Contact allergy in children with and without atopic dermatitis: An Italian multicentre study

Abstract Background Contact allergy and atopic dermatitis (AD) are both common inflammatory T cell‐mediated diseases and many factors may influence the prevalence of contact allergy in AD patients. In children, their possible correlation was debated with conflicting results. Objectives The present study aimed to assess the prevalence of contact sensitivity in children and to investigate the association with AD. Materials and methods A retrospective multicentre study on children aged from 0 to 14 years patch tested between January 2017 and December 2018 was performed. Children were consecutively patch tested with the SIDAPA (Società Italiana Dermatologia Allergologica Professionale Ambientale) baseline series. Results Among the 432 children investigated for contact allergy, 125 (28.9%) showed a positive reaction to at least one of the allergens tested, with a higher prevalence of positive patch test reactions in girls (32.3%) than in boys (25.0%). The most frequent contact allergens were nickel sulphate (10.2%), cobalt chloride (6.7%), methylisothiazolinone (3.7%), fragrance mix‐2 (3.2%), potassium dichromate (2.8%), fragrance mix‐1 (2.1%) and methylchloroisothiazolinone/methylisothiazolinone (2.1%). One‐hundred‐three children (23.8%) suffered from AD showing a higher prevalence of positive patch test (36.9%) compared to children without AD (26.4%). Conclusions Despite the topic being still controversial, the present study suggests a consistent prevalence of contact allergy among children with higher sensitivity rate among children with AD than without AD.


| INTRODUCTION
In the past, contact allergy was considered rare and probably underestimated in children due to the immaturity of the childhood immune system and the low frequency of exposure to contact sensitizers in paediatric population. 1 In the last decade, few large-scale studies on childhood contact allergy published in Europe and North America showed that contact sensitization in children is more common than previously thought with rates of sensitization ranging from 36.2% to 62.3%. [2][3][4][5] Sensitization to contact allergens can occur as early as infancy 6 and patch testing is the gold standard to diagnose contact allergy in children. 2,4,6,7 The prevalence of childhood contact allergy is influenced by several factors (new fashion in body piercing, use of personal care products, sports, and hobbies) and the most frequent sources of contact allergy in children are metals (nickel sulphate, potassium dichromate, cobalt chloride), fragrances, topical antibiotics (neomycin sulphate and bacitracin), emollients and emulsifiers (propylene glycol), and surfactants (cocamidopropyl betaine). 1 Besides metals, the most frequent contact allergens in all ages, contact allergens in children vary according to age: neomycin sulphate, methylchloroisothiazolinone/methylisothiazolinone (MCI/MI) and lanolin alcohols in 1-5 years old children, neomycin sulphate, Myroxylon pereira, and fragrance mix-1 in 6-12 years old children, and p-phenylenediamine, fragrance mix-1, and MCI/MI in 13-16 years old children. 4 Nowadays, the role of atopic dermatitis (AD) as a favouring factor for contact allergy is debated and conflicting data have been reported in literature with a high prevalence range (from 27.0% to 95.6%) depending on study designs. [8][9][10][11][12][13][14][15][16][17][18][19][20][21][22] In the past, murine and human models suggested that AD could be protective against contact allergy 23,24 since a prevalent Th2 response may lead to a relative cell-mediated immune deficiency. 25 Recent literature data have demonstrated increased risk of contact allergy in patients with AD due to multiple factors, such as a constitutionally reduced skin barrier function, 26 also damaged by the frequent use of irritant chemicals, 27 the continuous local use of emollients and anti-inflammatory ointments with potential sensitizing properties, 28 and a reduced heterogeneity of the AD skin microbiome. 29,30 Moreover, the relationship between contact allergy and AD seems to be even more complex as different immune pathways (Th1, Th2, and even Th17 mediated ones) may be shared by both entities. 31 The most frequently reported contact allergens in AD are metals (nickel sulphate, cobalt chloride, and potassium dichromate), lanolin alcohol, neomycin sulphate, formaldehyde, sesquiterpene lactone mix, Compositae mix, and fragrances. 10,[12][13][14][15][16][17][18][19][20] Considering that also some "hypoallergenic" personal care products can contain powerful contact allergens, 28,32 lanolin and fragrances were recently reported as the most common allergens in AD children by European 22 and North American 33 researchers.
In this multicentric retrospective study, we analysed the prevalence of contact sensitivity in children aged from 0 to 14 years undergoing patch testing for eczematous dermatitis, also highlighting the possible correlations with gender and atopic dermatitis.  35 Being the study retrospective, it was not possible to establish the relevance of all positive patch test results. Patch test results were analysed according to five age groups (0 to 3, ≥3 to <6, ≥6 to <9, ≥9 to <12, ≥12 to <15 years) and the presence of AD at the time of testing. The diagnosis of atopic dermatitis was made according to Hanifin and Rajka criteria. 36 The study protocol was approved by the ethics committees of the participating centres. Signed informed consent was obtained from patients' parents. Differences of paired discrete data were tested by Fisher's exact test and were used to analyse categorical variables. All statistical analyses were performed using IBM-SPSS version 26.0 (IBM Corp., 2019) and using R software, version 4.0.3. In all analyses, a two-sided p value ≤0.05 with Bonferroni correction was considered significant. According to age groups, the highest prevalence (37.0%) of positive patch test results was observed in the oldest age group (12-14 years), with a decreasing trend in the 9-11 and 6-8 years age groups (31.9% and 27.7%, respectively), while the lowest prevalence (20.7%) was reported in the 3-5 years age group. The higher prevalence of positive patch test reactions in girls than in boys was confirmed in all of the age groups, except for the 6-8 years age group, where this prevalence was higher in boys than in girls (15.4% vs. 12.3%).
Face, hands, arms, and body folds were more frequently involved in children with AD than in those without AD (

| DISCUSSION
In the paediatric population, the prevalence of contact allergy is difficult to precise and the reported sensitivity rate in children largely T A B L E 2 Positive contact allergens according to gender and atopic dermatitis  with T-cell inflammation and the T-helper 2 cell-mediated pathways that worse damage of epidermal barrier. 39 Literature data changed during the last decades, suggesting an increasing role of AD as risk factor for developing contact allergy. In fact, studies conducted until 2010 mostly showed a lower prevalence of contact allergy in children with AD. 1 Studies performed in the following years reported an increasing prevalence in children with AD and recently, a higher prevalence of contact allergy in AD children population was documented, probably due to increased attention by dermatologists in AD diagnosis, a more frequent patch testing in refractory AD to investigate contact allergic component as potential aggravating factor, and increased use of cleansing and moisturizing products specifically formulated for AD 2 ( Table 3).
The most frequent sensitizers were metals (nickel sulphate, cobalt chloride, potassium dichromate), covering 45.9% of all 185 positive patch test reactions, followed by fragrances (14.6%) and isothiazolinones (13.5%). According to the current literature 1,2,4 and disappointing the 2001 EU Nickel Directive, 40 our results confirmed nickel sulphate as the most frequent contact allergen (10.2%), especially in girls (12.1%). This is probably due to the still wide diffusion of nickelcontaining products from non-EU countries, 2,38 such as jewellery, toys, and electronics. Cobalt chloride, almost always as nickel sulphate cosensitivity (93.1%), 2 is the second most common contact allergen (6.7%) being children exposed to metal-plated products, crayons, and deodorants. 31 We observed a prevalence of fragrance allergy similar to that of recent studies, 2,4 even if in our study a higher sensitivity rate for fragrance mix-2 (3.2%) than fragrance mix-1 (2.1%) was documented. The greater role of new fragrances than old fragrances as contact sensitizers in children, also recently observed by others, 41 confirms the necessity to periodically re-evaluate the fragrance mix composition according to EU cosmetic legislation. 42 Moreover, the significant difference of positive patch test reactions to fragrance mix-1 between children with and without AD is controversial in literature, confirming the findings of previous studies 22,43 and differing from others. 44 Our data seem to be confirmed in adults with and without AD, although with a lower difference. 44 Regarding isothiazolinones, MI prevalence (3.7%) was F I G U R E 1 Sites involved in 125 children with at least one positive patch test reaction according to presence or absence of atopic dermatitis.
considerably and surprisingly higher than MCI/MI prevalence (2.1%), confirming that also in children, it is important to separately test MI at higher concentration to avoid false-negative results. 45,46 Patch test concentration of MI (0.2%) was recently confirmed in children, 2 demonstrating that the high prevalence observed by us is probably due to exposure to other than personal care products. The latter were regulated in 2014 by the European Commission Scientific Committee on Consumer Safety that banned MCI/MI from leave-on products, allowing it in rinse-off products not exceeding 1.5 ppm. 47 In Italy, besides cosmetics, children are still exposed to declared and undeclared MI, such as toys, glue, slime, water-based paint. 48 Considering the eight most frequent positive allergens, all showed higher prevalence in children with than without AD, and in particular for fragrance mix-

CONFLICTS OF INTEREST
The authors declare that there are no conflicts of interest.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. This study included also 91 children with unknown atopic dermatitis status.