Aspects of chronic autoimmune thyroiditis in children and adolescents

Abstract: The aims of this thesis were to 1) compare different methods for detection of antibodies against thyroglobulin (TgAb) and thyroid peroxidase (TPOAb), particulary in children with chronic autoimmune thyroditis (AIT), 2) investigate if maternal enterovirus infection is a risk factor for development of AIT in offspring, 3) ascertain the prevalence of thyroid autoantibodies in cord blood sera from children who later developed AIT, 4) develop a new reference standard for interpretation of normative thyroid volumes in Swedish children, and 5) evaluate the effect of Levothyroxine (L-thyroxine) treatment on thyroid size in children and adolescents with AIT. We found that 1) Methods based on radioimmunoassay (RIA) were more efficient in detecting both TgAb and TPOAb in children with AIT than methods based on agglutination. TgAb (RIA) was the only method able to detect antibodies in all children with AIT. We believe that methods for detection of TgAb and TPOAb should be based on immunoassays, such as RIA, and that TgAb still is valuable in the clinical diagnosis of AIT. 2) Immunoglobulin M (IgM) against enterovirus, indicating an enterovirus infection late in pregnancy were more common among mothers of children with AIT, as compared to controls (16% vs. 7%). The difference did not reach statistical significance. However, children of enterovirus IgM-positive mothers were significantly younger and more often overtly hypothyroid upon diagnosis of AIT than children of enterovirus IgM-negative mothers, suggesting a promoting effect on the autoimmune process in genetically-susceptible individuals. 3) Children with AIT had had a significantly higher prevalence of thyroid autoantibodies in cord blood, as compared to controls (44% vs. 16%). Whether this intrauterine exposure to maternal thyroid autoantibodies had a pathogenic role in the development of AIT in these children, or simply mirrored the inheritance of AIT, remains to be investigated. 4) Thyroid volume correlated independently with age and with anthropometrics, to a significant degree. To accurately interpret thyroid volume, we feel that both age and anthroprometrics need to be considered. To accomplish this, we propose applying a multivariate model using age and body surface area (BSA) as predictors. 5) Based on the muiltivariate model for interpretation of thyroid volume, L-thyroxine treatment was found to reduce thyroid size in children with goiter due to AIT, irrespective of thyroid function when treatment commenced. We, therefore, recommend L-thyroxine for treatment of goiter in children with AIT.