Serum levels of Vitamin A and Atopic Rhinoconjunctivitis in Swedish adolescents

Aim: Vitamin A plays a role in mucosal immunity and tolerance, but the association between vitamin A status and allergy is still unclear. The aim of the study was to analyze the levels of vitamin A in serum from adolescents with or without atopic rhinoconjunctivitis. Method: Thirteen-year-old children with atopic rhinoconjunctivitis (n = 53) and non-allergic, nonsensitized controls (n = 52) were randomly selected from a population based prospective birth cohort comprising 1228 children in Northern Sweden born in 1996-1997. Vitamin A (retinol) concentrations in serum were measured with high performance liquid chromatography mass spectrometry. Multiple logistic regression was used to evaluate the association between allergy prevalence and serum vitamin A levels. Results: Multiple logistic regression analysis showed no association between serum vitamin A levels and atopic rhinoconjunctivitis prevalence; OR = 1.00 (95% confi dence interval 1.00-1.00), p = 0.81. Stratifi cation for gender revealed a trend for a higher risk for having atopic rhinoconjunctivitis with higher concentrations of vitamin A in serum for females, OR = 1.02 (1.00-1.05), p = 0.07. No such associations were found in male subjects OR = 0.99 (0.97-1.01), p = 0.15. A dose-response relationship between allergy and vitamin A concentrations were also calculated but no such relationships were found, neither for all subjects nor for male and females separately. Conclusions: Serum levels of vitamin A could neither be positively nor negatively associated with atopic rhinoconjunctivitis in Swedish teenagers. Research Article Serum levels of Vitamin A and Atopic Rhinoconjunctivitis in Swedish adolescents Malin Barman1,2*, Nils-Gunnar Carlsson1, Anna Sandin3, Agnes E. Wold4 and Ann-Sofi e Sandberg1 1Food and Nutrition Science, Department of Biology and Biological Engineering, Chalmers University of Technology, Göteborg, Sweden 2Department of Obstetrics and Gynecology, Sahlgrenska University Hospital, Gothenburg, Sweden 3Paediatrics, Department of Clinical Sciences, Umeå University, Umeå, Sweden 4Institute of Biomedicine, Department of Infectious Diseases, University of Gothenburg, Göteborg, Sweden Dates: Received: 31 August, 2017; Accepted: 10 September, 2017; Published: 12 September, 2017 *Corresponding author: Malin Barman, Chalmers University of Technology, Department of Biology and Biological Engineering, Food and Nutrition Science, SE-412 96 Göteborg, Sweden, Tel: +46(0)31-7723811; Fax: +46(0)31-7723830; E-mail:


Introduction
Allergy is the most prevalent chronic disease in affl uent countries with a Western lifestyle, and its prevalence has increased globally. Decreased exposure to environmental microbes [1,2], and changes in dietary patterns, including an increased intake of margarine and oils [3], are different factors that may explain the rise in the prevalence of allergy.
Vitamin A (retinol) is a fat-soluble vitamin that is important for vision, gene expression and tissue differentiation. Vitamin A defi ciency has profound effects on the immune system [4], and is associated with increased mortality from infectious disease in developing countries [5]. Animal studies show that vitamin A-defi cient animals have reduced mucosal immune responses [6], and a reduced capacity to develop oral tolerance [7]. A particular subset of antigen-presenting dendritic cells (DC), the CD103+DC subset, possesses the vitamin A converting enzyme RALDH [8], and is central in mucosal immunity and tolerance [7]. These cells traffi c the intestinal mucosa, where they pick up antigens, migrate to the mesenteric lymph nodes and present the antigens to naïve T cells traffi cking the lymph nodes. Retinoic acid (a metabolite of vitamin A) produced by the CD103+DCs is needed to imprint a gut-migrating phenotype on the naïve T cells [8], and to convert naïve T cells to regulatory T cells [9]. Retinoic acid has been shown both to promote Th2 cell differentiation by increasing the ratio of Th2 cytokines relative to Th1 cytokines [10], and to downregulate Th2 immune responses by inhibiting IL-6 driven induction of Th17 cells and increasing the proliferation of T-regulatory cells [11,12]. in children [13]. Similar results were reported in another metaanalysis from 2010, which revealed lower serum levels of vitamin A in children with asthma compared with controls [14].
The aim of the present study was to analyze the association between vitamin A in serum and allergy in Swedish adolescents selected from a prospective and population based mother and child birth cohort. Allergy has many diagnoses and nuances with different, and sometimes transient, age infl uenced symptoms in combination with environmental changes, and is best observed in longitudinally prospective cohorts. In the present study, we had the opportunity to select non-allergic children with no sensitization or reported allergic symptoms neither at 1, 4 nor 13 years of age. For a well-defi ned group of children with allergy at 13 years of age (time point for retinol analyses) we chose to study allergic rhinoconjunctivitis, defi ned as coherently reported symptoms and sensitization to airborne allergens.

Study subjects
Cases with atopic rhinoconjunctivitis (n = 53) and nonsensitized non-allergic controls (n = 52) were selected at 13 years of age from a population based birth cohort consisting of 1,228 children born during one year between February 1996 and January 1997 at the Östersund Hospital in Northern Sweden [15]. The children were followed with skin prick tests and questionnaires regarding allergic symptoms at 1, 4 and 13 years of age. At 13 years of age, 789 adolescents responded to the questionnaire (together with a parent) and participated in sensitization test. The subjects were skin prick tested with standardized extracts for ten allergens, including milk, egg, fi sh, wheat, soy, cat, dog, horse, timothy grass, and birch (ALK, Hørsholm, Denmark), as earlier described [16,17].

Allergy diagnosis
Atopic rhinoconjunctivitis was defi ned as a positive skin prick test to one or more inhalant allergens (cat, dog, horse, birch or timothy), in combination with a positive answer to the question "Have you had any allergic symptoms from eyes and/or nose in contact with pollen or furred animals during the last 12 months?". Hence, all subjects in the allergic group had symptoms in contact with pollen or furred animals in combination with a positive skin prick test against an airborne allergen. In total, 174 of the subjects in the cohort fulfi lled the criteria for atopic rhinoconjunctivitis. Non-allergic, nonsensitised controls were defi ned as having neither allergic symptoms nor any positive reaction in the skin prick test at 13 years of age or in any of the previous follow-ups (n = 331).

Selection of subjects
Fifty-three subjects who had atopic rhinoconjunctivitis were choosen randomly; of these, 19 subjects also had asthma, 4 subjects also had food allergy and 30 subjects had only rhinoconjunctivitis. As controls we selected a similar number of subjects that had been non-allergic and non-sensitised in all follow-ups at 1, 4, and 13 years of age.

Collection of serum
Venous blood (10 mL) was drawn, allowed to clot and centrifuged. Serum was separated, aliquoted and frozen within three hours. The serum samples were stored for a maximum of fi ve years at -80 o C before analyses.

Analysis of vitamin A in serum
Serum vitamin A concentrations were measured using liquid chromatography mass spectrometry (LC-MS) simultaneously as 25-hydroxy vitamin D. The results from the vitamin D analysis has been reported previously [18]. Samples were prepared for the analysis according to Turpeinen et al. [19], with some modifi cations as earlier described in detail [18]. Briefl y, 200 μl serum was mixed with 150 μl methanol:iso-propanol (80:20), containing 100 ng of retinoic acid as internal standard (Sigma Aldrich, Saint Louis, USA). After extraction twice with 2 ml hexane, the organic phases were evaporated and dissolved in 100 μl of methanol and transferred to vials. Retinol and retinoic acid were analyzed using an LC-MS system (Agilent 1260 Infi nity Binary LC and Agilent 6120 Quadrupole LC/MS, Agilent Technologies, Santa Clara, California, USA). The LC-MS setup has previously been described in detail [18]. The instrument was operating in selected-ion monitoring mode: 269 for both retinol and retinoic acid. Retinol and retinoic acid were identifi ed based on retention time.
A control serum sample was extracted and analyzed in every run. Study samples and control samples were prepared and analyzed in duplicates. Relative standard deviation (%RSD) for intra-assay precision was 4.7% for a serum sample with a mean retinol concentration of 116 μMol/L. Relative standard deviation was 7.5% for between day variability. Retinol concentrations decreased from 116 μMol/L to 111 μMol/L after 24 hours in room temperature. After three freeze-thaw cycles, i.e. thawed in room temperature and frozen for 24 hours, retinol concentration was down to 99 μMol/L.

Statistical analysis
For analysis of differences in background variables (Table  1), chi-square test was used. Mann-Whitney U test was used to compare group means of serum vitamin A concentrations due to non-parametric data ( Figure 1). To analyse the trend for dose-response in fi gure 2 chi-square for linearity was used. Multiple logistic regression analysis was used to analyze the association between allergy and serum vitamin A levels, taking the following potential confounders into consideration: gender, siblings, breastfeeding, maternal allergy, paternal allergy and residence of school (Table 2). Statistical analyses were performed using IBM SPSS Statistics version 19 (IBM Corporation, New York) and a two-tailed p-value ≤0.05 was considered signifi cant.

Ethical considerations
The study was approved by the local ethical committee in Umeå, Sweden (Dnr 09-110M) and was conducted according to the Declaration of Helsinki. At 13 years of age, the adolescents verbally approved their participation in skin prick tests and Participation was voluntary and the adolescents were informed that they were free to decline participation at any time, without stating the reason and that their decision would not lead to any disadvantage.

Characteristics
The characteristics of the individuals selected from each group as well as the non-selected individuals, are shown in Table 1. Statistical analyses were performed to compare both selected with non-selected subjects as well as to compare cases and controls. For differences between selected and non-selected subjects in the two groups, a larger percentage of selected non-allergic adolescents attended schools in the only major city of the region, Östersund, compared to the non-selected non-allergic adolescents (Table 1). No signifi cant differences were found between selected and non-selected cases. For the comparisons between cases and controls: subjects with atopic rhinoconjunctivitis were more often males and had more often mothers with a history of allergy than selected non-allergic controlsA;also paternal allergic heredity tended to be higher in cases than in controls (Table 1).

Differences in serum vitamin A concentrations
Vitamin A was extracted and quantifi ed in serum samples of 105 individuals representing the cases (n = 53) and controls (n = 52). The mean (SD) serum concentration of vitamin A was 176 Defi ned as a positive skin prick test to an inhalant allergen, in combination with a positive answer to the question: "Have you had any signs of pollen allergy or allergy to furry pets during the last 12 months?" 2 Pearsons Chi-Square test 3 Exclusively breast feeding at 4-month-of-age. 4 Defi ned as an affi rmative answer to the question "Has the childs mother/father ever had asthma, hay fever or eczema?"; answered at 13-years-of-age. 5 Answered at four years of age. 6 Recidens of school, city of Östersund or in the more urban surroundings, at 13 years of age. ns: p > 0.20

Multiple logistic regressions
Multiple logistic regression was used to analyse the association between serum vitamin A concentration and allergy diagnosis at age 13, either as crude odds ratio, or adjusted for gender, siblings, breastfeeding, maternal allergy, paternal allergy and residence of school. Serum levels of vitamin A was not a risk factor for being allergic at 13 years of age in neither the crude nor the adjusted model ( Table 2). The logistic regression models were also performed separately for males and females. For females, there was a tendency for a higher risk for atopic rhinoconjunctivitis for higher vitamin A levels in serum. This association was not found for male subjects ( Table   2).
Of the 53 selected subjects with atopic rhinoconjunctivitis, 19 subjects had both asthma and rhinoconjunctivitis. The logistic regression analyses were performed stratifi ed for 'having asthma' and 'not having asthma'. No signifi cant difference in serum vitamin A levels was found between cases with or without asthma and controls (data not shown).

Dose-response relationship
To analyze if there was a dose-dependent difference between vitamin A and allergy, all subjects were divided in four groups depending on vitamin A status. Proportions of allergic subjects in each vitamin A group are displayed in Figure 2.
We found no signifi cant difference in prevalence of allergy between the four groups stratifi ed according to serum vitamin A concentration (Chi-square for linearity, Figure 2).

Discussion
Serum levels of vitamin A were measured in 105 adolescents at the age of 13 years, living in the County of Jämtland in Northern Sweden. The subjects were followed from birth onwards with regular skin prick tests and questionnaires probing for allergic symptoms. From a well-defi ned group of children with allergy diagnosis based on questionnaire and skin prick test, we randomly selected cases with atopic rhinoconjunctivitis, i.e. subjects with a positive prick test to one or more airborne allergen (cat, dog, horse, birch or timothy) together with symptoms from the eye or nose in contact with furred animals or pollen. Moreover, we randomly selected non-sensitized, non-allergic controls, i.e. subjects without sensitization and symptoms at 13 years of age or in any of the previous follow-ups. No differences in serum vitamin A levels were found between adolescents with atopic rhinoconjunctivitis and healthy controls of the same age, in neither the crude nor the adjusted logistic regression model.
Since the distribution of the two genders were different in the cases and the controls, with more males in the cases than in the controls, the logistic regression models were also performed separately for females and males. For females only, vitamin A tended to be associated with an increased risk for having atopic rhinoconjunctivitis. This was not shown for males separately.
Hence, it seems likely that there might be a gender difference in the association between vitamin A and allergy. When the mean values of vitamin A in cases and controls was visualized with a bar chart, a slightly higher mean vitamin A level were seen in female cases than in female controls, however this difference was not statistically signifi cant.

Previous studies on vitamin A in serum and allergy have
shown mainly inverse associations [13,14,[20][21][22][23][24], but also no associations [25][26][27][28][29], between vitamin A in serum and asthma. Most of the previous studies have been focusing on asthma which has mainly been diagnosed as recurrent cough and/or wheezing, sometimes together with previous demonstration of improvement of symptoms in response to asthma medication. In young children, asthma is often diagnosed as wheezing, which may also be a sign of respiratory infection. Furthermore, asthma symptoms in adults are often triggered by viral infections.
As this birth cohort used questionnaires for symptoms and did not have any objective method i.e. lung function test, we considered the diagnostic tools not sharp enough for a clear allergic asthma diagnose. Instead we chose to study allergic rhinoconjunctivitis, defi ned as coherently reported symptoms We have previously shown that neither serum levels of 25-hydroxy vitamin D nor proportions of long chain polyunsaturated fatty acids (LCPUFAs) in serum were associated with allergy in the same study population cohort in a cross-sectional analysis of the same cohort at 13 years of age [17,18]. On the other hand, we found that higher proportions of LCPUFAs in the cord blood of both the n-3 and n-6 families, were associated with being allergic at 13 years of age in this cohort [16]. This suggests that perinatal exposure to fatty acids Allergy was assessed at fi ve, eleven and 20 years of age [22].
Overall, the study found no association between serum vitamin A levels and atopic disease. The only signifi cant relation was an inverse association between vitamin A status at 2 months of age and atopic allergy at 20 years of age [22]. In a recent study, serum carotenoids analysed at various time points during early childhood were related to parental reported asthma at fi ve years of age [31]. Carotenoids have antioxidant capacity and are also precursors for vitamin A. The study found no associations between serum carotenoids early in life and the risk of asthma at fi ve years of age [31].

Conclusion
We found no association between levels of vitamin A in serum and atopic rhinoconjunctivitis at 13 years of age in adolescents, selected from a population-based cohort followed from birth onwards and with well-defi ned allergic manifestations.