Atopic dermatitis: correlation of distinct risk factors with age of onset in adulthood compared to childhood.

BACKGROUND: Atopic dermatitis (AD) has long been regarded as a primarily pediatric disease. However, there is growing evidence for a high rate of adult-onset AD. We aimed to characterize factors associated with adult-onset versus (vs.) childhood-onset AD and controls.

METHODS: We analyzed cross-sectional data of the CK-CARE-ProRaD cohorts Bonn, Augsburg, Davos, Zürich of 736 adult patients stratified by age of AD onset (childhood-onset <18 years: 76.4% (subsets: 0-2; ≥2-6; ≥7-11; ≥12-18); adult-onset ≥18 years: 23.6% (subsets: ≥18-40; ≥41-60; ≥61) and 167 controls (91 atopic, 76 non-atopic).

RESULTS: We identified active smoking to be associated with adult-onset AD vs. controls (adjusted Odds Ratio (aOR)= 5.54 [95% Confidence Interval:1.06-29.01] vs. controlsnon-atopic , aOR= 4.03 [1.20-13.45] vs. controlsatopic ). Conjunctivitis showed a negative association vs. controlsatopic (aOR= 0.36 [0.14-0.91]). Food allergy (aOR= 2.93 [1.44-5.96]), maternal food allergy (aOR= 9.43 [1.10-80.95]), palmar hyperlinearity (aOR= 2.11 [1.05-4.25]) and academic background (aOR=2.14 [1.00-4.54]) increased the odds of childhood-onset AD vs. controlsatopic . Shared AD-associated factors were maternal AD (4-34x), increased IgE (2-20x), atopic stigmata (2-3x) with varying effect sizes depending on AD-onset and control group. Patients with adult-compared to childhood-onset had doubled odds of allergic rhinitis (aOR= 2.15 [1.12-4.13]), but reduced odds to feature multiple (3-4) atopic comorbidities (aOR= 0.34 [0.14-0.84]). Adult-onset AD, particularly onset ≥ 61 years, grouped mainly in clusters with low contributions of personal and familial atopy and high frequencies of physical inactivity, childhood-onset AD, particularly infant-onset, mainly in "high-atopic"-clusters.

CONCLUSIONS: The identified associated factors suggest partly varying endo- and exogeneous mechanisms underlying adult-onset vs. childhood-onset AD. Our findings might contribute to better assessment of the individual risk to develop AD throughout life and encourage prevention by non-smoking and physical activity as modifiable lifestyle factors.