Food allergy concurrent with atopic dermatitis: Correlation with age of onset and severity

Abstract

Background:

Data on the frequency of food sensitivity (FS) and food allergy (FA) in patients with atopic dermatitis (AD) differ among studies.

Objectives:

The aim of this study was to determine the frequency of FS and FA in different AD phenotypes according to the age of onset and severity of AD. In addition, we aimed to investigate the risk of FA in these patients.

Methods:

Patients diagnosed with AD between 2022 and 2024 were included in the study. All patients with AD admitted during this period were analyzed for the coexistence of FA and FS.

Results:

The study included 257 children with AD. Of these patients, 147 of 257 were girls (57.2%). The median (interquartile range [IQR]) age of onset of AD was 6 months (2.5–30 months). By the age of AD onset, FS and FA were present in 60.3% and 32.5%, respectively, in patients with moderate-to-severe AD, with disease onset between ages 0 and 3 months. Among the patients with disease onset between ages 4 and 11 months, the corresponding rates in moderate-to-severe cases were 59.6% for FS and 17.4% for FA. FS was present in 39.3% of moderate-to-severe cases with AD onset age after 12 months, but none had FA. Being in the moderate-to-severe category for AD increased the risk of FA 14–16 times compared with the mild AD group.

Conclusion:

FS is significantly more prevalent than FA in patients with AD. In children in whom FS test results are positive, the diagnosis of AD-FA coexistence should not be made without performing an elimination diet and oral food challenge test. This approach will help prevent unnecessary food elimination.

Atopic dermatitis (AD), also referred to as atopic eczema, is a chronic inflammatory skin condition characterized by severe itching and recurring eczematous lesions.^1,2 The prevalence is reported to be 15–20% in children.³ However, the frequency of AD in patients with food allergy (FA) is relatively high and AD is also a risk factor for the development of FA.⁴ Early-onset AD, and increased severity and chronicity are associated with a higher risk of FA.⁵ The prevalence of FA that accompanies AD varies between different studies. Studies have reported that the prevalence of FA varies between 14.8% and 60%.^3,6–12 İn addition, other studies have shown that food sensitization (FS) is 48.4%–66%.^8,13,14 There are very few prospective studies^8,9 that have investigated the incidence of FA and FS based on AD onset age and severity in pediatric patients with AD.

In the present study, the aim was to determine the rates of FS and FA in children according to different AD onset age and severity, and to determine the risk factors in the FA-related group with a prospective study design.

METHODS

This study included patients ages 0–18 years and with AD who attended Kocaeli University Medical Faculty Allergy and Immunology outpatient clinic between January 2022 and December 2024. The study was designed as a cross-sectional study. The patients were prospectively enrolled and evaluated by using the same questionnaires and same panel of tests. However, the onset time of AD was retrospectively documented based on information obtained from the parents.

This study was approved by the Clinical Trials Ethics Committee, Kocaeli University Medical Faculty with the protocol number E-80418770-730.99-287092. Written informed consent was obtained from all the participants (from parents and/or legal guardians for children ages 0–18 years). The data collection and de-identification process were carried out in accordance with the protocol approved by the ethics committee. All procedures were conducted in accordance with the ethics standards of the Declaration of Helsinki and relevant guidelines.

Patients and Data Acquisition

Patients with physician-diagnosed AD according to the Hanifin Rajka criteria were prospectively evaluated.¹⁵ Those with immunodeficiency, malignancy, autoimmune, chronic disease, and those >18 years of age were excluded from the study. During admission, all the participants were administered a standard questionnaire that collected demographic data such as age of AD onset, gender, mode of delivery, diet in the first 6 months of life, and age of introduction to complementary feeding.

Groups by AD Onset Age and Severity

According to the age of onset of AD, patients were divided into four subgroups: ages 0–3 months, 4–11 months, 12–72 months, and >72 months. The severity of eczema was determined according to the SCORing Atopic Dermatitis (SCORAD) index, and patients were divided into groups according to disease severity. Patients with a SCORAD score <25 were included in the mild AD group, patients with a score between 25 and 50 were included in the moderate AD group, and patients with a score >50 were included in the severe AD group, as previously described.^6,16

Determination of Serum Specific Immunoglobulin E and Skin-Prick Test

Patients were analyzed for concentrations of specific immunoglobulin E (spIgE) antibody titers to food allergens (milk, egg, hazelnut, peanut, walnut, wheat, beef, chicken meat) determined by fluorescence enzyme immunoassay by using the Pharmacia CAP-systems (Pharmacia Diagnostics, Uppsala, Sweden). Those with an spIgE level >0.35 kU/L were considered positive.^7,8 In addition, the eosinophil count and the percentage and total IgE levels were measured in all the participants. The skin-prick test (SPT) was carried out to detect IgE-mediated sensitization for food allergen extracts, including milk, egg, hazelnut, peanut, walnut, wheat, beef, chicken meat (ALK-Abelló, Horsholm, Denmark). In addition, SPTs were performed to detect aeroallergens sensitization. Negative (saline solution) and positive (histamine) controls were also used. The size of the skin response was calculated as the mean of the longest vertical and horizontal diameters of the indurations measured 15 minutes after the application of the SPT. A response at least > 3 mm of the negative control was considered positive.¹⁷

Oral Food Challenges

Standardized oral food challenges (OFC) were performed in children according to the European Academy of Allergy and Clinical Immunology (EAACI) guideline.¹⁸ All children ages <2 years and with AD and with positive food-spIgE and/or SPT results underwent OFC after a 4-week elimination diet. In patients >24 months of age and with moderate-to-severe AD, those who did not respond to conventional treatment and who had positive food-sIgE and/or SPT results also underwent OFC after a 4-week elimination diet. In patients in whom OFCs were postponed due to illness or other circumstances, the interval was extended to 6 weeks.

In addition, challenges were performed in children with an spIgE positive result who would take the food for the first time. Examination and follow-up before and after OFC were performed by the same physician (IO) who had received allergy training. Patients with at least one of the typical IgE-mediated symptoms, such as urticaria, angioedema, hypotension, tachycardia, bronchospasm, rhinitis, and cardiovascular collapse within 2 hours, were considered as early type. Patients with an increase of >10 points in the SCORAD index within 6–72 hours were described as late type and mixed type in patients with both early and late reactions.^19–21

Statistical Analysis

All statistical analyses were performed by using IBM SPSS for Windows, version 29.0 (IBM Corp., Armonk, NY). The Kolmogorov-Smirnov and Shapiro-Wilk tests were used to assess the normality assumption. Continuous variables are presented as median and interquartile range (IQR). Categorical variables are summarized as counts and percentages. Comparisons between groups were carried out by using independent samples to test for normally distributed variables and the Mann-Whitney U test for non-normally distributed variables. Associations between categorical variables were examined by using the χ² test. Multivariable analysis was performed by binary logistic regression analysis.

RESULTS

Demographic Characteristics of Patients

Of 257 patients, 147 were girls (57.2%). Because the data of the patients were documented at the time of admission, the age of admission and the age at data collection were the same. The median (IQR) age at data collection was 32.5 months (12–66 months). Similarly, the median (IQR) age at admission was 32.5 months (12–66 months). The median (IQR) age at onset of AD was 6 months (2.5–30 months). The age of onset of AD was within the first 3 months in 95 of 257 (37%), between 4 and 11 months in 70 (27.2%), between 12 and 72 months in 71 (27.6%), and, in 21 (8.2%), it was after 72 months. In terms of the SCORAD index, 66 of 257 (25.7%) had mild AD, 146 (56.8%) had moderate AD, and 43 (16.7%) had severe AD. Furthermore, 33 of 257 (12.8%) had asthma and 72 of 257 (28%) had allergic rhinitis (Table 1).

Table 1.

Characteristics of the study population

IQR = Interquartile range; AD = atopic dermatitis; SCORAD = SCORing Atopic Dermatitis; IgE = immunoglobulin E.

Reactions with Oral Provocation

Of the 38 patients diagnosed with FA, 33 (86.8%) were diagnosed by oral provocation, whereas 5 (13.2%) were diagnosed based on clinical history and positive laboratory test results. Anaphylaxis developed in four of the patients diagnosed with FA by OFC. Of the 42 reactions proven by oral provocation, 19 of 42 (45%) were early-type reactions, 16 of 42 (38%) were late-type reactions, and 7 of 42 (16.6%) were mixed-type reactions. Types of reactions that developed with oral provocation and doses of food allergen are given in Supplemental File 1.

In the five patients diagnosed with lgE-mediated FA by clinical history and laboratory, no oral provocations were performed due to contraindications or expected severe symptoms. Among these five patients diagnosed without oral provocation, four had IgE-mediated early-type reactions (urticaria and/or angioedema) within the past 6 months, and one had a history of anaphylaxis within the past year.

Prevalence of FAs and FS

FA was present in 38 of 257 (14.8%) and FS was present in 115 of 257 of all patients with AD (44.7%). By the age of onset of AD grouping in patients with moderate-to-severe AD, the rates of FS and FA were 60.3% and 32.5% in the 0–3 months group, 59.6% and 17.2% in the 4–11 months group, respectively. FS was present in 39.3% of patients in moderate-to-severe AD cases > 1 year of age, but none had FA (Fig. 1). For patients with mild AD, the rates of FS were 44.4%, 20%, and 20% in the AD onset groups 0–3, 4–11, and >12 months, respectively. There was only one patient with FA in the 0–3 months mild AD onset group This patient was diagnosed with the development of lgE-mediated early-type reaction (urticaria/angioedema) after eating an egg during the complementary feeding period. FA was not diagnosed in any of the patients with mild-AD cases with onset from 4 months of age upward. The frequency of FA according to onset age of and severity of AD is shown in Table 2.

Figure 1.

Frequency of FA and FS in patients with moderate-to-severe AD according to age of onset AD. FA = Food allergy; FS = food sensitivity; AD = atopic dermatitis.

Table 2.

Frequency of food allergy according to onset age and severity of AD

AD = Atopic dermatitis; FA = Food allergy.

Thirty patients (78.9%) had one FA, five patients (13.2%) had two FAs, and three patients (7.9%) had three FAs. Egg allergy was detected in 25 (54%), cow milk allergy in 12 (26%), and allergy to tree nuts in 9 (20%) of the patients with FA. Only one patient (2.8%) had peanut allergy. This case was diagnosed based on clinical history and splgE positivity. Of the patients with FS, 84 (38%) had sensitization to egg, 42 (19.5%) to cow's milk, 44 (20.2%) to tree nuts, 25 (11.4%) to peanut, 25 (11.4%) to wheat, 17 (7.8%) to beef, and 8 (3.7%) to chicken meat.

Risk Factors Related to FA in AD

İn multivariate logistic regression analysis, moderate AD increased the risk of FA compared with mild AD 14-fold (odds ratio [OR] 14.08 [95% confidence interval {CI}, 1.77–112.28]; p = 0.013], and severe AD increased the risk of FA 16-fold compared with mild AD (OR 16.81 [95% CI, 1.96–144.07]; p = 0.01). We found that there was an inverse relationship between the age of onset of AD and the development of FA (OR 0.83 [95% CI, 0.72–0.95]; p = 0.011), and that there was an association between an increase in eosinophil count and FA (OR 1.00 [95% CI, 1.000–1.002]; p = 0.036), but there was no correlation between FA and total lgE level (Table 3).

Table 3.

Univariant and multivariant associated risk analysis in patients with AD with food allergy^*

AD = Atopic dermatitis; OR = odds ratio; CI = confidence interval; ref. = reference; IgE = immunoglobulin E.

^*Statistical logistic regression analysis was used for multivariant and univariant risk analysis for food allergy.

The age of onset of AD, SCORAD score, and laboratory values in the AD groups with and without FA are shown in Table 4.

Table 4.

Age of onset, SCORAD and laboratory evaluations in AD groups with and without FA

SCORAD = SCORing Atopic Dermatitis; AD = atopic dermatitis; FA = food allergy; IQR = interquartile range; IgE = immunoglobulin E; spIgE = specific IgE; SPT = skin-prick test.

*Mann Whitney U test.

DISCUSSION

This study provides information on the frequencies of FS and FA, together with risk factors in pediatric patients with different AD phenotypes, stratified by onset age and severity. In the whole AD cohort, the prevalence of FA and FS were 14.8% and 44.7%, respectively. In patients with moderate-to-severe AD, in the earliest onset age group (0–3 months), the prevalence of FS and FA were 60.3% and 32.5%, and, in the 4–11 months onset group, FS prevalence remained at 59.6%, whereas FA decreased to 17.4%. FS was detected in 39.3% of patients in moderate-to-severe cases > 1 year of age, whereas FA was not detected in any patient. The FA risk was found to be 14–16 times higher in the moderate-to-severe group compared with the mild group, and no significant difference was detected between the moderate and severe groups in terms of FA.

İn pediatric patients, AD-FA coexistence has been the subject of many studies but the results of these studies vary. These variable results may have been because of heterogeneous populations in terms of age groups and disease severity, and methodological differences. Although AD-FA coexistence was reported at rates of between 30% and 60% in studies that only included patients with moderate-to-severe AD^{8,11,13,22,23}; the rate dropped to 14–17% in studies in which all the pediatric patients with AD were investigated, regardless of severity, and food elimination and challenge were performed.^9,10 Similar to these earlier studies, we found AD-FA coexistence in only 14.8% of all the patients with AD.

Many studies have shown that an early age of onset (in the first 2 years) is an important risk factor for the association of AD and FA.^8,10,13,24 However, the results (39.2–59%) show wide variability because the evaluation interval of this early-onset age may be as simple as in “the first 2 years of age”.^8,13,25 There are few studies that had been conducted with smaller age group^8,10 ranges according to the time of the onset and the severity of AD. Gray et al.⁸ reported that, in patients with moderate-to-severe AD, AD-FA coexistence was seen in 66% of cases in the first 6 months of life, 28% in ages 6–12 months, and only in 3 patients after ages 12 months. Similarly, Mailhol et al.¹⁰ reported that 69.6% of patients with moderate-to-severe AD with AD-FA were seen before 3 months of age, 24.6% between 3 and 12 months, and only 5.8% (four patients) after 12 months of age. In the present study, according to the age of onset of AD, we also detected 71.1% of patients with FA in the first 3 months and 28.9% in 4–11 months. FA was not detected in any of our patients whose age of onset of AD was >11 months. The results of our study¹⁰ and other studies stratified by severity and age of onset of AD show that three fourths of AD-FA coexistence occurred in patients with moderate-to-severe cases with onset in the first 3 months. Consequently, we observed that patients with moderate-to-severe AD with onset in the first 3 months of life constituted the highest risk group for AD-FA coexistence.

Unnecessary elimination diets begin with unnecessary tests. Pepper²⁶ emphasized that unnecessary food testing can lead to unnecessary dietary restrictions and various subsequent adverse outcomes. The Atopic Dermatitis (Eczema) Guidelines 2023²⁷ states that unnecessary eliminations lead to the development of IgE-mediated FAs. Although sensitization is observed in approximately one in two patients with AD, it is known that the frequency of allergy is lower. The prevalence of FS among children with AD in the general population has been reported to range from 47% to 66% in various studies.^5,9,13,14 Gray et al.⁸ reported sensitization rates of 86% in the first 6 months, 67% between 6 and 12 months, and 42% > 12 months in their study of patients with moderate-to-severe AD with a risk of FA. In our study, we identified sensitization in 60.3% and allergy in 32.5% of patients with moderate-to-severe AD ages 0–3 months, whereas, in the 4–11 months age group, sensitization was detected in 59.6% and allergy in 17.2% of patients with moderate-to-severe disease. Among patients ages >12 months and with moderate-to-severe AD, sensitization was identified in 39.3% of the patients, whereas no clinical allergies were detected. The findings demonstrated that the rate of sensitization without clinical allergy was remarkably high in patients with AD. In addition, the avoidance of unnecessary elimination diets remains crucial in preventing adverse clinical outcomes, which encompasses immunologic tolerance deterioration, compromised nutritional status, and subsequent growth and developmental impairment.

In general, although studies to date show that patients with moderate-to-severe AD are at more risk for AD-FA coexistence than are patients with mild AD,^{11,21,25,28,29} there are some studies that show that only being in the severe AD group was a risk factor.^8,10,13 Our study showed that the moderate-to-severe group increased the risk of FA 14–16 times compared with the mild group and that there was no difference between severe and moderate in terms of FA risk.

The present study showed that FA coexistence was present in one of the two patients with moderate-to-severe AD and FS in the first 3 months of life, and in only one of the three patients between 4 and 11 months. We did not detect FA in any patient >1 year of age. These results may be explained by the dual allergen exposure hypothesis. It has been suggested that the timing and balance of cutaneous and oral exposure determine whether a child develops allergy or tolerance to foods³⁰ This hypothesis suggests that, when food protein first enters the immune system through the skin and is not sufficiently encountered by the gastrointestinal immune system, it may result in FA, whereas FA may be prevented when AD develops after a given food protein is first encountered by the gastrointestinal tract and tolerance is induced. Therefore, in the first 3 months, an inadequate oral exposure to food proteins in the infant may result in a lack of stimulation that would induce tolerance in the gastrointestinal lymphoid tissues. Complementary feeding, which is often started between 4 and 6 months of life, becomes the main form of nutrition for infants until they are 1 year old, which enables the activation of tolerance mechanisms for food proteins taken orally. This may explain why the AD-FA coexistence is more common in the first months of life.

This study's strengths comprise the inclusion of all pediatric age groups with AD and the standardized conducting of elimination and challenge tests by a single allergy-trained physician (IO). In addition, there was a reasonable number of participants. Among our limitations are the fact that the patients in our cases belong to a tertiary health-care institution and the potential for recall bias with regard to the age of AD onset in older children. Another limitation was the lack of data with regard to the effect of an AD-FA association on the natural course of AD in the long term.

CONCLUSION

This study demonstrated that patients with moderate-to-severe AD, particularly those with onset in the first 3 months, were at higher risk for the coexistence of AD and FA. In addition, the moderate-to-severe AD group increased the risk of FA by 14-16–fold compared with the mild AD group. FS is significantly more prevalent than FA in patients with AD. Therefore, a decision for FA should not be made without food elimination-challenge tests. This approach will help prevent unnecessary food elimination.