Histomorphology and Immunophenotype of Eczematous Skin Lesions Revisited—Skin Biopsies Are Not Reliable in Differentiating Allergic Contact Dermatitis, Irritant Contact Dermatitis, and Atopic Dermatitis

Abstract

Lesions of allergic contact dermatitis (ACD), irritant contact dermatitis (ICD), and atopic dermatitis (AD) share similar clinical features and thus, their diagnosis can be challenging. The aim of this study was to reassess histopathology and immunophenotyping properties to distinguish between ACD, ICD, and AD. Charts of patients with eczema, who had undergone complete routine diagnostic workup (skin biopsies, patch tests, skin prick tests, and respectively or serum IgE levels), were reviewed. Thirty-five skin biopsy specimens of 28 patients (mean age 64 ± 15 years; ♀ = 13 ♂ = 15) with clear diagnosis of ACD (n = 15), ICD (n = 6), or AD (n = 14) were analyzed. Histomorphological and immunohistochemical (CD3, CD4, CD8, CD11c, CD34, CD123, S100, and IL-17) parameters were evaluated using Kruskal-Wallis test, Wilcoxon test, Fisher exact test, and decision tree analysis. Eosinophils were statistically significant (P = 0.0184), more often observed in AD than in ACD or ICD. No other statistically significant differences were found with regard to epidermal patterns, patterns of dermal infiltrates, or immunophenotyping. Using predictive modeling approaches, dermal eosinophils were found to be associated with AD, necrotic epidermal keratinocytes with ICD, and a focal type of parakeratosis with ACD. As an additional finding, pseudo-Pautrier microabscesses, which were present in the skin of 2 AD and 2 ACD patients, contained myeloid dendritic cells (CD11c). Differentiation of ACD, ICD, and AD should be based on clinical features and results of allergy tests. Histopathology does not reliably differentiate between ACD, ICD, and AD, but helps to exclude psoriasis, tinea, or T-cell lymphoma.