IntroductionAutoimmune thyroid disease is common, affecting approximately 1% of the population while subclinical, focal thyroiditis and/or circulating thyroid autoantibodies can be found in about 15% of otherwise healthy subjects who are euthyroid. This frequency, combined with historical precedence, ready h e s s to the target organ and well established animal models, has led to a considerable research effort aimed at understanding the initiation and pathogenesis of thyroid autoimmunity, often with the hope that the lessons learned may be applicable to more serious but less accessible autoimmune diseases. In this review, I shall summarize recent developments in autoimmune hypothyroidism (Hashimoto's thyroiditis and primary myxoedema), only touching on the important insights gained from experimental autoimmune thyroiditis (EAT) and other animal models, detailed elsewhere (Weetman, 1991). Similar autoimmune processes occur in Graves' disease, which is uniquely distinguished by the presence of thyroid stimulating antibodies. These and other types of TSH receptor antibody are discussed in a separate review (Munro, 1992) and therefore will not be considered further here.
PredispositionAs in many autoimmune disorders, it is the interplay of genetics and environment which determines the initiation of autoimmune thyroiditis, with a further modifying influence provided by endogenous factors such as age and sex hormones.
lmmunogenetic factorsHLA-linked genes The major histocompatibility complex (MHC) of genes, called HLA in man, controls (in part) immune responsiveness to a variety of foreign and self antigens. In many autoimmune conditions, associations have been reported with certain MHC alleles, particularly those encoded by the class I (HLA-A,B,C) and class I1 (HLA-D) regions. On the whole, such associations tend to be stronger with class I1 than class I region genes. Many of the alleles within and between these regions are in linkage disequilibrium, that is, closely linked genes occur together more frequently than expected with random distribution. It is therefore difficult to determine from population-based surveys whether associations of a n autoimmune disease with a particular allele are truly an effect of this allele rather than one in linkage disequilibrium with it. To complicate matters, HLA genes have been renamed recently. A clear exposition of the old and new terminology can be found elsewhere (Tait &Harrison, 1991). As most analyses of HLA in thyroiditis to date have used serologically defined specificities, this nomenclature will be preserved below.Initial studies showed a n association of primary myxoedema with HLA-B8 in Caucasians, but none between class I region genes and Hashimoto's thyroiditis (Irvine, 1978;Farid et al., 1981). With the availability of serological reagents for HLA-DR typing, this dichotomy was further emphasized, with reports of a n HLA-DR3 or DR5 association with atrophic thyroiditis (primary myxoedema) and Hashimoto's thyroiditis respectively (Weissel et al., 1980;Farid et al., 19...