Association between thyroid dysfunction and dysglycaemia: a prospective cohort study.

AIMS: To compare the incidence of hyperglycaemia among participants with low, elevated and normal serum thyroid-stimulating hormone concentration, as well as the incidence of abnormal thyroid function test results among participants with normal blood glucose and those with hyperglycaemia.

METHODS: In a prospective study, a cohort of 72 003 participants with normal, low and elevated serum thyroid-stimulating hormone concentration were followed from the study beginning to the first report of diabetes and prediabetes. A proportional hazards regression model was used to calculate the hazard ratios and 95% CIs for each outcome, adjusting for age, sex, education level, smoking, alcohol consumption and obesity. Analyses for the association between dysglycaemia and incident abnormal thyroid function test were also conducted.

RESULTS: During a median 2.6 year follow-up, the incident rates for dysglycaemia, particularly prediabetes, were substantially higher in participants with elevated thyroid-stimulating hormone concentrations at baseline, while the rates for participants with normal and low thyroid-stimulating hormone were similar. After controlling for risk factors, participants with elevated thyroid-stimulating hormone retained a 15% increase in risk of prediabetes (adjusted hazard ratio 1.15, 95% CI 1.04-1.26), but were not at greater risk of diabetes (adjusted hazard ratio 0.96, 95% CI 0.64-1.44). By contrast, participants with normal and low thyroid-stimulating hormone concentrations had similar dysglycaemia risks. Participants with diabetes and prediabetes were not at greater risks of developing abnormal thyroid function test results when compared with participants with euglycaemia.

CONCLUSIONS: People with elevated serum thyroid-stimulating hormone concentration are at greater risk of developing prediabetes. Whether this includes a greater risk of developing frank diabetes may require an extended period of follow-up to clarify.