Atopic disease and cardiovascular risk factors in US children

Abstract

Atopic diseases, including asthma, hay fever, and eczema, are common chronic inflammatory disorders of childhood. They are associated with chronic peripheral and organ-specific inflammation, including upregulated TH2 inflammation.1Broide D.H. Molecular and cellular mechanisms of allergic disease.J Allergy Clin Immunol. 2001; 108: S65-S71Abstract Full Text Full Text PDF PubMed Scopus (203) Google Scholar Aside from chronic inflammation, these disorders may be anxiety-provoking and have been found to be associated with chronic sleep disturbances, functional limitations of physical activity, and use of various corticosteroids and other immunosuppressive medications. In turn, each of these sequela was previously found to be associated with increased risk for cardiovascular disease. Indeed, a recent study demonstrated increased cardiovascular risk in 2 studies of US adults with eczema.2Silverberg J.I. Greenland P. Eczema and cardiovascular risk factors in 2 US adult population studies.J Allergy Clin Immunol. 2015; 135: 721-728.e6Abstract Full Text Full Text PDF PubMed Scopus (172) Google Scholar However, few studies have examined the impact of atopic disease on cardiovascular risk in children. I used the 2012 National Health Interview Survey (NHIS), an in-person household survey administered in English and Spanish. One child per household was randomly selected for the child questionnaire. Using data from the US Bureau of the Census, sampling weights were adjusted for age, sex, race, ethnicity, household size, and educational attainment of the most educated household member. The questions used to assess for history of atopic and cardiovascular disease are presented in Table E1 in this article's Online Repository at www.jacionline.org. This study was approved by the institutional review board at Northwestern University. All data processing and statistical analyses were performed with SAS, version 9.4, software (SAS institute, Cary, NC). Bivariable survey logistic regression models were used to determine the associations of asthma, hay fever, and eczema (independent variables) with obesity, hypertension, hyperlipidemia, and diabetes (dependent variables). Multivariate models included age (continuous), sex (male/female), race/ethnicity (white, black, Hispanic, multiracial/other), household income (0-0.9, 1-2.9, 3-3.9, and ≥4.0 poverty index ratio), highest level of household education (<high school, high school or GED, >high school), birthplace in the United States, and insurance coverage (yes/no). Adjusted odds ratios and 95% CIs were estimated. Two- and 3-way statistical interactions between covariates were tested and included in the above models if P was less than .01, cell frequencies were more than 5 for each level of interaction, and estimates were modified by more than 20%. Complete case analysis was performed; that is, missing values were excluded. A 2-sided P value of less than .05 was taken to indicate statistical significance. However, the multiple dependent tests performed increase the risk of falsely rejecting the null hypothesis. A total of 13,275 children aged 0 to 17 years were included in the 2012 NHIS. The US prevalences (95% CI) of asthma, eczema, and hay fever were estimated to be 14.0% (13.3% to 14.8%), 12.0% (11.3% to 12.7%), and 16.6% (15.7% to 17.4%), respectively. In multivariable models, pediatric asthma and hay fever were associated with higher odds of overweight and obesity, hypertension, and hyperlipidemia, but not diabetes (Table I). However, eczema was associated with higher odds of overweight and obesity, but not hypertension, hyperlipidemia, or diabetes. There were no significant interactions with other covariates in any of the above models.Table IAssociations between pediatric asthma and cardiovascular risk factorsVariableAsthmaHay feverEczemaNo(n = 11,340)Yes(n = 1,923)No(n = 11,088)Yes(n = 2,161)No(n = 11,658)Yes(n = 1,603)% Prev(95% CI)% Prev(95% CI)aOR(95% CI)P value% Prev(95% CI)% Prev(95% CI)aOR(95% CI)P value% Prev(95% CI)% Prev(95% CI)aOR(95% CI)P valueBMI-for-age and sex (%)∗Questions for height and weight were asked for ages 12-17 years. % Prevalences for BMI-classifications are presented for children ages 12-17 years. <54.9 (4.5-5.3)4.4 (3.5-5.4)0.99 (0.77-1.27).924.8 (4.4-5.2)5.0 (4.1-5.9)1.08 (0.86-1.36).524.9 (4.5-5.3)4.1 (2.8-5.3)0.72 (0.51-1.00).05 5-8466.6 (65.7-67.5)59.0 (57.1-61.0)1.00 [ref]—65.9 (65.0-66.8)62.8 (61.0-64.7)1.00 [ref]—65.8 (65.0-66.7)60.2 (57.4-62.9)1.00 [ref]— 85-9415.8 (15.1-16.5)17.7 (16.1-19.2)1.19 (1.05-1.35).00615.7 (15.0-16.4)18.0 (16.4-19.5)1.24 (1.09-1.41).000816.0 (15.3-16.6)17.9 (15.8-20.0)1.19 (1.00-1.41).045 ≥9512.7 (12.0-13.3)18.8 (17.3-20.4)1.50 (1.33-1.69)<.000113.6 (13.0-14.3)14.2 (13.0-15.5)1.22 (1.08-1.39).00213.3 (12.7-13.9)17.9 (15.7-20.0)1.46 (1.24-1.69)<.0001 Hypertension†Questions for hypertension, hypercholesterolemia, and diabetes were asked for ages 6-17 years. % Prevalences for these outcomes are presented for ages 6-17 years.0.8 (0.5-1.1)2.1 (1.2-3.0)1.93 (1.04-3.58).040.8 (0.5-1.1)1.9 (1.0-2.9)2.61 (1.31-5.20).0061.0 (0.7-1.3)1.1 (0.4-1.9)1.12 (0.53-2.34).77 Hypercholesterolemia†Questions for hypertension, hypercholesterolemia, and diabetes were asked for ages 6-17 years. % Prevalences for these outcomes are presented for ages 6-17 years.0.8 (0.6-1.1)1.7 (0.9-2.6)2.02 (1.09-3.74).030.8 (0.6-1.0)1.7 (0.9-2.4)2.50 (1.42-4.40).0020.9 (0.6-1.1)1.8 (0.8-2.7)1.72 (0.83-3.56).14 Diabetes†Questions for hypertension, hypercholesterolemia, and diabetes were asked for ages 6-17 years. % Prevalences for these outcomes are presented for ages 6-17 years.0.2 (0.1-0.3)0.3 (0.0-0.7)2.06 (0.59-7.16).260.2 (0.1-0.3)0.3 (0.0-0.6)0.93 (0.20-4.31).930.2 (0.1-0.3)0.1 (0.0-0.2)0.42 (0.09-1.99).28Multivariable survey weighted logistic regression models were constructed with caregiver-reported history of hypertension, hypercholesterolemia, and diabetes as dependent variables and history of asthma, hay fever, and eczema as independent variables. Adjusted odds ratios (aORs) and 95% CI were estimated in multivariable models, including age, sex, highest level of education in the household, size of family, race/ethnicity, birthplace in the United States, and insurance status. Refusal to answer a particular question or response of “don't know” occurred for the questions pertaining to eczema in 14 (0.1%), asthma in 12 (0.08%), hay fever in 26 (0.2%), BMI-for-age and sex percentile in 11 (0.2%), history of hypertension in 3 (0.03%), high cholesterol in 2 (0.02%), diabetes in 9 (0.06%), sex in 0 (0.0%), race in 0 (0.0%), Hispanic/Spanish origin in 0 (0.0%), and highest level of education in the household in 69 (0.2%), respectively. Boldface indicates statistically significant estimates.BMI, Body mass index.∗ Questions for height and weight were asked for ages 12-17 years. % Prevalences for BMI-classifications are presented for children ages 12-17 years.† Questions for hypertension, hypercholesterolemia, and diabetes were asked for ages 6-17 years. % Prevalences for these outcomes are presented for ages 6-17 years. Open table in a new tab Multivariable survey weighted logistic regression models were constructed with caregiver-reported history of hypertension, hypercholesterolemia, and diabetes as dependent variables and history of asthma, hay fever, and eczema as independent variables. Adjusted odds ratios (aORs) and 95% CI were estimated in multivariable models, including age, sex, highest level of education in the household, size of family, race/ethnicity, birthplace in the United States, and insurance status. Refusal to answer a particular question or response of “don't know” occurred for the questions pertaining to eczema in 14 (0.1%), asthma in 12 (0.08%), hay fever in 26 (0.2%), BMI-for-age and sex percentile in 11 (0.2%), history of hypertension in 3 (0.03%), high cholesterol in 2 (0.02%), diabetes in 9 (0.06%), sex in 0 (0.0%), race in 0 (0.0%), Hispanic/Spanish origin in 0 (0.0%), and highest level of education in the household in 69 (0.2%), respectively. Boldface indicates statistically significant estimates. BMI, Body mass index. The absolute risk differences for cardiovascular disease risk factors between atopic and nonatopic children were fairly small because of the low prevalence of cardiovascular risk factors. For example, asthma and hay fever were associated with only a 1.28% (0.32% to 2.24%) and 1.14% (0.17% to 2.11%) increased risk of hypertension and 0.91% (0.06% to 1.77%) and 0.88% (0.09% to 1.66%) risk of hypercholesterolemia, respectively. Previous studies demonstrated associations between obesity, hypertension, hyperlipidemia, and/or metabolic syndrome in clinical cohorts of children with asthma.3Granell R. Henderson A.J. Evans D.M. Smith G.D. Ness A.R. Lewis S. et al.Effects of BMI, fat mass, and lean mass on asthma in childhood: a Mendelian randomization study.PLoS Med. 2014; 11: e1001669Crossref PubMed Scopus (69) Google Scholar, 4Ross K.R. Hart M.A. Storfer-Isser A. Kibler A.M. Johnson N.L. Rosen C.L. et al.Obesity and obesity related co-morbidities in a referral population of children with asthma.Pediatr Pulmonol. 2009; 44: 877-884Crossref PubMed Scopus (36) Google Scholar, 5Bibi H. Shoseyov D. Feigenbaum D. Genis M. Friger M. Peled R. et al.The relationship between asthma and obesity in children: is it real or a case of over diagnosis?.J Asthma. 2004; 41: 403-410Crossref PubMed Scopus (106) Google Scholar The present study builds on these studies and demonstrates an indirect public health burden of allergic disease from increased cardiovascular risk in US children. Recently, we demonstrated increased obesity, hypertension, hyperlipidemia, and adult-onset diabetes in US adults with eczema, due in part to increased smoking, alcohol consumption, and sedentary lifestyle.2Silverberg J.I. Greenland P. Eczema and cardiovascular risk factors in 2 US adult population studies.J Allergy Clin Immunol. 2015; 135: 721-728.e6Abstract Full Text Full Text PDF PubMed Scopus (172) Google Scholar In the present study, I could not examine these health behaviors because they were not asked in the NHIS child questionnaire. Smoking and alcohol are less likely to be major contributing factors in younger children/preadolescents, though decreased physical activity might be a contributing factor. Children often avoid intense physical activities, which may aggravate their airway disease and/or cutaneous itch. Moreover, chronic inflammation occurring in eczema and other allergic diseases may directly contribute toward the increased cardiovascular risk. Future studies are needed to determine the mechanism(s) of association between pediatric allergic and cardiovascular disease. Moreover, clinical studies are needed to determine whether early and aggressive treatment of allergic disease is able to mitigate increased cardiovascular risk. The lack of association between eczema and hypertension is in contrast to a recent multicenter pediatric dermatology case-control study of 132 children with moderate-severe atopic dermatitis and 143 age-matched healthy controls, which found higher overall systolic and diastolic blood pressures and increased odds of systolic blood pressure (>90%) in moderate-severe atopic disease.6Silverberg J.I. Becker L. Kwasny M. Menter A. Cordoro K.M. Paller A.S. Central obesity and high blood pressure in pediatric patients with atopic dermatitis.JAMA Dermatol. 2015; 151: 144-152Crossref PubMed Scopus (72) Google Scholar In the present study, however, I was unable to assess the severity of eczema or allergic disease. The lack of association between atopic disease and diabetes may be related to reduced power owing to the relative rarity of childhood diabetes and/or heterogeneity from type 1 and 2 diabetes. Previous studies found inverse associations between childhood eczema and type 1 diabetes.7Thomsen S.F. Duffy D.L. Kyvik K.O. Skytthe A. Backer V. Relationship between type 1 diabetes and atopic diseases in a twin population.Allergy. 2011; 66: 645-647Crossref PubMed Scopus (21) Google Scholar In future, even larger studies are needed to determine whether childhood eczema is associated with type 2 diabetes. The strengths of this study include prospective data collection, US population–based, large-scale diverse sample, complex sample weighting that allows for generalization of results to the entire US population, and controlling for multiple confounding variables in multivariable models. However, this study has limitations. All exposures and outcomes in the study were assessed by caregiver report and not verified by physical examination and may be subject to misclassification. However, a recent multicenter validation study of the eczema question used in the NHIS found good sensitivity, specificity, and positive and negative predictive values.8Silverberg J.I. Patel N. Immaneni S. Rusniak B. Silverberg N.B. Debashis R. et al.Assessment of atopic dermatitis using self- and caregiver-report: a multicenter validation study.Br J Dermatol. July 17, 2015; ([published online ahead of print])https://doi.org/10.1111/bjd.14031Crossref Scopus (85) Google Scholar Moreover, self-report of asthma has been previously validated.9Senthilselvan A. Dosman J.A. Chen Y. Relationship between pulmonary test variables and asthma and wheezing: a validation of self-report of asthma.J Asthma. 1993; 30: 185-193Crossref PubMed Scopus (31) Google Scholar Thus, I believe that the case definitions for allergic disease are sufficiently valid for epidemiological study. Nevertheless, confirmation of these findings using objective measures is warranted. In conclusion, pediatric allergic disease was found to be associated with increased odds of obesity, hypertension, and hyperlipidemia. Future studies using clinical examination and objective measures of adiposity and metabolic syndrome are needed to confirm these associations. Table E1Questions used in this studyVariableQuestionEczemaDuring the past 12 mo, have you been told by a doctor or other health professional that (child) had eczema or any kind of skin allergy?Ever asthma historyHave you ever been told by a doctor or other health professional that (child) had asthma?Hay feverDuring the past 12 mo, have you been told by a doctor or other health professional that (child) had hay fever?Height∗Questions were asked for ages 12-17 years.How tall is (child) without shoes?Weight∗Questions were asked for ages 12-17 years.How much does (child) weigh without shoes?High blood pressure†Questions were asked for ages 6-17 years.Have you ever been told by a doctor or other health professional that (child) had hypertension, also called high blood pressure?Diabetes†Questions were asked for ages 6-17 years.Has a doctor or other health professional ever told you that (child) had diabetes?High cholesterol†Questions were asked for ages 6-17 years.Have you ever been told by a doctor or other health professional that (child) had high cholesterol?∗ Questions were asked for ages 12-17 years.† Questions were asked for ages 6-17 years. Open table in a new tab