Introduction

The house dust mites are a major cause of allergic diseases throughout the world [27]. The most frequently responsible mites are Dermatophagoides pteronyssinus and Dermatophagoides farinae which produce efficient allergens able to induce sensitization and clinical disease [38]. The spectrum of allergic manifestations includes atopic dermatitis (AD) [35], which is particularly frequent in children, rhinitis and asthma [13, 26], and, very rarely, anaphylaxis [10]. The knowledge of the mechanisms underlying the kind of sensitization and the type of clinical expression is far from exhaustive. It was long believed that the load of mite allergens which a subject is exposed to could be an essential factor in determining simple sensitization or clinical allergy. However, recent data have indicated that this relationship is not linear, as the highest exposure to dust mites was not directly associated with the risk of allergic outcome [34]. Instead, the mite species seems to have an influence on the type of allergy which subjects develop: In a large group of around 1,700 children, respiratory allergy was associated with both D. pteronyssinus and D. farinae exposure, while only the latter one was associated with AD [19].

As regards allergy testing, the skin prick test (SPT) or the measurement of specific IgE antibodies in serum is used to indicate sensitization, but they must be combined with the subject’s case history to diagnose clinical allergy. The atopy patch test (APT), which consists of applying the suspected allergen to the skin by the same method of patch testing used for contact dermatitis, was introduced as a technique to evaluate sensitization to aeroallergens in subjects with AD [29]. It was later confirmed as a valid test, in particular for dust mites [4, 5, 14, 31, 36]. It is certainly of interest that recent studies have found that the APT can be the only positive test in patients with AD and also with respiratory allergy [6, 11].

We investigated the role of SPT, IgE measurement, and APT in mite allergy by analyzing the relationship between their results and the clinical expression in subjects with different allergic manifestations.

Materials and methods

Patients

From subjects consecutively referred to the Pediatric Allergy Service in San Severo (Foggia, Italy) for AD, 399 patients (239 males and 160 females) aged 0.4–17.6 years (mean ± SD 6.5 ± 4.0; median 5.8) were included in the study. They were referred mostly by pediatricians but also by general practitioners, as they had a history of AD or respiratory symptoms (rhinitis, asthma) with uncertain etiology. The area where the patients lived includes 61 municipalities and 690,992 inhabitants, about 150,000 of whom are in pediatric age; in this area, there are 90 family pediatricians. To be included, the patients had to have a negative history for food allergy and negative skin prick tests to food extracts. The patients who took part in the study were divided into three groups: group A, current AD (40 patients); group B, current AD with respiratory symptoms (156 patients); and group C, past AD with respiratory symptom (203 patients). Sixty-six patients with rhinosinusitis or asthma but without AD (40 males and 26 females) aged 2.8–13.6 years (mean ± SD 6.6 ± 2.8; median 6.5) examined during the same period served as the control group.

Methods

AD was diagnosed according to the criteria set down by Hanifin and Rajka [15], and rhinosinusitis and asthma were diagnosed on the basis of the consensus document Allergic Rhinitis and Its Impact on Asthma [2].

All patients underwent SPT with inhalant allergen extracts (Stallergénes, Antony, France) with test positivity assessed according to the guidelines from the European Academy of Allergology and Clinical Immunology [9] and to APT using house mite extract (Merck, Milan, Italy). In particular, the material used for testing was composed of inert dust mite bodies purified to 20% (D. Pteronyssinus and D. Farinae 1:1), with white vaseline 20% and white mineral oil as excipients. The substance to be tested was applied onto intact skin of the lower back and held firmly in position using adhesive patch tests made up of aluminum Finn chambers 8 mm in diameter, with an area of 50 mm2 and a volume of about 20 μl. The application period was 48 h. The test was read no less than 30 min after removal to avoid margin effect. Results were interpreted according to the American Academy of Dermatology for APT [17], with a scale ranging from 1+ (weak reaction) to 3+ (very strong reaction). Only reactions of 2 and 3+ were considered positive for the purpose of the study.

Specific IgE to dust mites was measured by ImmunoCap (Phadia, Uppsala, Sweden).

Statistical analysis

Data are presented as percentages and as mean ± SD for qualitative and quantitative variables, respectively. Statistical analysis was performed using χ 2 test for categorical variables and analysis of variance (ANOVA) and Bonferroni multiple comparison tests for continuous variable. Contingency tables were used to analyze the association between the categorical variables (sex and positive family history) and the tests used in the study (APT, SPT, and radioallergosorbent test (RAST)) and to perform the univariate analysis of the positivities to APT, SPT, and RAST in the four different groups. Determinants of the single disease were evaluated by means of multivariate analysis performed using logistic regression analysis. We performed several stepwise forward logistic regressions used as dependent variables for each analysis: the control group vs the three atopic eczema/dermatitis syndrome (AEDS) groups, the current AEDS with respiratory group vs the current AEDS group and the past AEDS with respiratory symptom group, respectively, and the past AEDS with respiratory symptom group vs the current AEDS group. In each model, the predictor variables used were age as continuous variable and sex, positive family history, APT, SPT, and RAST as categorical variables. Statistical significance was assumed for P < 0.05. All statistical analysis was performed using a software program (BMPD version 2007; BMPD statistical software Inc., Los Angeles, CA, USA).

Results

The characteristics of patients included in the four study groups are shown in Table 1. Group 1 subjects, i.e., patients with current AD, were significantly younger than subjects of the other three groups. The results of APT, SPT, and RAST in the different groups are reported in Table 2; the APT was significantly more frequently positive in groups with current AD (groups A and B) or past AD (group C) than in the control group, while SPT and RAST were significantly more frequently positive in the control group. Dividing the patients into three different age groups (0–2, 2–12, more than 12 years), a strong association between age group and current AD vs past AD could be seen (Table 3).

Table 1 Demographic and baseline characteristics of the different groups
Full size table
Table 2 APT, SPT, and RAST results in the different groups
Full size table
Table 3 Correlation between age categorized in three classes and presence of allergy
Full size table

Tables 4 and 5 show the results of univariate and multivariate analysis, respectively. As far as univariate analysis was concerned, for APT, a number of significant differences were found between the different study groups, and all the differences between each study group and the control group were highly significant, whereas for SPT and RAST, the significant differences were in favor of the control group vs the three study groups. As regards multivariate analysis, for APT, significant differences were found only in the comparison between group A vs group B (odds ratio (OR) 1.55) and between group A vs group C (OR 1.81).

Table 4 Positive results to APT, SPT, and RAST in the different groups, univariate analysis
Full size table
Table 5 Results to APT adjusted for sex and age in the different groups, multivariate analysis
Full size table

Discussion

AD has been thoroughly investigated in recent years in its clinical and pathogenetic aspects [1, 20–23]. Despite its denomination, when population studies are performed, the majority of patients with AD in the community do not have atopic sensitization. In the most recent study, which was carried out on 562 children, sensitization to environmental allergens was estimated by specific IgE measurement at about 25% at the age of 3 years [18].

This value is comparable with the 22% we found for IgE tests to dust mite—which is the prominent allergen source—while no population studies using the APT are as yet available. As regards the form sustained by hypersensitivity, AD fits at best the concept of systemic disease with multi-organ involvement [28], as made evident by the natural history of atopic disease. Atopic disease is characterized by early presentation with AD and subsequent respiratory manifestations with asthma and rhinitis, which is often summarized using the term “atopic march” [33]. A number of factors are implicated in this complex pathophysiologic development, which includes skin barrier dysfunction, the consequent exposure to allergens, the interaction with dendritic cells [23], and the activation of an initially local Th2 but later Th1 response along with a systemic Th2 response inducing isotype switching to IgE synthesis and the involvement of eosinophils [1].

Regarding skin barrier dysfunction, in many examples of recent data, interest has been focused on filaggrin, which consists of filament-associated proteins which are bound to keratin fibers in epidermal cells [30]. It was found that loss-of-function null mutations of the filaggrin gene predispose to AD [24] and that this specific gene is related to sensitization to allergens, to more severe phenotypes of AD [3] and asthma [25]. These mechanisms may explain previous experimental observations in which sensitization by the epicutaneous route leads to localized dermatitis and to bronchial hyperresponsiveness [32].

From a clinical point of view, the best tool to investigate the occurrence of these mechanisms seems to be the ATP, which reveals a reaction with T lymphocyte-mediated allergen-specific immune response [37] rather than referring to just the simple presence in the skin or in the blood of specific IgE, as SPT and RAST do. We recently demonstrated that in patients with asthma or rhinitis, a positive APT to house dust mite was strongly correlated to the presence of current or past AD, while most subjects with respiratory disease but a negative history for AD had instead a positive SPT [12].

In the course of our study, we examined a large population of subjects sensitized to dust mites by dividing them according to their clinical manifestations: current AD, current AD with respiratory symptoms, and past AD with respiratory symptoms, using patients with rhinitis or asthma but with no history of AD as a control group. The differences in rates of positive ATP were highly significant in favor of patients with current or past AD vs control subjects, while concerning SPT and RAST less strong differences were detected in favor of control subjects. The univariate analysis showed for APT significant differences in the rate of positivity in patients with current AD and respiratory symptoms compared with patients with only AD or with past AD and respiratory symptoms and highly significant differences in all single study groups compared with control group. The multivariate analysis including only subjects with current or past AD and adjusted for sex and age showed that APT is a good prognostic factor in the comparison between the group with only AD vs the group with AD and respiratory symptoms and vs the group with past AD and respiratory symptoms. The striking differences in response to APT suggest that different immunologic mechanisms underlie the various manifestations of hypersensitivity to dust mites. It seems conceivable that in subjects with a negative history for AD sensitization occurs by respiratory route and leads to the development of a Th2 pattern of response with ongoing production of specific IgE and consequent positive SPT and RAST. Instead, if mite allergens enter through the skin, as occurs in exposure to common indoor concentrations of the major allergen Der p 1 [16], such entering being facilitated by its proteolytic activity [8] and by the filaggrin-associated skin barrier dysfunction, a different sequence of events is likely to take place. This is ultimately revealed by positive APT and negative SPT and RAST.

Our observations were obtained through the study of a large population of patients with AD referred for allergologic evaluation because of suspected allergic origin of AD or respiratory symptoms, this being a possible bias for the high rate of positive APT we found. In any case, these data need to be confirmed by future studies to gain further importance. If confirmed, the definition of atopy as a genetic predisposition to react to SPT to environmental allergens should be reconsidered, also taking into account recent data on sensitization to arthropods and, in particular, to dust mites [7].

In conclusion, we found that subjects sensitized to dust mites with current or past AD, with or without respiratory symptoms, have a different response to diagnostic tests, which is characterized by a highly significantly more frequent positive APT in respect to subjects with respiratory symptoms but a negative history for AD, who show the common response to SPT and RAST.