Corresponding Author: J Dejewska, Department of Dermatology, Central Clinical Hospital of Military Institute of Medicine, Warsaw, Szaserów 128, Warsaw 00-909, Poland (Email: jdejewska@yahoo.pl)

doi: 10.1007/s12519-012-0381-1

Background: Atopic dermatitis (AD) diagnosis in children gives rise to many problems. Although the Hanifin and Rajka criteria are acknowledged and used universally in diagnosing AD, their evaluation is a major source of difficulty in pediatrician's daily practice. The simplified criteria, revised by Williams et al, seem to be more useful and represent the most common clinical manifestations of AD. The aim of the present study was to compare the efficacy of the two different criteria for diagnosing AD in children.

Methods: This study involved 250 children with AD. All of the patients underwent clinical examinations and the diagnostic procedures according to the criteria of Hanifin and Rajka as well as those of Williams.

Results: According to the Hanifin-Rajka criteria, AD was diagnosed in 173 children, of whom 153  were diagnosed positive by the criteria of Williams. Among the 77 children who were diagnosed without AD according to the Hanifin and Rajka criteria, 4 were detected with AD by the Williams criteria. Four children with scabies and seborrhoeic dermatitis were misdiagnosed as having AD according to the Williams criteria due to the atypical locations of their lesions and the histories of asthma or hay fever and xerosis.

Conclusions: Although the criteria of Hanifin and Rajka are the gold standard for diagnosing AD, the Williams criteria are also very useful in children older than 4 years. The most useful Williams criteria for AD diagnosis in children is pruritus with history of lesions in characteristic locations and history of generally dry skin. The significant increase in the detectability of AD is found by raising the borderline age for the first skin lesion development from 2 to 5 years of age.

Key words: atopic dermatitis; diagnosis; Williams criteria