The association of early infections and atopic dermatitis: A retrospective case-control study in Chinese children in primary care in Hong Kong

Abstract

Objectives To investigate the association of early infections and atopic dermatitis (AD) in Chinese children in primary care in Hong Kong.

Methods Our setting were two primary care surgeries run by a family physician. We searched our database with “atopic dermatitis” and “atopic eczema” and retrieved medical records of all infants and children (below 8) with AD diagnosed between 1 April 2003 and 31 March 2004. An age-limit of 8 was used as the surgery commenced service 9 years ago. Records of children above 8 might not have reliable early history documented.

The diagnoses were all made by one of us (AC) with training in paediatrics and dermatology. We checked whether the diagnostic criteria according to the United Kingdom Working Party was fulfilled, and excluded children who do not fulfil the diagnostic criteria, not of Chinese origin, or not cared by us since birth.

For each child with AD, we retrieved the record of the next child of same sex and similar age (±1 year) who consulted us but did not have AD as a paired control.

For all study and control subjects, we retrieved the following information from records: (1) whether the child had a febrile episode or a clinical infection before age of 6 months, number of such episodes, diagnoses made, (2) whether the child was hospitalised before 6 months, number of such episodes, diagnoses made, and (3) whether the child was prescribed antibiotics before 6 months, number of such episodes, diagnoses made. For (1), febrile episodes clearly documented to be related to vaccinations were not included.

One of us then re-checked the records for validity of diagnosis and accuracy of information.

Results Records of 375 children were retrieved. Of such, 182 were not cared by us since birth. Of the remaining 193, 132 (74 boys and 58 girls) had documented evidence of fulfilling the diagnostic criteria. They aged from 4 months to 8 years (mean: 3.8 years). The 132 control subjects aged from 5 months to 8 years (mean: 3.6 years).

33 (25%) children with AD had a total of 48 febrile episodes or infection before 6 months, while 39 (31%) controls had 59 febrile episodes or infection (p=0.46). Viral respiratory tract infections were the commonest, followed by otitis media. 4 (3%) children with AD and 3 (2%) controls were hospitalised during the first 6 months (p=0.46). 9 (7%) children with AD and 8 (6%) controls were given antibiotics during the first 6 months (p=0.82).

The odds ratio of having AD for children with no febrile episode or infection to those with one or more febrile episodes or infection was 1.26 (95% CI: 0.71–2.18).

Conclusions There is no significant association between early infection and the risk of AD in our sample of Chinese children in primary care in Hong Kong.