BACKGROUND: Eosinophil cationic protein (ECP) is a cytotoxic agent secreted by activated eosinophils during allergic and inflammatory processes. The aim of the study was to determine the ECP level, absolute and relative eosinophil count and IgE antibodies in children with atopic dermatitis (AD) compared with those of nonatopic children, and to assess the correlation of these laboratory parameters with the clinical severity of AD. ----- METHODS: This prospective study comprised 70 children. There were 49 children with AD aged 3-36 months, and the control group comprised 21 children with a negative personal and family history for atopic diseases. Detailed history, serum ECP levels (UniCAP FEIA), relative and absolute eosinophil counts and total serum IgE antibodies were determined in both groups. In the children with AD, skin involvement was measured by the SCORAD index. ----- RESULTS: The calculated SCORAD index was between 16 and 83. IgE antibodies, relative and absolute eosinophil counts showed a significantly wider range of values and a statistically higher median (P < 0.001) in the patients with AD compared with the control group. These laboratory parameters did not correlate with the severity of AD. The serum ECP median level, in the children with AD, was 16.2 microg/L (range 3.01-65.30) compared with 5.92 microg/L (range 2.76-21.90) in the control group. Correlation of the total SCORAD index and the serum ECP levels was negative, weak (r = -0.065) and statistically not significant (P > 0.05). The same was found for the correlation of serum ECP and intensity of skin changes (r = -0.095) and serum ECP and subjective symptoms (r = -0.045). The correlation was positive, but weak and statistically not significant for the serum ECP and extent of the skin lesions (r = 0.079, P > 0.05). ----- CONCLUSION: Elevated levels of ECP, relative and absolute eosinophil counts, as well as IgE antibodies were determined in the patients with AD. As these laboratory findings did not correlate with the severity of AD, they can be considered only as additional methods in the evaluation of patients with AD.