Allergic contact dermatitis in 136 children patch tested between 2000 and 2006


Mark Hammonds Mayo Clinic Jacksonville 4500 San Pablo Rd Jacksonville FL 32224, USA


Background  Allergic contact dermatitis is often under-recognized in the pediatric population but it may affect greater than 20% of this age group.

Methods  We conducted a retrospective review of the Mayo Rochester, Jacksonville and Arizona patch test database of all children 18 or younger over a 7-year period (January 1, 2000–December 31, 2006).

Results  One-hundred thirty-six children were patch tested from age 3 to 18. Females constituted 66% of those tested and males 34%. Eighty percent of the children were equally distributed between age groups 11–15 and 16–18, with the remainder being 10 years or younger. Sixty-one percent of the children tested positive to at least one allergen. Fifty-three percent of these reactions were deemed to be of current relevance, 31% questionable relevance, 5% past relevance, and 10% not relevant. Males younger than 10 were most likely to have a positive patch test. However, the percent of positive tests in males decreased with increasing age. Females younger than age 10 were less likely to have a positive test than older females. The most common allergens were nickel, cobalt, gold, and thimerosal.

Conclusion  In children suspected to have allergic contact dermatitis, 61% were confirmed to have a positive reaction to at least one allergen. The utility of patch testing children whose clinical presentation is suggestive for allergic contact dermatitis is high.


Allergic contact dermatitis was once viewed as a disorder of the adult population. Children were thought to be spared due to a lack of exposure to potential allergens and a deficient reactivity of the immune system. Patch testing has also been limited in this population because of technical difficulty owing to small physical size and the belief that irritant reactions dominate in children, leading to extensive false-positive results.

More recently, it has been estimated that 20% of the pediatric population is affected by allergic contact dermatitis and some studies have suggested an even greater number.1 The most common allergens in children have historically been nickel, thimerosal, neomycin, rubber compounds, potassium dichromate, and fragrances.1 In order to further evaluate the prevalence of allergic contact dermatitis and common allergens in children we reviewed our patch testing data over a 7-year period.

Materials and Methods

Approval for this study was obtained from the Mayo Clinic Institutional Review Board. We retrospectively reviewed the results of patch testing in patients 18 and younger during the 7-year study period.

Patch testing

Patients were tested to different series based on clinical presentation. The fewest allergens tested were 25 in two patients and the most were 185. The mean number of antigens tested was 92 and median was 82.


Patients age 0–18 who underwent patch testing for suggested allergic contact dermatitis from January 1, 2000 to December 31, 2006 were identified from a clinical database and medical records. Patients generally were not patch tested if they took immunosuppressive drugs (including oral corticosteroids) or had medical conditions that could compromise the evaluation of skin responsiveness. These medical conditions included connective tissue disease, immunobullous disease, and blistering diseases. In addition, patients who had denied research authorization were excluded from the analysis.

Interpretation of patch test results

The method of patch test application, chambers used, and reading times was the same as that used by the North American Contact Dermatitis Group (NACDG).2,3 Testing was conducted with Finn Chambers, and patches were applied to the patients and removed after 48 h. Reactions were evaluated initially at 48–72 h and again at 96–168 h after patch application. Patch test reactions were interpreted using the following criteria: negative reaction, 1+ (macular erythema), 2+ (papular erythema), 3+ (vesiculation), 4+ (bulla formation or spreading erythema), and irritant reaction. An irritant reaction was defined as a reaction that diminished between the first and second evaluations. Reactions were further interpreted as having current relevance, past relevance, questionable relevance or no relevance by the patch testing physician. We grouped current, past, and questionable relevance together as relevant reactions.

For the purposes of this study, a positive allergic patch test result was defined as 2+ or greater. Assignment of relevance did not play a factor in the definition of a positive reaction in these cases. Macular erythema was also accepted as a positive reaction if it was judged to be relevant. Irritant reactions were excluded.


One hundred thirty six children children were patch tested during the 7-year study period. Positive reactions to at least one allergen were found in 83 children, or 61% of those tested. The mean age of the study group was 14.4 years (range 3.6–18.9). Nineteen percent (n = 26) of children were 3–10 years of age (group A), children ages 11–15 (group B) composed 40% (n = 55), and those aged 16–18 (group C) amounted to 40% (n = 55) of the total population.

Sixty-two percent of children both in groups A and B had at least one positive patch test. Those in group C tested positive to at least one allergen 60% of the time.

Females made up 65% (n = 89) of the group with the remaining 35% (n = 47) being males. Females tested positive 65% of the time whereas only 53% of males had a positive reaction to at least one allergen. Fifty-six percent of females in group A, 69% in group B, and 67% in group C had positive patch tests (Table 1). Seventy-five percent of males in group A, 54% in group B, and only 38% of males in group C had positive patch tests (Table 1). There were no significant differences between age groups although there was a trend in the males to have fewer positive reactions with increasing age (P = 0.18).

Table 1.   Positive reactions by age and gender
Age groupTotal tested (n = 89)Tested positivePercent positive (%)
A (3–10)181056
B (11–15)292069
C (16–18)422867
Age groupTotal tested (n = 47)Tested positivePercent positive (%)
A (3–10) 8 675
B (11–15)261454
C (16–18)13 538
Age groupTotal tested (n = 136)Tested positivePercent positive (%)
A (3–10)261662
B (11–15)553462
C (16–18)553360

A total of 12,499 allergens were tested and there were 247 positive patch tests. Ninety-four percent of patients were tested to our standard series. Other common series tested were rubber, shoe, steroid, cosmetics, plastics/glue, and oral flavoring/preservative.

The fewest allergens tested were 25 in one patient and the most were 185 in one. This patient was tested to the standard, plastics/glue, rubber, shoe, and steroid series for foot dermatitis. Seven patients were tested to fewer than 40 allergens. Three of these were tested specifically to the metal series. In two of these, there was a history of metal allergy and the patch test was performed in a preoperative fashion for an orthopedic surgeon and a dentist. Two patients were tested only to the oral series. One child was tested to the rubber series for suspected allergic contact dermatitis to her eyeglass nosepiece. A 3-year-old child with suspected adhesive tape allergy was tested to a limited series of 25 allergens due to her small size.

The number of allergens to which an individual tested positive ranged from 1 to 16, with a mean of 3.0 allergens. Fifty-three percent of positive patch tests were scored as current relevance, 31% questionable relevance, 5% past relevance, and 10% not relevant. Twenty-nine percent of the positive tests were 1+ reactions with current, questionable, or past relevance.

Nickel was the most common offender, positive in 22% (87% relevant) of those tested (Table 2). Seventeen percent of children reacted to cobalt (87% relevant), 10% to gold (86% relevant), 7% to thimerosal (70% relevant) and 7% to benzalkonium chloride (100% relevant). Six percent had positive patch tests to fragrance mix, neomycin, and potassium dichromate. Those reactions of 2 + or greater scored as “not relevant” and included as a positive reaction in the database were evaluated. Thimerosal was the only allergen in the aforementioned group with a considerably lower proportion of positive reactions deemed relevant compared to the collective database. A positive reaction for thimerosal was graded relevant in only 70% of cases, compared to 89% relevance for positive reactions in the overall study. The breakdown of relevance for other antigens is noted in Table 2.

Table 2.  Most common allergens
AllergenTestedPositivePercent Positive (%)*Percent Relevant (%)
  • *

    Percent positive calculated as positive reactions/136.

  • Reactions scored as current, past, or questionable relevance.

Nickel1353022 87
Cobalt1362317 87
Gold1191410 86
Thimerosal 5910 7 70
Benzalkonium Chloride126 9 7100
Fragrance Mix128 8 6100
Neomycin127 8 6100
Potassium Dichromate129 8 6 88

The prevalence of positive tests to specific allergens among age groups and genders was calculated (Table 3). Females consistently tested positive more frequently than males in all allergens except fragrance mix; however, the overall difference between genders was not significant (P = 0.20). Nickel allergy was more common in the youngest patients and gold allergy in female patients. Dermatophagoides mix was the most common irritant and was therefore discontinued as a component of our patch testing series.

Table 3.  Allergen percent positive by age and gender
AllergenGroup A (3–10) (%)Group B (11–15) (%)Group C (16–18) (%)Female (%) (n = 89)Male (%) (n = 47)Total (Relevant*)
  • *

    Reactions scored as current, past, or questionable relevance.

Nickel352018251722% (87%)
Cobalt191320181517% (87%)
Gold12 71315 210% (86%)
Thimerosal 8 7 7 9 4 7% (70%)
Benzalkonium Chloride 4 411 7 6 7% (100%)
Fragrance Mix 8 7 8 4 8 6% (100%)
Neomycin 8 5 5 7 4 6% (100%)
Potassium Dichromate 4 5 7 7 4 6% (88%)


Allergic contact dermatitis in children is more common than was once recognized. A recent study of an Italian pediatric population by Seidenari et al.4 found the prevalence to be 52.1% in a limited patch series of 1094 children. This rise in allergic contact dermatitis is thought to be due to an increase in patch testing of the pediatric population, greater awareness in pediatricians and dermatologists, and more chemical exposures than in the past.1 In our study, 61% of children had a positive patch test. This figure is higher than most studies have noted in the past. This may be explained by the expanded patch series utilized with a mean of 92 chemicals tested per patient. An increased clinical suspicion for allergic contact dermatitis in those patients selected for testing may also be responsible for the elevated numbers. Another factor in our study was accepting macular erythema as a positive test if the clinician deemed it to be relevant. This represented 29% of all positive patch tests. In previous studies, macular erythema was not consistently regarded as a positive test. Indeed, we recognize that assignment of relevance in our population is difficult and controversial since some of our patients had limited follow-up, being referred specifically for patch testing.

It is also commonly accepted that allergic contact dermatitis increases with age as a result of the opportunity for more exposures. However, the aforementioned Italian study found the highest prevalence of sensitization in those below 3 years of age. This is consistent with other previous works.5,6 In our patch series there was no significant difference between age groups with A, B, and C having 62%, 62%, and 60% positive, respectively. Considering vaccinations, piercing, topical medicaments, and cosmetics in younger patients, this age group has many potential exposures. When evaluating prevalence by gender, a trend emerged with males having a higher prevalence in the younger groups which decreased with age (Fig. 1). Younger females had a lower prevalence, but this leveled out in groups B and C. None of these differences were significant yet the trends were unexpected. This may be attributable to the small numbers in group A, with males only making up eight patients and six of these testing positive.

Figure 1.

Percent positive by age and gender

The most common positive allergens in our study were nickel, cobalt, gold, thimerosal, benzalkonium chloride, fragrance mix, neomycin, and potassium dichromate. These findings are consistent with other previous studies. Nickel and cobalt made up the majority of positive tests, with a total prevalence of 39% combined. This was not surprising as they are common cosensitizers. Although thimerosal was found to be a frequently positive allergen, and our officially documented relevance was 70%, we question its relevance to the patients’ dermatitis. Since positive patch test reactions to thimerosal likely reflect that the patient was exposed to immunizations preserved with this allergen, we have discontinued its use in our standard series.


Allergic contact dermatitis is common in the pediatric population, with 61% testing positive to at least one allergen in the current study. Positive reactions were considered to have current, questionable, or past relevance in 89%. Nickel (87% relevant), cobalt (87% relevant), gold (86% relevant), and thimerosal (70% relevant) were the most frequent allergens in which there was a positive reaction. Females had a higher prevalence of positive patch tests; however, the subset of males age 3–10 had the highest overall prevalence. Given that our relevant positive patch test reaction rates are considerably higher than in previous pediatric studies, we believe that patch testing should be utilized more frequently as a valuable diagnostic tool in the pediatric setting when the clinical presentation is suggestive of allergic contact dermatitis.