Blood eosinophil related to maternal allergic rhinitis is associated with the incidence of allergic rhinitis in offspring: COCOA study

This study was conducted as part of the Cohort for Childhood Origin of Asthma and Allergic Disease (COCOA). COCOA is a general prospective population study as a hospital-based birth cohort designed to reveal the impact of environmental exposure in South Korea. COCOA includes demographic variables such as maternal height and weight, disease history, smoking history, tracks children’s height, environmental questionnaire responses, clinical symptoms, and disease history every year. A detailed description of the COCOA study design has been published previously [14]. A total of 3,004 pregnant women with gestational age less than 26 weeks were recruited for the COCOA study. Of these, 2,846 children were born, excluding mothers with inappropriate or withdrawn consent and those with missing follow-up. Women with missing data for major variables, such as the eosinophil level and Der f-IgE during pregnancy and the child’s sex, were excluded. Finally, 983 mother–child pairs were included in this study (Fig. 1). Written informed consent was obtained from the parents of all infants. Ethical approval for this study was obtained from the Institutional Review Board (IRB) of Asan Medical Center (IRB no. 2008–0616 and 2015–1031), Samsung Medical Center (IRB no. 2009–02–021), Severance Hospital (IRB no. 4–2008 − 0588), CHA Medical Center (IRB no. 2010–010), Seoul National University Hospital (IRB no. H-1401-086-550), and Korea Disease Control and Prevention Agency (IRB no. 2020-03-06-C-A).

IgE (total IgE and Der f-IgE) levels and leukocyte levels in the mother were measured at the time of the infant’s birth. IgE (total IgE, milk-specific IgE, and egg-specific IgE) levels and leukocyte levels in the children were measured at 1 and 3 years of age. Total serum IgE levels and serum levels of specific IgE to egg and milk were measured using a fluorescent enzyme immunoassay (AutoCAP System; Pharmacia Diagnostics AB, Uppsala, Sweden). Leukocytes were calculated using an automatic blood-cell counter (XE-100;Sysmex Co., Kobe, Japan) [14].

The questionnaire based on the International Study of Asthma and Allergies in Childhood (ISAAC) was translated into Korean. The parents of infants filled out a modified version of the questionnaire. The questionnaire consists of (1) demographic factors such as gender, date of birth, height, and weight; (2) symptom history related to AR; (3) environmental factors related to allergic diseases such as residential patterns and indoor air quality [15].

SPT was performed in the mother, 6 months old infant, and in the child on the third birthday. SPT was conducted for 18 allergens (i.e. Der p, Der f, German cockroach, grasses mixture, alder, birch, oak, Japanese hop, mugwort, ragweed, dog epithelium, cat epithelium, Alternaria alternata, Aspergilus fumigatus, peanut, cow’s milk, egg white, and soybean) in children and 14 allergens (i.e. Der p, Der f, German cockroach, grasses mixture, Trees I, Trees II, mugwort, ragweed, dog epithelium, cat epithelium, Alternaria alternata, Aspergilus fumigatus, oak, and alder) in mothers using normal saline and histamine as a negative and positive control, respectively. A mean wheal size greater than 3 mm was considered a positive reaction.

AR diagnositic information was used when the child was 3 years old. AR was suspected when the participant complained symptoms of AR, such as runny nose and stuffy nose. AR was diagnosed by a physician considering SPT results (Supplementary Tables 1, 2) and the participant’s history of: (1) showed symptoms of runny nose, stuffy nose or sneezing; (2) were treated with AR; or (3) were diagnosed with AR [16].

All statistical analyses were performed using the R program (R Foundation, Austria, available from http://​www.​r-project.​org/​foundation/​main.​html). Summary statistics are expressed as mean ± standard deviation (SD) or number (percentage). Two groups were compared by t-test. ANOVA was performed to determine whether maternal blood parameters differed by the maternal Der f-IgE levels at delivery. The children were divided into three groups according to the tertiles of mothers’ Der f-IgE and eosinophil levels. ANOVA was also performed to compare blood parameters in children aged 1 and 3 years according to the Der f-IgE and eosinophil levels of the mothers classified into three groups. The values divided into groups are as follows: (1) the maternal Der f-IgE: low(≤ 0.01), middle(> 0.01 to ≤ 0.92 ), and high(> 0.92); (2)maternal eosinophil: low(≤ 0.6), middle(> 0.6 to ≤ 1.2 ), and high(> 1.2); (3) infant’s 1y eosinophil: low(≤ 2.0), middle(> 2.0 to ≤ 3.5 ), and high(> 3.5); (4)infant’s 3y eosinophil: low(≤ 2.1), middle(> 2.1 to ≤ 3.7 ), and high(> 3.7). Also, logistic regression analysis was performed to determine whether maternal and childhood eosinophil counts were associated with 3-year-old AR in children. Three types of logistic regression models were constructed. The models are as follows: (i) Model 1, maternal age, BMI, total IgE at birth and secondhand smoke in pregnancy; (ii) Model 2, sex, and total IgE at 1 year old; (iii) Model 3, sex, and total IgE at 3 years of age. A logistic regression was performed to examine the association between higher eosinophilic levels in both mothers and children and the risk of AR incidence at age 3, using sex, maternal age at birth, and secondhand smoke in pregnancy as confounder variables. Significance was set at p-value < 0.05.