Fruit, vegetable and dietary antioxidant intake in school age, respiratory health up to young adulthood

Dietary antioxidants may protect the lung against oxidative damage and prevent chronic respiratory disease. We aimed to investigate fruit, vegetable and antioxidant intake (measured as total antioxidant capacity, TAC) at age 8 years in relation to asthma and lung function up to 24 years.


| INTRODUC TI ON
Asthma is a common chronic respiratory disease affecting both children and adults. Asthma usually starts in childhood, and although some outgrow their disease, a large proportion will suffer from lifelong symptoms, reduced quality of life and long-term lung function deficits. 1,2 However, asthma may also debut in adulthood, and while it is more common among males in early childhood, it becomes more common among females after puberty. 3 Several epidemiological studies have investigated the potential modulatory effect of diet on respiratory health and suggested that diets rich in antioxidants may protect the lung against oxidative damage and prevent chronic respiratory disease. [4][5][6] Most longitudinal studies have focused on maternal exposure during pregnancy, reporting lower risk of wheeze or asthma or improved lung function in the offspring. [7][8][9] Fewer studies have examined antioxidant intake during childhood. A study from the Dutch PIAMA birth cohort found inverse associations between childhood fruit intake and asthma symptoms up to age 8 years, 10 and similar results were observed in a Japanese cohort followed from age 7 to 10 years for the onset p-trend .072) up to 24 years. Associations were more pronounced among males, and regarding asthma, participants sensitized to aeroallergens.
Conclusions: Antioxidant intake in school age may improve asthma and lung function up to young adulthood. Although our results should be interpreted with caution, they emphasize the importance of following current dietary guidelines regarding consumption of antioxidant-rich foods as part of a balanced diet.

G R A P H I C A L A B S T R A C T
This longitudinal study of 2506 participants from the population-based birth cohort BAMSE aimed to investigate the association of fruit, vegetable and dietary total antioxidant capacity (TAC) at age 8 years, with repeated measures of asthma and lung function up to 24 years. Diet and asthma were assessed by questionnaires and lung function by spirometry. Results from this study suggest that high fruit and dietary antioxidant intake in school age may improve asthma and lung function up to young adulthood.

Key Messages
• Antioxidant intake in school age may improve asthma and lung function up to young adulthood.
• Males and participants sensitized to aeroallergens may benefit more from an increased intake.
• General dietary recommendations regarding intake of antioxidant-rich foods may be important for respiratory health.
of respiratory allergic symptoms. 11 Furthermore, although there is some support for an association between high fruit intake and improved lung function in young adulthood from cross-sectional studies, 12 there is a lack of prospective studies on antioxidant intake in childhood with follow-up beyond school age.
Since single antioxidants may not reflect the total antioxidant power of diet, total antioxidant capacity (TAC) has been used as a cumulative measure of antioxidants consumed in the whole diet and as a proxy to describe overall diet quality. 13  pre-specified response categories ranging from "never" to "≥three times/day". 18 The consumption frequencies were converted into average daily consumption and summarized as fruits and vegetables, respectively.
Individual TAC estimates were obtained by combining the information on frequency of consumption of specific food items with information from a database of common foods analysed with the oxygen radical absorbance capacity (ORAC) method 19

| Assessment of asthma and aeroallergen sensitization
Participants were considered to have asthma if they had at least two of the following three criteria, based on parental (age 8 years) and participant (

| Assessment of lung function
Lung function (primary outcome) was tested according to ATS/ERS spirometry criteria at age 8 years using a 2200 Pulmonary Function Laboratory (SensorMedics), 16 years using a Jaeger MasterScreen-IOS system (CareFusion Technologies) and 24 years using a Vyaire Vyntus system (Vyaire Medical). 23 The same spirometry test protocol was used at all time-points. Z-scores for forced expiratory volume in the first second (FEV 1 ) and forced vital capacity (FVC) were computed for each participant, using the GLI reference equations. 24 Pre-bronchodilator measurements were used.

| Statistical analyses
Intake of fruits, vegetables and TAC at age 8 years was divided into tertiles, and when numbers were small (in stratified analyses and analyses of incidence and remission), it was dichotomized (by the median). Potential confounding factors were identified from existing literature. 25 In the multivariable models, we adjusted for sex, total energy intake (kcal·day −1 ) at 8 years, parental education (elementary school, high school, university), parental ethnicity (born in or outside of Sweden), parental history of atopic disease (yes/no), maternal age at delivery <25 years (yes/no), maternal smoking in pregnancy and/ or infancy (yes/no) and older siblings (yes/no). Fruit and vegetable intakes were mutually adjusted through inclusion in the same multivariable model. Variables additionally adjusted, which did not influence the results and were not included in the final models are listed in the supplement.
We stratified analyses with prevalent asthma by sex and aeroallergen sensitization, and lung function analyses additionally by asthma at age 8 years and tested possible reverse causation (i.e. that the disease would have influenced the exposure) by excluding children who reported allergic symptoms related to fruits or vegetables, and/or who avoided any of these due to allergic symptoms at age 8 years. We also performed sensitivity analyses, including long-term fruit and vegetable intake, defined as mean intake in grams/day at ages 8 and 16 years, and adjusting for intake at 24 years.
Participants with baseline questionnaire data, data on fruit, vegetable and TAC intake at age 8 years with a mean energy intake within ±3 log SD and information on asthma at 8, 16 and/or 24 years were included in the study population (n = 2506); participants with anthropometric measurements and spirometry results at 8, 16 and/ or 24 years were included in lung function analyses (n = 1483). The 12-year follow-up was not included, due to lung function data not being available. The flow chart of the study is shown in Figure S1. All analyses were performed using the statistical software STATA V.16 (StataCorp).

| RE SULTS
The study population (n = 2506) was comparable to the original cohort (n = 4089) with regard to distribution of selected characteristics, apart from higher proportions of parents with university education (54.5% vs. 52.9%) and allergic disease ( and FVC) more than doubled from 8 to 16 years and continued to increase from 16 to 24 years, apart from FEV 1 among females which did not change significantly.
Participants with higher fruit and vegetable intake (≥ median) had similar distribution of various dietary, lifestyle and demographic characteristics compared to those with lower intake, apart from a higher energy and fish intake (Table S2). Moreover, they were more often females, had a parent with university education and had less often a young mother at baseline. Participants with higher TAC intake (≥ median) had a similar distribution by sex and maternal age, but were also more often overweight/obese, had a parent born out of Sweden and older siblings.

| Associations with prevalent asthma up to age 24 years
Overall associations between antioxidant intake at age 8 years and prevalent asthma up to 24 years are displayed in Figure 1.   models for all participants (n = 2506) and after exclusion of participants with food-related allergic symptoms (to specific fruits or vegetables, n = 242), adjusted for total energy intake, parental education, ethnicity and history of atopic disease, maternal age at delivery, smoking in pregnancy and/or infancy, and older siblings. p interaction between antioxidant intake and sex.  Further adjustment for fruit intake at age 24 years had no major influence on the results (data not shown). No association was observed between long-term vegetable intake and prevalent or incident asthma; however, an inverse association between long-term vegetable intake and remittent asthma was observed (OR 0.60; 95% CI 0.39-0.91).

| Associations with lung function up to age 24 years
We observed no association between intake of fruits or vegetables at age 8 years and lung function up to 24 years (Table 5). However, a higher TAC intake was associated with increased mean FEV 1 -z (T3 vs. T1, +0.11; 95% CI 0.01-0.21, p-trend .036) and a tendency to increased mean FVC-z (+0.09; 95% CI −0.01-0.20, p-trend .072) up to 24 years.
Age-and sex-specific analyses between TAC intake and lung function are shown in Figure 2. In age-specific analyses, associations were statistically significant at ages 8 years (≥median vs. <median, In sex-specific analyses, associations tended to be stronger among males, but the interaction terms were not significant. In analyses of lung function change, we found no interaction between time and TAC, thus no difference in change over time for lung function between the different TAC intake groups (data not shown). In analyses stratified by asthma at age 8 years, there was no significant difference between participants with or without asthma (Table S3). Likewise, no effect modification by aeroallergen sensitisation was observed (data not shown).

| DISCUSS ION
In this study of 2506 young adults from the population-based birth cohort BAMSE, a higher fruit intake at age 8 years was associated with a tendency to reduced odds of prevalent asthma, with reduced odds of incident asthma and increased odds of remittent asthma up to age 24 years. In contrast, no association was observed with vegetable intake. The cumulative action of dietary antioxidants, measured as a higher TAC of the diet, was associated with reduced odds of prevalent asthma and increased mean FEV 1 . In sex-specific analyses, associations between fruit and TAC intake were more pronounced among males. However, due to the large number of associations tested, results should be interpreted with caution. Note: OR (95% CI): odds ratio (95% confidence interval). Median intakes (<median reference): fruits 1.38 times/day, vegetables 2.32 times/day and TAC 10,009 μmol TE/day. Generalized estimating equations (GEE) models for all participants and after exclusion of participants with food-related allergic symptoms (to specific fruits or vegetables, n = 242), adjusted for sex, total energy intake, parental education, ethnicity and history of atopic disease, maternal age at delivery, smoking in pregnancy and/or infancy, and older siblings.

TA B L E 4
Overall associations between antioxidant intake at age 8 years and incidence and remission of asthma up to 24 years To our knowledge, this is the first prospective study to investigate the association between antioxidant intake in childhood and asthma and lung function up to young adulthood. Several epidemiological studies in children [26][27][28][29] and adults 30-32 have demonstrated a reduced risk of asthma in relation to a high fruit and vegetable intake, but most have been cross-sectional, thus limiting causal inference and being unable to address incidence and remission of disease. The results from our study are in agreement with previous studies that a potential association seems to be stronger for fruits compared to vegetables. 4 Fruits contain a complex mixture of bioactive compounds with antioxidant and immunomodulatory properties, such as vitamins, minerals and polyphenols, that are able to enhance endogenous antioxidant systems, scavenge free radicals produced in the inflammation process and protect the airways against endogenous and exogenous sources of oxidative stress. 5 Additionally, fruits are rich in fibres, which by inducing changes in the gut microbiome, might impact on immune responses and lung disease. 33 Therefore, the additive and synergistic effects of the different nutrients present in fruits may be responsible for their potent antioxidant activity. 34 Higher diversity of vegetable consumption has recently been associated with decreased risk of asthma in school-aged children. 35 Potential contamination of vegetables by pollutants, such as cadmium, which have been associated with increased risk of asthma, could contribute to the absence of association with vegetables in the study 36,37 ; however, this question was not possible to address with our data. The inverse association observed between long-term vegetable intake and remittent asthma could be due to chance, as it was not consistent with the analyses of intake at 8 years.
TAC reflects the sum of dietary antioxidant intake, taking synergistic effects into account, and correlates with plasma total antioxidant capacity. 38 Consistent with lung function trajectories up to adulthood, 39 we observed relatively stable coefficients for the association of high TAC intake and increased mean FEV 1 and FVC across ages, which suggest general effects on lung growth rather than on airway obstruction specifically. A previous study from the BAMSE cohort on TAC intake in school age has shown improved lung function development up to adolescence among children with asthma. 15 In the present study, we observed no significant difference among participants with and without asthma, which may be explained by the longer duration between exposure and outcome and the lower number of participants compared to the previous study. In this study, we did not observe associations between fruit or vegetable  In the present study, associations between fruit intake and asthma, and TAC intake and FEV 1 and FVC were more pronounced in males compared to females. Males had a lower dietary antioxidant intake at both ages 8 and 16 years. Sex differences in the same direction as in our study have previously been described, suggesting that oxidative stress may be associated with airflow limitation in males. 48,49 In addition, sex hormones may influence the development of asthma in females. 50 At younger ages asthma is more prevalent among boys, whose lung function development continues up to young adulthood, whereas females reach the peak lung function earlier in life and in general have a later onset of asthma. 3,51 Furthermore, effect modification by aeroallergen sensitization status for fruit intake and prevalent asthma has been observed in our study, and male sex has been associated with increased risk of Ig-E sensitization to airborne allergens up to young adulthood. 16 Diet modifications due to allergy to fruits and vegetables, and cross-reactivity between some fruit and vegetable and pollen allergens, might affect intake and need to be considered in studies of diet and allergic disease. 52 In our study, exclusion of children with foodrelated allergic symptoms attenuated the association between TAC intake and prevalent asthma, indicating that parts of the observed association may be due to disease-related modification of consumption. However, the association between fruit intake and incident asthma remained after the exclusion, suggesting less influence of disease-related modification of consumption on incident disease. An alternative explanation for the attenuated associations is the exclusion of sensitized participants, among whom associations seemed to be stronger. Additionally, asthma definition was based on information collected by questionnaires administered 8 years apart, which might have led to some misclassification of the outcome and over time. Although our analyses were adjusted for several potential confounders, residual confounding cannot be completely ruled out.
In conclusion, results from this longitudinal study suggest that fruit and dietary antioxidant intake in school age may influence asthma and lung function development up to young adulthood, particularly among males. Median intakes of fruits and vegetables at age 8 years in the highest compared to the lowest tertile differed more than threefold. Our results are generally in line with WHO recommendations to consume at least five servings or 400 g per day of fruits and vegetables as part of a balanced diet. Therefore, our findings emphasize the importance of following current F I G U R E 2 Age-and sex-specific associations between TAC intake at age 8 years and lung function up to 24 years. Median TAC intake (<median reference): total 10,016 μmol TE/day, females 10,364 μmol TE/day and males 9546 μmol TE/day. Analyses for all participants (n = 1483, including 775 females and 708 males), obtained by mixed-effects linear regression, adjusted for sex (totals), total energy intake, parental education, ethnicity and history of atopic disease, maternal age at delivery, smoking in pregnancy and/or infancy, and older siblings recommendations for the general population and highlight the need for further studies using repeated dietary assessment and biomarkers of dietary intake from childhood up to young adulthood.

ACK N OWLED G M ENTS
The authors would like to thank the children and parents participating in the BAMSE cohort and all staff involved in the study through the years.

CO N FLI C T O F I NTE R E S T
The authors declare no conflict of interest.

E TH I C A L A PPROVA L S TATEM ENT
The study was approved by the Swedish Ethical Review Authority

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.