ESPE Abstracts

Introduction: The prevalence of thyroid peroxidase antibody (anti-TPO) positivity is estimated to be around 3-4% in healthy children and is remarkably higher in children with type 1 diabetes (T1D). However, anti-TPO positivity in children with HLA-conferred susceptibility to T1D who are not yet diagnosed with T1D, is not well studied. The aim of this study was to describe the prevalence of positive anti-TPO and its effect of thyroid function in children with a genetic susceptibility to T1D.

Methods and subjects: Two hundred twenty-five children (114 boys) aged 7.5 to 10.4 years from the Estonian DIABIMMUNE birth cohort were studied. None of them had a known thyroid disease, two children were previously diagnosed with T1D and two were diagnosed during the study period. Anti-TPO concentration was measured with ECLIA method. In line with the definition provided by the test manufacturer, anti-TPO concentration of >18 kU/l was considered positive. Clinically significant anti-TPO value was defined as >100 kU/l. When anti-TPO was positive, thyroglobulin antibody (anti-TG), TSH and free T4 were measured. All subjects had diabetes associated antibodies measured (IAA, IA2A, GADA, ZnT8A and ICA when one of the previous was positive). Statistical analysis was performed with Mann-Whitney U test, P<0.05 was considered statistically significant.

Results: Girls had a higher median anti-TPO concentration than boys (12 vs 11 kU/l, P=0.001). A positive anti-TPO concentration occured in 32 subjects (20 girls), that is 14.2% of the cohort. Girls with a positive anti-TPO concentration had a higher median anti-TPO value than boys, but this was not statistically significant (27,5 vs 23,5 kU/l, P=0.495). Four subjects (1,8%) had anti-TPO levels >100 kU/l, suggesting a very likely autoimmune thyroiditis. Anti-TG was measured in 28 children (four lost to follow-up), it was remarkably risen (>100 kU/l) in five subjects, of whom three had a high anti-TG without a clinically significant anti-TPO rise. Thyroid function was normal for all of these children. Only one of the 32 subjects had a positive diabetes associated antibody – IAA.

Conclusion: Girls with HLA-conferred susceptibility to T1D had higher anti-TPO concentrations than boys. The prevalence of positive anti-TPO in our cohort was 14.2% which is remarkably higher than previously described prevalence in healthy children. All of the subjects had a normal thyroid function, therefore, in children with HLA-conferred susceptibility to T1D, routine anti-TPO measuring is not justified, at least up to the age of ten.