Vaccinations may induce diabetes-related autoantibodies in one-year-old children.

ABSTRACT

Vaccinations have been discussed as one among many environmental candidates contributing to the immune process that later may lead to type 1 diabetes. ABIS (All Babies in Southeast Sweden) is a prospective cohort study following a nonselected birth cohort of general population. In a randomly selected sample collection from 4400 children, GADA and IA-2A have been determined at the age of 1 year. The information on vaccinations was collected from questionnaires answered by the parents and was related to beta cell autoantibodies. When studying the induction of autoantibodies using the autoantibody level of 90th percentile as cutoff level, hemophilus influenza B (HIB) vaccination appeared to be a risk factor for IA-2A [OR 5.9 (CI 1.4-24.4; p = 0.01)] and for GADA [OR 3.4 (CI 1.1-10.8; p = 0.04)] in logistic regression analyses. Furthermore, the titers of IA-2A were significantly higher (p < 0.01 in Mann-Whitney test) in those children who had got HIB vaccination. When 99th percentile was used as cutoff to identify the children at risk of type 1 diabetes, BCG vaccination was associated with increased prevalence of IA-2A (p < 0.01). We conclude that HIB vaccination may have an unspecific stimulatory polyclonal effect increasing the production of GADA and IA-2A. This might be of importance under circumstances when the beta cell-related immune response is activated by other mechanisms.