Distribution of atopic dermatitis lesions in United States adults

Background: The distribution of atopic dermatitis (AD) lesions and its impact on quality of life (QOL) is not well established in the US adult population. Objective: To elucidate the distribution of AD lesions and its impact on QOL in US adults with AD. Methods: A cross-sectional, population-based study of 602 adults was performed. AD was determined using modified UK Diagnostic Criteria, and its lesional distribution was assessed. QOL was assessed using Dermatology Life Quality Index (DLQI). Latent class analysis (LCA) was used to determine distinct phenotypes of AD lesional distribution. Multivariable logistic regression was used to determine the relationship between DLQI and distinct phenotypes. Results: The most common sites of skin lesions were reported to be the popliteal fossae, lower legs, dorsal feet and antecubital fossae. Most persons reported partial (19.0%) or complete (63.0%) symmetry of lesions on the extremities. Lesions on the trunk were significantly more common in blacks and Hispanics. Age ≥ 60 years was associated with significantly lower proportions of active lesions on the face and scalp, and significantly higher proportion of lesions on the buttocks or genitals. LCA identified 5 classes of lesional distribution: 1. lower probabilities of lesions affecting any sites; 2. Higher probability of lesions involving the anterior and posterior neck and trunk; 3. lesions involving the antecubital fossae and upper extremities; 4. lesions involving the arms, posterior hands, genitals and buttocks, and to a lesser extent face, palms and legs; 5. lesions affecting all sites. Class-2 (multivariable logistic regression; adjusted odds ratio [95% confidence interval]: 7.19 [3.21–16.07], class-3 (7.11 [3.20–15.80]), class-4 (6.90 [3.07–15.50]) and class-5 (7.92 [3.54–17.71]) were all significantly associated with higher DLQI scores compared to class 1. Conclusion: AD is associated with heterogeneous distribution of AD lesions, and distinct phenotypes that are associated with QOL impact.