Atopic dermatitis (AD) is the most common chronic inflammatory skin disease, affecting 15%–20% of children and 1%–3% of adults.1 Hand eczema (HE) is one of the most characteristic—yet nonspecific—manifestations of adult AD. A history of AD is one of the main risk factors for HE. Nevertheless, HE is also frequent in the general population, and we might expect it to affect up to 20% of adults during their lifetime; in fact, up to 10% report compatible symptoms within the previous 12 months.2,3 A recent systematic review and meta-analysis revealed that a previous history of AD was associated with an increased prevalence of DM compared with baseline, 1 year, and the patient's lifetime.4 As for pathogenesis, some experimental studies have shown that increased transepidermal water loss in patients with atopy leads to greater penetration and absorption of irritants and allergens than in patients with healthy skin.5,6 AD has been identified as a risk factor for HE in patients aged <30 years,7 although its influence decreases with age.8,9

Few studies have compared the clinical characteristics of atopic HE patients with those of nonatopic HE patients. In particular, data for large, representative populations from different geographical and occupational settings should be sought.

The objective of this study was to compare the epidemiological, clinical, and allergy profile of patients with AD and HE who undergo patch testing with that of nonatopic HE patients.

The study was based on data from the Spanish Contact Dermatitis Registry (Registro Español de Dermatitis de Contacto [REIDAC]), a prospective Spanish multicenter registry (https://dermatitis.contacto.aedv.es/), for the period January 2018 to June 2020. We recorded data for all patients who underwent patch testing for HE and compared those with and without AD.

REIDAC is a centralized registry developed by the Spanish Contact Dermatitis and Skin Allergy Research Group (Grupo Español de Investigación en Dermatitis de Contacto y Alergia Cutánea [GEIDAC]), together with the Research Department of the Healthy Skin Foundation (Fundación Piel Sana) and the Spanish Academy of Dermatology and Venereology (Academia Española de Dermatología y Venereología [AEDV]), which brings together the main contact dermatitis departments in Spain. The epidemiological, clinical, and allergy-related variables of patients who underwent patch testing at the participating centers were recorded in the registry. The methodology and data collection have been described elsewhere.10 In this study, we recorded sex, age, other sites affected, association with occupational factors, occupation, and duration of symptoms. The patient was considered to fulfill the criteria for AD if he/she had ever been diagnosed with AD or fulfilled the criteria at the time of consultation for predominantly flexural eczema, together with other atopy-related data. Patients with only minor criteria of AD such as hyperlinear palms, keratosis pilaris, xerosis, or chronic cheilitis were not included in this diagnosis.11 The tests were performed according to the recommendations of the European Society of Contact Dermatitis.12 In REIDAC, the allergy workup is performed using the Spanish standard series and complementary series at the clinician's discretion.13 This study only took into account data from the GEIDAC standard series. Only patients with lesions affecting the hands as the main site were considered to have HE, although patients with areas other than the hands affected (secondary involvement) could also be included. The main causal factor of a picture of HE (allergic dermatitis, irritant dermatitis, or AD) was based on clinical criteria. Quantitative variables were analyzed using the t test; qualitative variables were analyzed using the χ2 test. The statistical analysis was performed using Stata Statistical Software: Release 16 (Stata Corp. 2015).

We included 1466 patients who had undergone patch testing for HE. Of these, 326 (22%) had a history of AD. The MOAHLFA index for the group as a whole was as follows: male, 32%; occupational, 49%; atopy, 22%; hand, 100%; leg, 3%; face, 5%; age >40, 62%.

Of the 1466 patients included, 996 were women (68% [72% of those with atopy] and 470 were men (32% [28% of those with atopy]) (P=.0931).

Table 1 shows the epidemiological characteristics of patients with HE.

Patients with AD were younger, had a longer mean duration of symptoms before the patch tests. The occupational history was less commonly considered the trigger than in nonatopic patients (38% vs. 53%).

By age, AD was more common in students (OR, 4.29 [2.69–6.86]) and less common in retirees (OR, 0.49 [0.27–0.84]).

The occupational history of patients who underwent patch testing is shown in Table 2. Most patients with HE and AD (81%) were actively employed, although occupational triggers were not as important as in the nonatopic patients. By profession, the only one for which differences between the groups was found was hairdressing, where patients with HE were mainly nonatopic (85%) (P=.0317). No differences were found between atopic and nonatopic patients for the remaining professions.

Table 3 shows the complete allergen series and the positive results. The most frequent allergens from the standard series in patients with HE were nickel sulfate (20%), methylchloroisothiazolinone/methylisothiazolinone (MCI/MI) (11%), cobalt chloride (5%), potassium dichromate (4%), fragrance mixes I and II (4%), and formaldehyde (4%). Results were positive with ethylenediamine more frequently in atopic patients with HE (3.56 [0.94–13.39]), whereas results were more frequently positive to nickel sulfate (0.70 [0.51–0.95]) and carba mix (0.30 [0.08–0.83]) in nonatopic patients with HE.

The main diagnoses after patch testing in atopic patients with HE were, in order, AD alone (147/326 [45% of the total]), allergic contact dermatitis, and irritant contact dermatitis (Table 4). The other diagnoses included psoriasis and dyshidrotic eczema.

Our study showed that HE was a frequent reason for consultation among atopic patients. This association has been reported elsewhere.4 In fact, 1152 patients from the general registry had a history of atopy; therefore, patients with HE accounted for 28% of all atopic patients. Compared with the nonatopic patient, the profile of the atopic patient with HE was that of a younger person with a longer duration of symptoms. This profile is coherent with the endogenous nature of AD, which is often present from childhood, with eczematous lesions at various sites, including the hands.6,14 In addition, the longer mean duration of symptoms before patch testing in this population could be explained by the fact that a diagnosis of AD itself is a plausible explanation for HE that often rules out the indication for patch testing.

The distribution by age is consistent with that of AD, with more students and, albeit to a lesser extent, retirees than among the nonatopic patients. By profession, the only differences were found for hairdressing, with a lower percentage of HE in atopic patients, compared with dermatitis that was mainly allergic or irritant. This seemingly paradoxical observation could be due to the bias arising from the nature of the REIDAC, which does not usually include patients with occupational skin diseases. However, some authors have reported an occupational history associated with hairdressing and HE where no differences were found for sensitization to allergens among patients with and without a history of AD.15 Working as a hairdresser is a standard indication for patch testing and an independent factor for positive patch test results with present relevance independently of a history of AD.16 Furthermore, having a history of AD from childhood, especially the severe forms, might be expected to affect a person's occupational history and choice of a particular profession,17 thus modifying the results.

The most frequent allergens in atopic patients with HE—nickel, MCI/MI, cobalt, potassium dichromate, fragrances, and formaldehyde—are generally the most commonly reported allergens in HE in various studies.18–22 The difference with respect to sensitivity to nickel in both atopic and nonatopic patients is noteworthy. The greater or lesser sensitization to nickel in atopic patients is controversial. While some authors report mutations in the filaggrin gene to be responsible for greater sensitization to nickel,23,24 others conclude that the frequency of sensitization to nickel and other allergens is similar to or even lower in atopic than in nonatopic patients.18,25 It is important to remember that the present study only took into account allergens included in the GEIDAC standard series. It is reasonable to assume that these findings could vary with the increasing frequency of other allergens, such as acrylates, which are especially present in some occupational settings.

Although the profile of positive allergy results was similar in both atopic and nonatopic patients, the interpretation of the relevance and the final diagnosis varied. Therefore, a final diagnosis of allergic contact dermatitis or irritant hand dermatitis was less common in atopic patients, probably because, in a certain percentage, the dermatologist thought that the endogenous dermatosis itself accounted for the symptoms, with some resistance toward explaining that a positive patch test result could be playing a role beyond that of the trigger. Thus, a final diagnosis of AD was the most frequent (45% of all the cases studied).

Dyshidrosis, which was diagnosed in some cases, was more frequent in nonatopic patients, consistent with its nonspecific nature as hand involvement associated with various types of eczema, also including, albeit nonexclusively, AD.26

Our study is subject to a series of limitations. The first is that of selection bias, since we only included patients whose reason for attending the clinic was to undergo patch testing. Atopic patients may be overrepresented, since they often undergo patch testing more frequently, even when the main suspected diagnosis is endogenous eczema, especially in adults. However, patch testing could also be delayed in some patients with AD, since a history of AD has been considered the main reason for the cutaneous symptoms. Furthermore, given that the participating centers belong to the Spanish National Health System, many working patients are excluded because they were assessed by their occupational health insurer, thus affecting the results for occupational exposure. Finally, we only took into account allergens included in the standard GEIDAC series. While this points to a degree of homogeneity between centers, it may limit the identification of emerging allergens such as acrylate and linalool in the study groups.

In conclusion, this multicenter study from Spain revealed that patients who undergo patch testing for HE commonly have a history of AD. Among HE patients as a whole, atopic patients are younger, with a mean duration of dermatitis that is greater than in nonatopic patients, and, apparently, with no broad variations between most of the professions studied. Overall, the sensitization profile was also similar to that observed in patients with HE. These findings could vary with the emergence of new allergens.

REIDAC received funding from Sanofi, GlaxoSmithKline, and Novartis in the form of unconditional donations. The funders did not participate in the study design or analysis or in the review of the article

The authors declare that they have no conflicts of interest.