shown). Additionally, season of birth did not confound any of the associations we found. We did not include it in our model because, at 6 months of age, each of our subjects had experienced only half of the year. In stratified analyses, we found no evidence of effect modification of any of the observed associations by the variables included in our model.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Race/Ethnicity
The increased risk of atopic dermatitis for blacks and Asians persisted after control for multiple maternal, socioeconomic, and family variables. Our findings are consistent with several previous reports of higher rates of atopic dermatitis in blacks and Asian/Pacific Islanders in the United States and abroad. Increased physician visits for atopic dermatitis by patients from these racial/ethnic groups in the United States has been reported recently.9 Williams et al10 reported an increased risk of atopic dermatitis in black Caribbean children born in the United Kingdom versus white children, as assessed through multiple measures, including a cross-sectional prevalence survey, dermatologist’s examination, and parental report. Golding and Peters6 also found children of West Indian origin to be at increased risk for eczema, as assessed by parental report of eczema in the first 5 years of life, compared with white infants. Increased referrals for atopic dermatitis among Asians compared with non-Asians in the United Kingdom have been reported,11 as has an increased prevalence of atopic dermatitis in Chinese versus white infants in Hawaii12 and Australia.13
Migration has been suggested as a risk factor for atopic dermatitis,12 perhaps because of exposure to new or increased concentrations of allergens on migration.14 However, in our analysis, racial/ethnic differences did not change materially after control for maternal country of birth (US-born versus not US-born), so maternal migration did not seem to explain the racial/ethnic differences in infant atopic dermatitis risk we observed in our study (data not shown). These differences also persisted after control for infant feeding and multiple socioeconomic variables. Biological differences such as differences in staphylococcal colonization,15 increased antigen sensitivity, increased susceptibility to cutaneous irritation and skin barrier dysfunction16 and differences in immunoglobulin E (IgE) levels resulting from selection for defense against helminths,17 although not assessed in our study, may also explain racial/ethnic differences in rates of atopic dermatitis.
Differential access to medical care18 or cultural norms regarding disease reporting5 may play a role in the differences we observed. Changes of flexural dermatitis and postinflammatory pigmentary changes that are more visible and persist longer in darker skin may lead to more-frequent diagnosis.9 Milder cases may be overrepresented,9 or pediatricians may have greater difficulty diagnosing or managing unclassified dermatoses in patients with darker skin, leading to a label of atopic dermatitis. This reason may explain the higher risk we observed in black infants but would be harder to invoke for higher risk among Asians or the lack of risk among Hispanic infants, although our null finding for Hispanics might be due to small numbers. Studies are needed that further isolate particular environmental, genetic, social, or cultural exposures associated with self-identified race/ethnicity.19
Infant Gender
In our cohort, the adjusted OR for atopic dermatitis in the first 6 months of life was 1.76 for male versus female infants. Many previous international studies of atopic dermatitis have found more affected girls than boys,5,20 with a 1.3:1 girl/boy ratio frequently cited. Most studies, however, have examined point prevalence later in childhood or have assessed cumulative disease incidence over a longer period of follow-up. A Swedish community birth cohort study21 found that, at the age of 2 years, more boys than girls had ongoing atopic dermatitis. Higher IgE levels have been reported in male versus female infants at birth22 and at 6 months and 6 years of age.23 Our data may parallel reports that wheezing and asthma are more common in boys than girls during infancy and childhood but that this pattern is reversed at puberty,24 although some feel that the early differences in wheeze relate to a mechanical rather than an immune male disadvantage. Prenatal hormonal exposures may program the fetal immune system,25,26 but the biological or cultural factors involved in shifting gender differences in atopic disease are not fully known.27
Birth Weight and Gestational Age
Prior studies have suggested that increasing gestational age at birth may predict increased risk of atopic dermatitis,7,28,29 but few analyses have examined both birth weight and length of gestation as predictors of the disease in infancy. A Danish cohort study28 of children up to the age of 7 years found that those born after term had a significantly increased risk of atopic dermatitis, as did children with birth weights that were high for gender and gestational age. A low prevalence of atopic dermatitis was found in a hospital-based follow-up cohort study of very low birth weight infants,30 although this analysis did not control for gestational age. A Japanese study31 found a low prevalence of atopic eczema in low birth weight infants at 18 months of age, but the overall prevalence of atopic eczema in the normal birth weight group was only 2.3%.
In our study we observed an increased risk of atopic dermatitis with increased gestational age at birth. We did not find that increasing birth weight for gestational age (z value) was a predictor. Our data thus suggest that the increased risk associated with increased birth weight may be ascribed more precisely to the effects of increasing length of gestation than to increased fetal growth.
These results raise the question of why increasing length of gestation may confer an increased risk of atopic dermatitis. Fetal insults such as transplacental allergen exposure or maternal nutritional factors during critical periods of late gestation could predispose toward fetal atopy; early postnatal antigen exposure in infants born at earlier gestational ages could also preferentially result in the development of tolerance instead of sensitization.32 Higher birth weight and postmaturity have been associated with increased serum total IgE, and postmaturity may be associated with a reduction in thymic weight that may alter the balance of T helper (Th) 1 and Th2 cell populations in the thymus in favor of Th2 cells.33 A longer period of exposure to the Th2 cytokines of pregnancy may also bias the fetal immune system toward atopy.34 Examination of placental material, cord blood IgE, and lymphocyte proliferation in future studies may elucidate the biological explanations for our findings.
Family History of Atopy
Our study found an increased risk of atopic dermatitis in infants born to parents with a history of atopy, with maternal history more predictive than paternal history. Risk was highest for infants born to mothers with a history of eczema. Paternal eczema was a predictor of infantile atopic dermatitis, but other paternal variables were not. These data are consistent with previous reports that children are likely to develop atopic diseases similar to those of their parents and that maternal history predicts greater risk for childhood eczema than does paternal history.35–37
Atopy may be inherited preferentially through the maternal line, or mothers may carry relatively more of the predisposing genes.38 Transplacental transfer of antigens, maternal antibodies, and maternally-derived cytokines may shape early atopy,39 and intense sharing of postnatal environmental factors through breastfeeding and a shared home environment may also play a role.40 In our study, the risk estimate associated with a maternal history of eczema was not altered when controlled for early infant feeding practices.
A history of atopy in a first-degree relative is 1 of the diagnostic criteria for atopic dermatitis in infancy;41 thus, it is likely that some infants with other dermatoses will be misclassified as having eczema based on family history. In assessing atopic dermatitis during infancy we may also be preferentially including children of atopic mothers, who have been shown to present earlier.35 In addition, because the mothers reported for the fathers, our assessment of paternal history may contain misclassification. Litonjua et al36 found that mothers were more likely to classify fathers correctly if a child had eczema than if none of their children had eczema.
Other Factors
In our cohort, household presence of cockroaches was associated with an adjusted OR of 1.86 (95% CI: 0.87, 3.96) for atopic dermatitis in the first 6 months of life. The association between sensitization to cockroach antigen and asthma is well documented,42,43 whereas associations with atopic dermatitis are suggested.44,45 Our data also suggest increased risk, but more studies are needed with increased exposure prevalence and objective measures of exposure.
Season of birth has been associated with risk of atopic dermatitis, with greater risk in those children born in the fall than those born in the spring,46 and exacerbations of atopic dermatitis during dry winter months are common. However, we did not include birth season in this analysis, because we assessed our outcome at only 6 months of age, before every infant had experienced each season.
Strengths and Limitations
We assessed atopic dermatitis in the first 6 months of life, a period during which the condition causes considerable morbidity, in terms of sleep and feeding disruption, parental stress, economic costs, and later atopic outcomes. These children with early atopic dermatitis may also represent a severe or long-lasting subgroup and thus are important to study. However, predictors of later-onset atopic dermatitis may be different, and we may have included some cases of dermatoses that are eventually reclassified as nonatopic.
Although 288 of the 1293 enrolled mother-infant pairs that had reached the 6-month evaluation did not complete the 6-month interview or were missing covariate data, the eligible subjects differed only slightly from the 1005 participants in several salient characteristics. Although our observed associations are likely valid for this study population, the predominantly urban/suburban population and relatively high socioeconomic position of our participants could limit generalizability. Our study is also limited by the use of maternal report of provider-diagnosed atopic dermatitis as the outcome variable. Maternal report of physicians’ diagnoses of eczema, however, has been previously validated versus clinical examination in the United States47 and versus chart-recorded diagnosis in the United Kingdom.48
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
We thank the participants and staff of Project Viva. We also thank Emily Oken, MD, for insight into birth weight and gestational age analyses; Stevan DeProsse for technical expertise; Elizabeth Baldwin for management of the Project Viva database; and Rachel Fournier for administrative and logistic support.
|
FOOTNOTES |
|---|
Address correspondence to Matthew W. Gillman, MD, SM, Department of Ambulatory Care and Prevention, Harvard Medical School and Harvard Pilgrim Health Care, 133 Brookline Ave, 6th Fl, Boston, MA 02215. E-mail: matthew_gillman{at}hms.harvard.edu
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. Sugiyama, H. Arakawa, K. Ozawa, T. Mizuno, H. Mochizuki, K. Tokuyama, and A. Morikawa Early-Life Risk Factors for Occurrence of Atopic Dermatitis During the First Year Pediatrics, March 1, 2007; 119(3): e716 - e723. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E K Chung, R L Miller, M T Wilson, S J McGeady, and J F Culhane Antenatal risk factors, cytokines and the development of atopic disease in early childhood Arch. Dis. Child. Fetal Neonatal Ed., January 1, 2007; 92(1): F68 - F73. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G Moro, S Arslanoglu, B Stahl, J Jelinek, U Wahn, and G Boehm A mixture of prebiotic oligosaccharides reduces the incidence of atopic dermatitis during the first six months of age Arch. Dis. Child., October 1, 2006; 91(10): 814 - 819. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Kilchevsky, E. James, J. Fong, and P. Murphy Atopic Dermatitis and Breastfeeding Pediatrics, October 1, 2004; 114(4): 1129 - 1129. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2004 American Academy of Pediatrics. All rights reserved. |
||
|
||
