SEARCH

SEARCH BY CITATION

Keywords:

  • eczema;
  • food allergy;
  • season of birth;
  • skin barrier;
  • vitamin D

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References

Background

Food allergy (FA) in childhood has been shown to be more prevalent in those born in autumn and winter. The mechanisms of this season-of-birth effect remain unclear, although shortage of vitamin D during infancy has been considered one possible mechanism. The purpose of this study was to investigate the effect of eczema on the season-of-birth effect on FA in infancy.

Methods

A questionnaire survey on the prevalence of allergic diseases was completed by the parents of 14 669 Japanese schoolchildren, aged 7–15 years, in Kyoto City, Japan. Logistic regression models were constructed to compare the prevalence of FA in infancy according to season of birth.

Results

Those born in autumn and winter had a significantly higher prevalence of FA in infancy compared to those born in spring and summer in a multivariate model (4.8% vs 3.6%, P = 0.001). The difference, however, was no longer significant when eczema before 6 months was included as either an additional or only confounding factor. The difference among those with and without eczema before 6 months was further analyzed, and it was found that, in both groups, there was no difference between those born in spring and summer and those born in autumn and winter.

Conclusions

The season-of-birth effect on FA in infancy was significantly affected by the existence of eczema before 6 months in Japanese children. Eczema before 6 months may be the factor directly related to the season-of-birth effect on FA in infancy.

Food allergy (FA) in childhood has recently been shown to be more prevalent in those born in autumn and winter.[1, 2] One possible mechanism for this season-of-birth effect on FA has been attributed to the level of vitamin D, in which the shortage of vitamin D during infancy, due to the lower exposure to sunshine during autumn and winter, may affect the immune system, promoting IgE sensitization.[3-6] Meanwhile, Lack proposed the dual allergen exposure hypothesis, in which epicutaneous exposure to food allergens may induce sensitization and subsequent FA.[7] We previously found, in a large-scaled, questionnaire-based survey, that atopic dermatitis was more prevalent in schoolchildren born in autumn and winter.[8] We speculated that disruption of the skin barrier in infancy due to cold, dry weather may produce suitable conditions for epicutaneous sensitization of allergens to occur and induce allergic inflammation of the skin. Most recently, Keet et al. showed that autumn birth is associated with increased risk of FA only among Caucasian subjects, an ethnic group that is most likely to have seasonal variation in vitamin D during infancy.[9] They also found that a history of eczema was another factor that interacted with the seasonal association with FA. These data prompted us to hypothesize that another mechanism for the higher prevalence of FA in those born in autumn and winter might be the disruption of the skin barrier during these seasons, which enhances epicutaneous food sensitization. In order to investigate this hypothesis, we sought to determine whether there really is a difference in the prevalence of FA in infancy according to the season of birth in Japanese subjects, a non-Caucasian ethnic group, and, if so, whether it is influenced by the existence of eczema in early infancy.

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References

In 2006, a questionnaire survey dealing with the prevalence of allergic diseases was completed by the parents of 14 669 Japanese schoolchildren, aged 7–15 years, in Kyoto City, Japan. The data for 13 179 children, accounting for 89.5% of the total, with known month of birth, were analyzed. The definition of FA has been described previously.[10] Briefly, the parents were asked the following questions: (i) does your child ever have allergic symptoms, such as skin symptoms like hives or respiratory symptoms like cough/wheeze, within 1–2 h after ingesting a particular food; and (ii) does your child avoid particular foods due to these symptoms? Those who answered ‘yes’ to both questions 1 and 2 were regarded as having either a past history of or current immediate-type FA. Of these, those who avoided any of the three major food allergens (eggs, milk, or wheat) at <1 year of age were defined as having FA in infancy. These three foods were chosen because they are the important food allergens during infancy in Japan, and 93.5% of the present subjects who avoided any foods from <1 year of age avoided at least one of these foods.[10] With respect to eczema before 6 months, the parents were asked the following questions: (i) has your child ever had eczema; and, (ii) if so, at what age was it initially observed? Those who answered ‘yes’ to question 1 and who answered age within 6 months after birth to question 2 were regarded as having eczema before 6 months.

The study was designated the Allergic Schoolchildren in Kyoto (ASK) study and was approved by the Ethics Committee of Kyoto University Graduate School of Medicine.

The chi-squared test or Student's t-test was used to compare the background characteristics between those born in March–August (spring and summer) and those born in September–February (autumn and winter). Logistic regression models were constructed to compare the prevalence of FA in infancy and eczema before 6 months according to season of birth. P < 0.05 was considered significant. STATA/IC 10.0 (Stata Corporation, College Station, TX, USA) was used for all analyses.

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References

There was no significant difference in background profile (except day care attendance in infancy) between those born in March–August and those born in September–February (Table 1). Prevalence of eczema before 6 months and FA in infancy according to month of birth are shown in Figure 1. On univariate analysis, those born in September–February had a significantly higher prevalence of both eczema before 6 months and FA in infancy, compared to those born in March–August (Table 2). The differences in the prevalence of these symptoms were still significant in a multivariate model with sex, age, birth order, gestational age, birthweight, nutrition in infancy, day-care attendance in infancy, and family history of allergy as confounding factors. For the prevalence of eczema before 6 months, it remained significant when FA in infancy was further included as an additional confounding factor. In contrast, for the prevalence of FA in infancy, it was no longer significant when eczema before 6 months was included as an additional confounding factor. Moreover, the significance disappeared when eczema before 6 months was included as the only confounding factor. The difference in the prevalence of FA in infancy was further analyzed among those with and without eczema before 6 months (Fig. 2), and it was found that the prevalence was much lower in those without eczema before 6 months than that in those with eczema before 6 months. Moreover, in both groups, there was no difference in the prevalence of FA in infancy between those born in March–August and those born in September–February (P = 0.58 and 0.59 for those with and without eczema before 6 months, respectively).

figure

Figure 1. Prevalence of (a) eczema before 6 months and (b) food allergy in infancy, according to month of birth.

Download figure to PowerPoint

figure

Figure 2. Prevalence of food allergy in infancy in those (a) with and (b) without eczema before 6 months, according to month of birth.

Download figure to PowerPoint

Table 1. Participant background characteristics
 Month of birthP
March–AugustSeptember–February
n66686511 
Age (years), mean ± SD10.0 ± 2.59.7 ± 2.50.80
M/F1.031.040.87
Gestational age (months), mean ± SD39.0 ± 1.939.0 ± 2.00.13
Birthweight (g), mean ± SD3059 ± 4223054 ± 4250.57
Family history of allergy (%)74.373.80.57
Feeding in infancy (%)   
Complete breastfeeding31.329.50.20
Mixed feeding63.064.6
Complete artificial feeding5.75.9
Day care attendance in infancy (%)15.213.10.001
Table 2. Month of birth and prevalence of eczema before 6 months and FA in infancy
 Birth monthPrevalence (%)UnivariateMultivariate
OR (95%CI)POR (95%CI)P
  1. Adjusted for sex, age, birth order, gestational age, birthweight, nutrition in infancy, day care attendance in infancy, family history of allergy. Adjusted for all aforementioned plus FA in infancy. §Adjusted for FA in infancy only. Adjusted for sex, age, birth order, gestational age, birthweight, nutrition in infancy, day care attendance in infancy, family history of allergy. ††Adjusted for all aforementioned plus eczema before 6 months. ‡‡Adjusted for eczema before 6 months only. CI, confidence interval; FA, food allergy; OR, odds ratio.

Eczema before 6 monthsMarch–August7.9Reference Reference 
September–February11.41.51 (1.34–1.69)<0.00011.51 (1.34–1.71)<0.0001
    1.48 (1.30–1.69)<0.0001
    1.47 (1.30–1.67)<0.0001§
FA in infancyMarch–August3.6Reference Reference 
September–February4.81.33 (1.13–1.59)0.0011.35 (1.12–1.61)0.001
    1.10 (0.91–1.33)0.32††
    1.10 (0.91–1.32)0.32‡‡

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References

Overall, these data clearly show that the season-of-birth effect on FA in infancy is significantly affected by the existence of eczema before 6 months. Based on the data, it can be argued that eczema before 6 months, rather than shortage of vitamin D, may be the factor directly related to the season-of-birth effect on FA in infancy. Vitamin D deficiency might induce eczema before 6 months but not FA in infancy, because the season-of-birth effect on FA in infancy was no longer observed after stratifying the participants into those with and without eczema before 6 months.

The present data are in accordance with those of Keet et al., who suggested that both vitamin D and the skin barrier may be implicated in seasonal associations with FA.[9] In their data, seasonal variation was observed only among Caucasian subjects, an ethnic group that is most likely to have seasonal variation in vitamin D.[11] Similar seasonal variation in vitamin D, however, was also observed in a normal Japanese population.[12] In these data, the lowest and highest plasma 25-hydroxyvitamin D levels according to season were 60.0 ± 21.4 nmol/L and 85.4 ± 33.0 nmol/L, respectively, in Caucasian subjects,[11] and 37.8 ± 17.8 nmol/L and 79.0 ± 14.0 nmol/L, respectively, in Japanese subjects.[12] Although direct comparison is difficult, these data suggest that the seasonal variation in vitamin D is similar or even stronger in Japanese subjects. The point of the present data is that, even in such an ethnic group with a high seasonal variation in vitamin D, the season-of-birth effect on FA in infancy was significantly affected by eczema before 6 months, and the effect disappeared after stratifying the participants into those with and without eczema before 6 months. Vitamin D may indirectly affect the season-of-birth effect on FA by affecting the prevalence of eczema before 6 months. Based on the data, it is tempting to speculate that, when born in autumn or winter, enhanced epicutaneous sensitization of food allergens during early infancy through barrier-disrupted skin, whether due to vitamin D deficiency or cold and dry weather, promotes the occurrence of FA in infancy. If that is the case, deliberate and thorough skin care beginning in early infancy, especially for those born in the autumn or spring, may reduce the prevalence of FA in infancy.

The limitation of the present data is that the validity of the definition of FA or eczema was not confirmed by physician diagnosis or any laboratory data because it was a large, population-based, questionnaire survey. Another limitation is the possible recall bias due to the retrospective design of the survey. Thus, further prospective investigation is needed to determine the true mechanisms of the season-of-birth effect on FA in infancy and to find ways to effectively prevent the occurrence of childhood FA.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References

This work was supported by Grants-in-Aid from the Ministry of Education, Culture, Sports, Science, and Technology, Japan.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. References