Sensitization to Ascaris lumbricoides and severity of childhood asthma in Costa Rica

Transcription

1 Sensitization to Ascaris lumbricoides and severity of childhood asthma in Costa Rica Gary M. Hunninghake, MD, a,b,c Manuel E. Soto-Quiros, MD, PhD, d Lydiana Avila, MD, d Ngoc P. Ly, MD, MPH, a,c,e Catherine Liang, MPH, a Jody S. Sylvia, a Barbara J. Klanderman, PhD, a,c Edwin K. Silverman, MD, PhD, a,b,c and Juan C. Celedón, MD, DrPH a,b,c Boston, Mass, and San José, Costa Rica Background: Little is known about sensitization (defined as a positive IgE) to helminths and disease severity in patients with asthma. Objectives: To examine the relationship between sensitization (defined as a positive IgE) to Ascaris lumbricoides and measures of asthma morbidity and severity in a Costa Rican population with low prevalence of parasitic infection but high prevalence of parasitic exposure. Methods: Cross-sectional study of 439 children (ages 6 to 14 years) with asthma. Linear regression and logistic regression were used for the multivariate statistical analysis. Results: After adjustment for parental education and other covariates, sensitization to Ascaris lumbricoides was associated with having at least 1 positive skin test to allergens (odds ratio, 5.15; 95% CI, ; P <.001), increased total serum IgE and eosinophils in peripheral blood, reductions in FEV 1 and FEV 1 /forced vital capacity, increased airway responsiveness and bronchodilator responsiveness, and hospitalizations for asthma in the previous year (odds ratio, 3.08; 95% CI, ; P 5.02). Conclusion: Sensitization to Ascaris lumbricoides is associated with increased severity and morbidity of asthma among children in Costa Rica. This association is likely mediated by an increased degree of atopy among children with asthma who are sensitized to Ascaris. Clinical implications: In areas with a low prevalence of helminthiasis such as Costa Rica, Ascaris sensitization may be an important marker of severe atopy and disease morbidity in children with asthma. (J Allergy Clin Immunol 2007;119: ) Key words: Ascaris, helminth, atopy, exacerbation, severity, asthma, Costa Rica From a the Channing Laboratory and b the Division of Pulmonary and Critical Care Medicine, Brigham and Women s Hospital, Boston; c Harvard Medical School, Boston; d the Division of Pediatric Pulmonology, Hospital Nacional de Niños, San José, Costa Rica; and e the Pediatric Pulmonary Division, Massachusetts General Hospital. Supported by grants HL04370 and HL66289 from the National Institutes of Health. Disclosure of potential conflict of interest: E. K. Silverman has consulting arrangements with GlaxoSmithKline, has received grant support from GlaxoSmithKline, has received honoraria from Bayer, and received money from Wyeth for a talk on chronic obstructive pulmonary disease genetics. The rest of the authors have declared that they have no conflict of interest. Received for publication June 5, 2006; revised November 30, 2006; accepted for publication December 4, Reprint requests: Juan C. Celedón, MD, DrPH, Channing Laboratory, 181 Longwood Avenue, Boston, MA juan.celedon@channing. harvard.edu /$32.00 Ó 2007 American Academy of Allergy, Asthma & Immunology doi: /j.jaci Abbreviation used FVC: Forced vital capacity Infection with Ascaris lumbricoides is the most prevalent helminthic disease in the world, 1 affecting an estimated 1.4 billion persons. 2 Ascariasis is very common in economically deprived areas of Hispanic America 3,4 but very rare in Costa Rica, 5 a prosperous Hispanic American nation. The prevalence of ascariasis in Costa Rica declined markedly between 1953 and 1996, likely because of improved hygienic conditions and widespread distribution of antihelminthic treatment by healthcare agencies. 5-7 Among children admitted to a tertiary hospital in San José (Costa Rica s capital) in 1953, the prevalence of ascariasis was moderately high (39.5%). 6 In 1996, the estimated prevalence of ascariasis among children and adults in a nationwide study in Costa Rica was ;2%. 5 The relation between infection with helminths such as A lumbricoides and atopy or asthma is controversial. 8 Helminthiasis has been associated with reduced risk of atopy 3,9 and/or asthma symptoms 4,10 in areas with high prevalence of parasitic infections, but with increased risks of atopy and asthma in areas with low prevalence of parasitic infections. 11 These findings suggest that the relation among helminthiasis, atopy, and asthma is complex and influenced by host factors, timing and intensity of infection, and concurrent environmental exposures. Little is known about the relation between past or current exposure to A lumbricoides and disease severity in individuals with asthma, 12 particularly in areas with low prevalence of helminthic infections. We examined the association between sensitization to A lumbricoides and asthma severity among children with asthma in Costa Rica, a country with a low prevalence of ascariasis but high prevalence of asthma (;23.7% among adolescents). 13 METHODS Study population Children who participated in this study were index cases for a genetic study of nuclear families of children with asthma in Costa Rica. From February 2001 to March 2005, short questionnaires were

2 J ALLERGY CLIN IMMUNOL VOLUME 119, NUMBER 3 Hunninghake et al 655 sent to the parents of 9054 children age 6 to 14 years who were enrolled in 95 schools in Costa Rica. Of the questionnaires distributed, 5355 (59.1%) were returned. Children were eligible for inclusion in the study if they had asthma (defined as physiciandiagnosed asthma and at least 2 respiratory symptoms [wheezing, cough, or dyspnea] or a history of asthma attacks in the previous year) and high probability of having at least 6 great-grandparents born in the Central Valley of Costa Rica. 14 Of the 5355 children whose parents returned the screening questionnaires, 1947 (36.4%) had asthma. Of these 1947 children, 439 (22.5%) unrelated children were willing to participate in the study along with their parents (for whom only genotypic information was collected) and had Central Valley ancestry. There was no significant difference in sex or grade in school for children who did and did not agree to participate in the study. Study procedures Study participants completed a protocol that included a questionnaire, spirometry, allergy skin testing, measurement of serum total and allergen-specific IgE, peripheral blood eosinophil count, and (on a separate visit) assessment of airway responsiveness to methacholine. In addition, a stool specimen was examined for ova and parasites in the first 137 study participants (because helminths were not found in any instance, stool samples were not collected for the remaining study subjects). Written parental consent was obtained for participating children, for whom written assent was also obtained. The study was approved by the Institutional Review Boards of the Hospital Nacional de Niños (San José, Costa Rica) and Brigham and Women s Hospital (Boston, Mass). Questionnaire Each participant completed a slightly modified version of the questionnaire used in the Collaborative Study on the Genetics of Asthma, 15 which was translated into Spanish. Pulmonary function tests and bronchodilator responsiveness Spirometry was conducted with a Survey Tach Spirometer (Warren E. Collins, Braintree, Mass) following American Thoracic Society recommendations. 16 Height was measured to the nearest half inch. Subjects were instructed to avoid use of short-acting bronchodilators for at least 4 hours before testing. Spirometric maneuvers were performed with subjects seated and wearing a noseclip. The best FEV 1 and forced vital capacity (FVC) were selected for data analysis of FEV 1 and FEV 1 /FVC. Because of a nonnormal distribution, the FEV 1 /FVC ratio was transformed to a log 10 scale for analysis. After completing baseline spirometry, subjects were given 200 mg (2 puffs) of an albuterol metered-dose inhaler using a spacer, and spirometry was repeated after 15 minutes. For the statistical analysis, bronchodilator responsiveness was treated as a binary (positive if there was an increment of at least 12% and at least 200 ml in baseline FEV 1 after administration of albuterol) and continuous (absolute difference, in milliliters, in FEV 1 measurements before and after administration of albuterol) outcome. Methacholine challenge testing After completion of baseline spirometry, subjects whose FEV 1 was at least 65% of predicted underwent methacholine challenge testing using a slightly modified version of the Chatham protocol. 17 The study protocol consisted of 5 breaths of saline solution followed by 1 breath of a 1 mg/ml methacholine solution, 1 and 4 breaths of a 5 mg/ml methacholine solution, and 1 breath of a 25 mg/ml methacholine solution. All inhalations, which were taken from a DeVilbiss 646 nebulizer (Sunrise Medical, Carlsbad, Calif), lasted 6 seconds and were followed by 2 seconds of breath holding. After each inhalation level, spirometry was performed at 180, 210, and 240 seconds. The test was terminated if the FEV 1 declined by at least 20% from the best FEV 1 value after inhalation of saline solution. For the statistical analysis, airway responsiveness was treated as a binary (positive if there was a 20% fall in baseline FEV 1 with a methacholine dose 1.98 mmol) and continuous (log 10 -transformed dose-response slope to methacholine) outcome. Allergy skin testing Skin testing was performed according to the protocol of the International Study of Asthma and Allergies in Childhood. 18 In addition to histamine and saline controls, the following antigens were applied to the volar surface of the forearm: Dermatophagoides pteronyssinus, Dermatophagoides farinae, Blatella germanica, Periplaneta americana, cat dander, dog dander, mixed grass pollen, mixed tree pollen, and Alternaria tenuis. A test was considered positive if the maximum diameter of the wheal was 3 mm after subtraction of the maximum diameter of the negative control. Serum total and allergen-specific IgE and peripheral blood eosinophil count Serum total and allergen-specific IgE levels were determined by the UniCAP 250 system (Pharmacia & Upjohn, Kalamazoo, Mich), with samples measured in duplicate. Total serum IgE levels were transformed to a log 10 scale for data analysis. Serum was assayed for IgE to 3 allergens: B germanica, D pteronyssinus, and A lumbricoides. IgEs to specific allergens were considered positive at a level 0.35 IU/mL. Eosinophil count was measured in peripheral blood by Coulter counter (Beckman Coulter, Fullerton, Calif) techniques and then transformed to a log 10 scale for data analysis. Examination of stool specimens for ova and parasites Microscopic examination of wet mounts of fresh stool specimens in saline and iodine was conducted at low and high power at the Parasitology Laboratory of the Hospital Nacional de Niños in San José (Costa Rica). Definition of other variables Sociodemographic, perinatal, and familial variables considered for inclusion in the multivariate analyses were the child s sex, age, height, parental education, number of persons in the household, number of older siblings, number of children sharing a bedroom, day care in the first year of life, in utero exposure to smoking, current exposure to cigarette smoking in the household, paternal history of asthma, maternal history of asthma, paternal history of hay fever, maternal history of hay fever, and use of medications for asthma (inhaled b 2 -agonists [long-acting and short-acting], inhaled corticosteroids, leukotriene inhibitors, theophylline, and oral corticosteroids) in the previous year. Statistical analysis Bivariate analyses were conducted by Fisher exact tests for categorical variables and 2-tailed t tests for categorical and continuous variables. Stepwise linear or logistic regression (as appropriate) was used to study the relation between a positive IgE to A lumbricoides (heretofore referred to as sensitization to Ascaris) and the outcomes of interest (markers of allergy, spirometric measures of lung function [FEV 1 and FEV 1 /FVC], airway responsiveness to methacholine, bronchodilator responsiveness, and asthma exacerbations [hospitalizations and emergency department visits the previous year]) while adjusting for potential confounders. All of the final multivariate models included age, sex, and parental education, as well as variables that satisfied a change-in-estimate criterion (10% in the

3 656 Hunninghake et al J ALLERGY CLIN IMMUNOL MARCH 2007 TABLE I. Characteristics of participating children with asthma in Costa Rica Sensitized to A lumbricoides N (%) Categorical variables Yes (n 5 171) No (n 5 268) for comparison* Male 119 (69.6) 157 (58.6).02 Parental education Less than high school 87 (50.9) 117 (43.7).33 Completed high school 39 (22.8) 73 (27.2) At least some college 45 (26.3) 78 (29.1) Current smoker living in the home 44 (25.7) 70 (26.3).91 Maternal asthma 49 (28.7) 90 (33.7).29 Paternal asthma 38 (22.5) 55 (20.8).72 Maternal hay fever 52 (30.4) 93 (34.8).35 Paternal hay fever 49 (28.8) 74 (27.8).83 Skin test reactivity to 1 allergen 162 (95.3) 213 (79.8) <.001 Positive IgE to dust mite 161 (94.2) 173 (64.6) <.001 Positive IgE to cockroach 119 (69.6) 63 (23.5) <.001 Airway responsiveness to methacholine 1.98 mmol 107 (70.4) 144 (58.3).02 Bronchodilator responsivenessà 32 (19.1) 21 (8.3).002 Hospitalized for asthma, last year 13 (7.6) 9 (3.4).07 Emergency department visits for asthma, last year 144 (84.2) 212 (79.1).21 *Comparison by Fisher exact test. Defined as the highest level of education of either parent. àan increase of at least 200 ml and at least 12% in FEV 1 after administration of inhaled albuterol. Information missing on some subjects for airway responsiveness to methacholine 1.98 mmol (n 5 40), bronchodilator responsiveness (n 5 18), presence of a current smoker in the home (n 5 2), maternal asthma (n 5 1), paternal asthma (n 5 5), maternal hay fever (n 5 1), paternal hay fever (n 5 3), and allergy skin testing (n 5 2). risk ratio or the odds ratio) or that were significant at P<.05. In addition, height was included in all models for measures of lung function. All analyses were performed using Statistical Analysis Software version 8.2 (SAS Institute, Cary, NC). RESULTS A comparison of selected characteristics of children who were (n 5 171) and were not (n 5 268) sensitized to Ascaris is presented in Tables I (for categorical variables) and II (for continuous variables). Children who were sensitized to Ascaris were more likely to be boys, to have been hospitalized for asthma in the previous year, and to have at least 1 positive skin test to allergens, a positive IgE to dust mite or cockroach, markedly increased airway responsiveness to methacholine (a 20% fall in FEV 1 with a methacholine dose 1.98 mmol), and significant bronchodilator responsiveness (an increment of at least 12% and at least 200 ml in baseline FEV 1 after administration of albuterol) than children who were not sensitized (Table I). Children who were sensitized to Ascaris were older and had greater total serum IgE, eosinophil count, bronchodilator responsiveness (as an absolute difference in FEV 1 before and after administration of bronchodilator), and airway responsiveness (expressed as log 10 -transformed dose-response slope to methacholine) but lower FEV 1 /FVC than nonsensitized children (Table II). There were no significant differences in parental education, FEV 1, and maternal or paternal history of asthma between children who were and were not sensitized to Ascaris (Tables I and II). Skin test reactivity to allergens, serum total and allergen-specific IgE, and eosinophil count The results of the bivariate and multivariate analyses of the relation between sensitization to Ascaris and binary and continuous markers of allergy and eosinophilia are shown in Tables III and IV. In bivariate analyses, sensitization to Ascaris was associated with a 5-fold increase in the odds of having 1 positive skin test to allergens; 9-fold and 7-fold increases in the odds of having a positive IgE to dust mite and cockroach, respectively; an increased number of positive skin tests to allergens; an increased total serum IgE level; and an increased eosinophil count. These results were not appreciably changed after adjustment for relevant covariates (model 1 in Tables III and IV). As an example of the results of the multivariate linear regression analysis of markers of allergy and eosinophilia (as continuous outcomes), sensitization to Ascaris would predict a 127 cell/m 3 increase in peripheral blood eosinophil count for a male subject of average characteristics in our study. Spirometric measures of lung function, airway responsiveness, and bronchodilator responsiveness The results of the bivariate and multivariate analyses of the relation between sensitization to Ascaris and binary and continuous indicators of lung function, airway responsiveness to methacholine, and bronchodilator responsiveness are shown in Tables III and IV. Sensitization to

4 J ALLERGY CLIN IMMUNOL VOLUME 119, NUMBER 3 Hunninghake et al 657 TABLE II. Characteristics of participating childrenz with asthma in Costa Rica Sensitized to A lumbricoides Continuous variables* Yes (n 5 171) No (n 5 268) for comparisony Age (y) 9.2 ( ) 8.5 ( ).004 Height (cm) 133 ( ) 129 ( ) No. of persons in the family 5 (4-6) 5 (4-6).74 No. of positive skin tests to allergens 4 (3-5) 3 (1-4) <.001 Total serum IgE (IU/mL) ( ) ( ) <.001 Eosinophil count (cells/ml 3 ) ( ) ( ) <.001 Baseline FEV 1 (L) 1.76 ( ) 1.65 ( ).28 Baseline FEV 1 /FVC (%) 81.3 ( ) 83.2 ( ).03 Absolute response to bronchodilator (ml) 80.0 ( ) 60.0 ( ).005 Dose-response slope to methacholine (mmol) 16.2 ( ) 13.2 ( ).006 *Results expressed as median (interquartile range). Comparison by 2-tailed t tests. àinformation missing on some subjects for the dose response slope to methacholine (n 5 40), absolute bronchodilator responsiveness (n 5 18), baseline FEV 1 (n 5 8), FEV 1 /FVC ratio (n 5 8), height (n 5 3), number of positive skin tests (n 5 2), and total eosinophil count (n 5 1). TABLE III. Sensitization to A lumbricoides and categorical measures of allergy, asthma morbidity, and asthma severity in Costa Rican children Odds ratio (95% CI) Adjusted* Outcomes Unadjusted Model 1 Model 2 Model 3 Skin test reactivity to 1 allergen 5.13 ( ) 5.15 ( ) 0.62 ( ) <.001 < Positive IgE to dust mite 8.84 ( ) 8.93 ( ) 9.02 ( ) <.001 <.001 <.001 Positive IgE to cockroach 7.45 ( ) 7.31 ( ) 6.25 ( ) <.001 <.001 <.001 Airway response to 1.98 mmol 1.70 ( ) 1.61 ( ) 1.30 ( ) 0.90 ( ) of methacholine Bronchodilator responsiveness 2.60 ( ) 2.60 ( ) 2.34 ( ) 3.12 ( ) Emergency department visits for 1.41 ( ) 1.53 ( ) 1.35 ( ) 1.09 ( ) asthma in the previous yearà Hospitalizations for asthma in the previous year 2.37 ( ) 3.08 ( ) 2.98 ( ) 1.97 ( ) *Models 1, 2, and 3 were adjusted for age, sex, and parental education level for all outcomes. For all outcomes, model 2 was additionally adjusted for skin test reactivity to 1 allergen and model 3 was additionally adjusted for a positive IgE to dust mite and a positive IgE to cockroach. All models for FEV 1 and FEV 1 /FVC were additionally adjusted for height, and all models for airway responsiveness and bronchodilator responsiveness were additionally adjusted for height and baseline FEV 1. Also adjusted for paternal asthma history. àalso adjusted for presence of a current smoker in the home. Also adjusted for use of anti-inflammatory medications (inhaled steroids and/or leukotriene inhibitors) in the previous year. Ascaris was associated with a 1.6-fold increase in the odds of having markedly increased airway responsiveness (a 20% fall in FEV 1 with a methacholine dose 1.98 mmol) and a 2.6-fold increase in the odds of having significant bronchodilator responsiveness (an increment of at least 12% and at least 200 ml in baseline FEV 1 after administration of albuterol; model 1 in Table III). In addition, sensitization to Ascaris was significantly associated with increased airway responsiveness (expressed as log 10 -transformed dose-response slope to methacholine) and a reduction in FEV 1, and nonsignificantly associated with increased bronchodilator responsiveness (as an absolute difference in FEV 1 before and after administration of bronchodilator) and a reduced FEV 1 /FVC (model 1 in Table IV). Asthma exacerbations Table III shows the results of the analysis of the relation between sensitization to Ascaris and asthma exacerbations in the previous year. Sensitization to Ascaris was associated with a 1.4-fold increase in the odds of at least 1 visit to the emergency department for asthma in the previous

5 658 Hunninghake et al J ALLERGY CLIN IMMUNOL MARCH 2007 TABLE IV. Sensitization to A lumbricoides and continuous measures of allergy, asthma morbidity, and asthma severity in Costa Rican children Coefficient estimate (95% CI) Adjusted* Outcomes Unadjusted Model 1 Model 2 Model 3 Total IgE (IU/mL) 0.58 ( ) 0.57 ( ) 0.46 ( ) 0.24 ( ) <.001 <.001 <.001 <.001 Eosinophil count 0.18 ( ) 0.20 ( ) 0.14 ( ) 0.10 ( ) (cells/m 3 ) <.001 <.001 < Number of positive 1.08 ( ) 1.06 ( ) 0.05 ( ) skin testsà <.001 < Baseline FEV 1 (L) 0.05 ( 0.04 to 0.14) (20.12 to 20.01) (20.11 to ) (20.12 to 0.004) Baseline FEV 1 /FVC (%) (20.02 to ) (20.01 to 0.001) (20.01 to 0.002) (20.01 to 0.01) Dose-response slope to 0.15 ( ) 0.15 ( ) 0.08 ( 0.03 to 0.19) (20.13 to 0.11) methacholine (mmol) Absolute response to ( ) ( ) ( 6.00 to 49.08) ( 4.11 to 57.03) bronchodilator (ml)k *Models 1, 2, and 3 were adjusted for age, sex, and parental education for all outcomes. For all outcomes, model 2 was additionally adjusted for skin test reactivity to 1 allergen and model 3 was additionally adjusted for a positive IgE to dust mite and a positive IgE to cockroach. All models for FEV 1 and FEV 1 /FVC were additionally adjusted for height, and models for airway responsiveness and bronchodilator responsiveness were additionally adjusted for height and baseline FEV 1. Variable was log 10 -transformed before analysis. àalso adjusted for number of older siblings. Also adjusted for number of children sharing the bedroom. kalso adjusted for use of inhaled corticosteroids. year; these results were not appreciably changed after adjustment for relevant covariates (results for neither the bivariate nor the multivariate analysis reached statistical significance). Sensitization to Ascaris was also associated with a 2.4-fold increase in the odds of having at least 1 hospitalization for asthma in the previous year. After adjustment for use of anti-inflammatory medications (inhaled steroids and/or leukotriene inhibitors), sensitization to Ascaris was associated with a 3-fold increase in the odds of having at least 1 hospitalization for asthma in the previous year. Exploratory analyses To examine whether atopy explains the observed association between sensitization to Ascaris and measures of asthma morbidity and severity, we repeated the multivariate analysis after adjustment for skin test reactivity to 1 aeroallergen. In that analysis, there was only a change in the association between sensitization to Ascaris and airway responsiveness, which was appreciably attenuated (model 2 in Tables III and IV). We also repeated the analysis after adjustment for positive IgE to dust mite and/ or cockroach. In that analysis, sensitization to Ascaris was still significantly associated with increased total serum IgE, increased eosinophil count, and bronchodilator responsiveness (as a binary variable; model 3 in Tables III and IV). Sensitization to each of 2 common allergens (dust mite and cockroach) was associated with various measures of asthma severity and asthma morbidity (Tables V and VI), suggesting that sensitization to allergens other than Ascaris is associated with disease severity in Costa Rican children with asthma. DISCUSSION To our knowledge, this is the first report of an association between sensitization to A lumbricoides and increased asthma morbidity. Among children with asthma in Costa Rica, sensitization to Ascaris was associated with increased odds of allergic sensitization, increased levels of total serum IgE and eosinophils in peripheral blood, increased airway responsiveness and bronchodilator responsiveness, reduced lung function, and hospitalizations for asthma in the previous year. Studies of children not selected on the basis of asthma in Brazil, 19 Germany, 20 and South Africa 21 have reported that sensitization to Ascaris is associated with current wheeze, 19 increased total serum IgE levels, 20 higher prevalence of asthma and hay fever (by parental report), 20 atopic asthma, 21 atopic airway responsiveness, 21 and increased odds of sensitization to 1 aeroallergen. 20,21 Those studies were limited by small sample size, 19 limited information on socioeconomic status, 19 lack of data on the prevalence of helminthiasis in the source population, 19 nonexamination of stool specimens for helminths, 19,20 and absent 19,20 or limited 21 assessment of lung function or airway responsiveness. Among 39 children with asthma in an area with high prevalence of helminthiasis (Coche

6 J ALLERGY CLIN IMMUNOL VOLUME 119, NUMBER 3 Hunninghake et al 659 TABLE V. Sensitization to D pteronyssinus and B germanica and categorical measures of allergy, asthma morbidity, and asthma severity in Costa Rican children Outcomes Airway responsiveness to 1.98 mmol of methacholine Bronchodilator responsiveness Emergency department visits for asthma in the previous year Hospitalizations for asthma in the previous yearà Odds ratio* (95% CI) Model 1 (dust mite) Model 2 (cockroach) 3.78 ( ) 2.32 ( ) <.001 < ( ) 0.98 ( ) ( ) 1.84 ( ) ( ) 2.59 ( ) *All multivariate models were adjusted for age, sex, and parental education. Models for airway responsiveness and bronchodilator responsiveness were additionally adjusted for height and baseline FEV 1. Also adjusted for presence of a current smoker in the home. àalso adjusted for use of anti-inflammatory medications (inhaled steroids or leukotriene inhibitors) in the previous year. Island, Venezuela), unblinded antihelminthic treatment for 1 year resulted in a 2-year reduction in asthma attacks and medication use but no significant change in pulmonary function or frequency of sensitization to Ascaris. 12 Together with findings from previous studies in children not selected on the basis of a diagnosis of asthma, 19,20 our results suggest that the observed association between sensitization to Ascaris and indicators of asthma severity (eg, increased airway responsiveness) and asthma morbidity (eg, asthma exacerbations) is likely mediated by an increased degree of atopy in children who are sensitized to Ascaris. In support of this hypothesis, the observed association between sensitization to Ascaris and airway responsiveness was attenuated and became nonstatistically significant after adjustment for having a positive skin to at least 1 of the common allergens in Costa Rica 22 (model 2inTables III and IV). However, sensitization to Ascaris was significantly associated with other measures of asthma morbidity and asthma severity, suggesting that sensitization to or cross-reactivity with other allergens is not the sole explanation for the observed findings. Although analytical adjustment for a positive IgE to dust mite and/or cockroach may be overly conservative because of strong collinearity among measured IgEs, sensitization to Ascaris remained significantly associated with increased total serum IgE, increased eosinophil count, and bronchodilator responsiveness (as a binary outcome) after such adjustment (model 3 in Tables III and IV). The latter finding is interesting in light of a reported association between polymorphisms in the gene for the B 2 -adrenergic receptor (ADRB2) and intensity of helminthic infection. 23 TABLE VI. Sensitization to D pteronyssinus and B germanica and continuous measures of allergy, asthma morbidity, and asthma severity in Costa Rican children Outcomes Total IgE (IU/mL) Eosinophil count (cells/m 3 ) Baseline FEV 1 (L)à Baseline FEV 1 / FVC (%) Dose-response slope to methacholine (mmol) Absolute response to bronchodilator (ml) Coefficient estimate* (95% CI) Model 1 (dust mite) Model 2 (cockroach) 0.84 ( ) 0.64 ( ) <.001 < ( ) 0.16 ( ) <.001 < ( (20.09 to 0.02) to 0.03) (20.02 to (20.02 to 0.001) ) ( ) 0.21 ( ) <.001 < ( ) ( to 26.30) *All models were adjusted for age, sex, and parental education. Models for FEV 1 and FEV 1 /FVC were additionally adjusted for height, and models for airway responsiveness and bronchodilator responsiveness were adjusted for height and baseline FEV 1. Variable was log 10 -transformed before analysis. àalso adjusted for number of children sharing the bedroom. Also adjusted for use of inhaled corticosteroids. Because of the cross-sectional design of our study, we cannot establish causality for the increased degree of atopy among children sensitized to Ascaris. A plausible explanation for our findings is that children with asthma and severe atopy (which is associated with increased asthma severity and morbidity) have enhanced immune responses against A lumbricoides and may thus be protected from infection with this helminth. 27,28 Among Venezuelan children living in 2 areas with high prevalence of ascariasis (51% to 57%), a group with high prevalence of atopy (80%) had higher levels of specific IgE to Ascaris but a lower intensity of Ascaris infection than a group with a low prevalence of atopy (20%). 27 More recently, Cooper et al 29 demonstrated that atopic children had higher levels of T H 2 cytokines (IL-4 and IL-5) and greater histamine release in response to Ascaris antigens than nonatopic children. Although sensitization to Ascaris could also be a result of cross-reactivity between proteins present in Ascaris, dust mite, and cockroach (eg, tropomyosin) 30 in children with severe atopy, this remains unproven. 31 Cross-reactivity is unlikely to be the sole explanation for our results because 213 (56.8%) of the 375 participating children who had at least 1 positive skin test to allergens and significant subsets of children who had a positive IgE to either dust mite (51.8%) or cockroach (34.6%) were not sensitized to Ascaris (Table I). Furthermore, 8 (12.9%) of the 62 participating children who had no positive skin tests to allergens were sensitized to Ascaris,

7 660 Hunninghake et al J ALLERGY CLIN IMMUNOL MARCH 2007 suggesting that although most children who are sensitized to Ascaris are highly atopic (43.7%), sensitization to Ascaris may occur in the absence of sensitization to common aeroallergens. Furthermore, results of recent genomewide linkage analyses of allergy phenotypes in large families of Costa Rican children with asthma suggest that having a positive IgE to Ascaris is under distinct genetic regulation from having a positive IgE to either B germanica or D pteronyssinus (J. C. Celedón, personal communication, August 2006). An alternative explanation for our results is that previous removal of immunoregulatory influences by helminthic infection (eg, stimulation of T-regulatory cells) 8,32 in susceptible children leads to more severe atopy and thus increased asthma severity and morbidity. Although a recent large randomized trial of antihelminthic therapy in Ecuador did not show an increased prevalence of atopy in the year after treatment, 33 reversible suppression of allergen sensitization by helminthiasis was shown in a clinical trial of antihelminthic treatment in Gabon (Africa) 34 and in an uncontrolled study of antihelminthic treatment in Venezuela. 3 The discrepant findings of the trials in Ecuador and Gabon could be explained by differences in frequency and intensity of helminthic infections, duration of follow-up, and concurrent environmental or infectious exposures. The current low prevalence of helminthiasis in Costa Rica is the result of improved hygienic conditions and systematic campaigns for antihelminthic treatment in a tropical country where exposure to helminths is common. Current Costa Rican national guidelines for primary care physicians recommend empiric antihelminthic treatment for all schoolchildren biannually. 7 The low prevalence of helminthiasis in Costa Rican schoolchildren (3.4% for Ascaris, 5.6% for Trichuris, and 1.9% for Necator in 1996) was confirmed by our negative findings in the first 137 children participating in the current study. 5 Our finding of a high prevalence of Ascaris sensitization coupled with a low prevalence of ascariasis is consistent with a recent report from South Africa. 21 Sensitization to Ascaris may result from past or current exposure to Ascaris and other helminths. 35,36 The ability to generate IgE in response to Ascaris may be a more important predictor of asthma severity than other markers of past infectious exposure, because Ascaris-specific IgG does not appear to be strongly associated with asthma. 37 Levels of IgE specific for Ascaris may rise in the first year after antihelminthic treatment, 12 with a subsequent decline over a period of 2 to 4 years. 20 Although we did not assess sensitization to other parasites, the prevalence of infection with helminths other than Ascaris (eg, Necator americanus, Strongyloides stercoralis) is very low (<0.5% for Strongyloides and 1.9% for Necator) in Costa Rican children. 5,7 Although infection with Trichuris trichiura is slightly more common in Costa Rican schoolchildren (3.4% vs 5.6%), 5 Trichuris does not have a systemic phase as part of its life cycle 38 and has not been frequently implicated in the pathogenesis of asthma or atopy. In particular, infection with Ascaris, but not with Trichuris, was associated with asthma in a recent meta-analysis. 39 We recognize that we cannot exclude low-grade helminthiasis in a small minority of children in the absence of serial stool examinations for ova and parasites. However, active and/or intense ascariasis is unlikely in most of the participating children given the relatively high sensitivity and specificity of a single stool examination in asymptomatic populations with low prevalence of Ascaris infection 40,41 and results from previous nationwide surveys. Selection bias and/or residual confounding by socioeconomic status are unlikely explanations for our results. Because parents were largely unaware of the pattern of allergic sensitization of their children, entry into the study was unlikely to be affected by the presence or absence of sensitization to Ascaris. All of our analyses were adjusted for level of parental education, and other variables potentially associated with low socioeconomic status (eg, crowding, cigarette smoking) were considered for inclusion in our final multivariate models. In addition, extreme gradations in household income and/or access to health care would not be expected among Costa Rican children attending public schools (Costa Ricans enjoy universal access to health care). In summary, we found a strong association between sensitization to A lumbricoides and increased asthma severity and asthma morbidity in a country with a low prevalence of helminthiasis. This association is likely mediated by an increased degree of atopy among children with asthma who are sensitized to Ascaris. Longitudinal studies are needed to determine whether this finding is a result of removal of helminthic suppression of atopy in children with asthma who received anthelminthic treatment and/ or increased protection against helminthic infection in children with asthma and severe atopy. Regardless of the mechanisms for our findings, sensitization to Ascaris appears to be an important marker of severe atopy, increased disease morbidity, and increased disease severity in children with asthma. We thank the participating families for their collaboration, the members of our field team in Costa Rica (Ligia Sanabria, Jenny Vega, Marvin Corrales, Adriana Gonzalez, Raquel Boza, Joaquín Acuña, Laura Rojas, Ana Castillo, Gabriela Ivankovich, Marcia Solano, Herminia Solano, and Heliberto Mena), and Ms Jaylyn Olivo for editorial assistance. REFERENCES 1. Chan MS, Medley GF, Jamison D, Bundy DA. The evaluation of potential global morbidity attributable to intestinal nematode infections. Parasitology 1994;109: Khuroo MS. Ascariasis. Gastroenterol Clin North Am 1996;25: Lynch NR, Hagel I, Perez M, Di Prisco MC, Lopez R, Alvarez N. Effect of anthelmintic treatment on the allergic reactivity of children in a tropical slum. J Allergy Clin Immunol 1993;92: Cooper PJ, Chico ME, Bland M, Griffin GE, Nutman TB. Allergic symptoms, atopy, and geohelminth infections in a rural area of Ecuador. Am J Respir Crit Care Med 2003;168: Ministerio de Salud. Encuesta Nacional de Nutricion, 1996: Helmintos Intestinales. E San Jose: Universidad de Costa Rica; Morales MT, Hernández BI. Frecuencia de cuatro nematodos intestinales en el Hospital Nacional de Niños. Acta Pediatr Costarric 1997;11:106-8.

8 J ALLERGY CLIN IMMUNOL VOLUME 119, NUMBER 3 Hunninghake et al Normas de Atencion Integral de Salud, Primer Nivel de Atención. San José, Costa Rica: Caja Costarricense de Seguridad Social; p Cooper PJ. Intestinal worms and human allergy. Parasite Immunol 2004; 26: Cooper PJ, Chico ME, Rodrigues LC, Ordonez M, Strachan D, Griffin GE, et al. Reduced risk of atopy among school-age children infected with geohelminth parasites in a rural area of the tropics. J Allergy Clin Immunol 2003;111: Scrivener S, Yemaneberhan H, Zebenigus M, Tilahun D, Girma S, Ali S, et al. Independent effects of intestinal parasite infection and domestic allergen exposure on risk of wheeze in Ethiopia: a nested case-control study. Lancet 2001;358: Palmer LJ, Celedon JC, Weiss ST, Wang B, Fang Z, Xu X. Ascaris lumbricoides infection is associated with increased risk of childhood asthma and atopy in rural China. Am J Respir Crit Care Med 2002;165: Lynch NR, Palenque M, Hagel I, DiPrisco MC. Clinical improvement of asthma after anthelminthic treatment in a tropical situation. Am J Respir Crit Care Med 1997;156: Worldwide variation in prevalence of symptoms of asthma, allergic rhinoconjunctivitis, and atopic eczema: ISAAC. The International Study of Asthma and Allergies in Childhood (ISAAC) Steering Committee. Lancet 1998;351: Escamilla MA, Spesny M, Reus VI, Gallegos A, Meza L, Molina J, et al. Use of linkage disequilibrium approaches to map genes for bipolar disorder in the Costa Rican population. Am J Med Genet 1996;67: Blumenthal MN, Banks-Schlegel S, Bleecker ER, Marsh DG, Ober C. Collaborative studies on the genetics of asthma: National Heart, Lung and Blood Institute. Clin Exp Allergy 1995;25(suppl 2): Standardization of Spirometry, 1994 update. American Thoracic Society. Am J Respir Crit Care Med 1995;152: Chatham M, Bleecker ER, Smith PL, Rosenthal RR, Mason P, Norman PS. A comparison of histamine, methacholine, and exercise airway reactivity in normal and asthmatic subjects. Am Rev Respir Dis 1982;126: Weiland SK, Bjorksten B, Brunekreef B, Cookson WO, von Mutius E, Strachan DP. Phase II of the International Study of Asthma and Allergies in Childhood (ISAAC II): rationale and methods. Eur Respir J 2004;24: Camara AA, Silva JM, Ferriani VP, Tobias KR, Macedo IS, Padovani MA, et al. Risk factors for wheezing in a subtropical environment: role of respiratory viruses and allergen sensitization. J Allergy Clin Immunol 2004;113: Dold S, Heinrich J, Wichmann HE, Wjst M. Ascaris-specific IgE and allergic sensitization in a cohort of school children in the former East Germany. J Allergy Clin Immunol 1998;102: Obihara CC, Beyers N, Gie RP, Hoekstra MO, Fincham JE, Marais BJ, et al. Respiratory atopic disease, Ascaris-immunoglobulin E and tuberculin testing in urban South African children. Clin Exp Allergy 2006;36: Soto-Quiros ME, Stahl A, Calderon O, Sanchez C, Hanson LA, Belin L. Guanine, mite, and cockroach allergens in Costa Rican homes. Allergy 1998;53: Ramsay CE, Hayden CM, Tiller KJ, Burton PR, Hagel I, Palenque M, et al. Association of polymorphisms in the beta2-adrenoreceptor gene with higher levels of parasitic infection. Hum Genet 1999;104: Siroux V, Oryszczyn MP, Paty E, Kauffmann F, Pison C, Vervloet D, et al. Relationships of allergic sensitization, total immunoglobulin E and blood eosinophils to asthma severity in children of the EGEA Study. Clin Exp Allergy 2003;33: Ponsonby AL, Gatenby P, Glasgow N, Mullins R, McDonald T, Hurwitz M. Which clinical subgroups within the spectrum of child asthma are attributable to atopy? Chest 2002;121: Carroll WD, Lenney W, Child F, Strange RC, Jones PW, Whyte MK, et al. Asthma severity and atopy: how clear is the relationship? Arch Dis Child 2006;91: Lynch NR, Hagel IA, Palenque ME, Di Prisco MC, Escudero JE, Corao LA, et al. Relationship between helminthic infection and IgE response in atopic and nonatopic children in a tropical environment. J Allergy Clin Immunol 1998;101: Cooper PJ. The potential impact of early exposures to geohelminth infections on the development of atopy. Clin Rev Allergy Immunol 2004;26: Cooper PJ, Chico ME, Sandoval C, Nutman TB. Atopic phenotype is an important determinant of immunoglobulin E-mediated inflammation and expression of T helper cell type 2 cytokines to ascaris antigens in children exposed to ascariasis. J Infect Dis 2004;190: Fernandes J, Reshef A, Patton L, Ayuso R, Reese G, Lehrer SB. Immunoglobulin E antibody reactivity to the major shrimp allergen, tropomyosin, in unexposed Orthodox Jews. Clin Exp Allergy 2003;33: Bernardini R, Mistrello G, Novembre E, Roncarolo D, Zanotta S, Lombardi E, et al. Cross-reactivity between IgE-binding proteins from Anisakis simplex and Dermatophagoides pteronyssinus. Int J Immunopathol Pharmacol 2005;18: Maizels RM, Yazdanbakhsh M. Immune regulation by helminth parasites: cellular and molecular mechanisms. Nat Rev Immunol 2003;3: Cooper PJ, Chico ME, Vaca MG, Moncayo AL, Bland JM, Mafla E, et al. Effect of albendazole treatments on the prevalence of atopy in children living in communities endemic for geohelminth parasites: a clusterrandomised trial. Lancet 2006;367: van den Biggelaar AH, Rodrigues LC, van Ree R, van der Zee JS, Hoeksma-Kruize YC, Souverijn JH, et al. Long-term treatment of intestinal helminths increases mite skin-test reactivity in Gabonese schoolchildren. J Infect Dis 2004;189: Turner KJ, Fisher EH, McWilliam AS. Homology between roundworm (Ascaris) and hookworm (N. americanus) antigens detected by human IgE antibodies. Aust J Exp Biol Med Sci 1980;58: Revoltella R, Jayakar SD, Tinelli M, Scaglia M, Peracino A, Desmarais JC, et al. Parasite-reactive serum IgE antibodies in African populations: relation to intestinal parasite load. Int Arch Allergy Appl Immunol 1980; 62: Karadag B, Ege M, Bradley JE, Braun-Fahrlander C, Riedler J, Nowak D, et al. The role of parasitic infections in atopic diseases in rural schoolchildren. Allergy 2006;61: Cooper E. Tropical infectious diseases. Philadelphia: Churchill Livingstone; Leonardi-Bee J, Pritchard D, Britton J. Asthma and current intestinal parasite infection: systematic review and meta-analysis. Am J Respir Crit Care Med 2006;174: Gyorkos TW, MacLean JD, Law CG. Absence of significant differences in intestinal parasite prevalence estimates after examination of either one or two stool specimens. Am J Epidemiol 1989;130: Morris AJ, Wilson ML, Reller LB. Application of rejection criteria for stool ovum and parasite examinations. J Clin Microbiol 1992;30: