Determinants of TSH change in a community-based cohort

Thyroid disorders are common with occurrence primarily determined by the availability of dietary iodine. Iodine fortification programmes are internationally recommended to ensure sufficient iodine intake in populations. An understanding of the role of thyroid hormone levels within the normal range, set points and etiological factors related to thyroid disease development is important for optimal prevention and treatment. Limited data, however, exist regarding the impact of iodine fortification on thyroid function development. Additionally, the relation between body weight and thyrotropin (TSH) within the normal range and the role of female reproductive factors in the etiology of thyroid autoimmunity is debated.

The aim of this PhD project was to analyse the effect of a nationwide iodine fortification programme on individual development in thyroid function and to identify concurrent determinants for the possible changes. Furthermore, we aimed to investigate the association between weight and serum TSH change as well as the association between female reproductive factors and change in TSH and thyroid peroxidase antibody (TPO-Ab) status.

A longitudinal population-based study of the DanThyr C1 cohort examined before (1997-1998) and after (2008-2010) the introduction of mandatory iodine fortification of salt on July 1 2000. A total of 2,465 individuals participated in the follow-up examination. The main outcome measure was change in serum TSH. Change in TPO-Ab status was additionally used in Paper III.

Urinary iodine excretion levels increased significantly during follow-up. Serum TSH also increased significantly, most pronounced in the region with the highest iodine intake, whereas the increase was not significant in the low-iodine-intake region. The presence of TPO-Ab at baseline and absence of goitre and multiple nodules were identified as determinants of TSH increase. Moreover, a low-normal TSH at baseline was a determinant of future decreased serum TSH, while likewise a high-normal baseline TSH values determined a TSH above normal reference range at follow-up. A positive association between 11-year serum TSH change and weight change was found, but without baseline body mass index being a determinant of future weight change and without baseline TSH being a determinant of future weight change. An inverse association between the time on HRT treatment and the risk of increased TPO-Ab status during follow-up was found, but the association was not significant when applying the Bonferroni adjusted significance level and not associated with TSH change. Parity, OCP use, abortions, age at menarche and menopausal status were associated neither with TSH change nor with increased TPO-Ab status during follow-up.

TSH increased significantly, and the difference be-tween regions with different iodine intakes could indicate that iodine, at least partly, explains the TSH increase. The identified determinants of TSH change may indicate that susceptible individuals were subject to well-known adverse effects of iodine fortification. The predictive value of TSH on future TSH levels suggests a gradual development of thyroid disease. Whether body weight and TSH are causally connected remains to be proven. These results are an important contribution to the discussion of the role of thyroid hormones level within the normal range, set points and the association with body weight. A minor role, if any, is suggested for the studied female reproductive factors in development of thyroid autoimmunity. The longitudinal study neither solves the problem of causality nor is of the optimal design to measure the impact of iodization of salt, but can be informative in the study of determinants.