Thyroid dysfunction in patients with impaired glucose metabolism: 11 year follow up from the Tehran Thyroid Study

OBJECTIVES This study aimed to assess the prevalence and incidence and predictive factors of thyroid disorders (TD) in patients with impaired glucose metabolism. METHODS Prevalence of TD was calculated in patients with impaired glucose metabolism compared to healthy controls, aged over 30 years in phase 1 of the Tehran Thyroid Study (TTS). Follow up assessments were conducted every 3 yrs, after which incidence of TD was calculated and its correlations with age, sex, smoking, blood pressure, body mass index (BMI), thyroid peroxidase antibody (TPOAb), thyrotropin (TSH), insulin resistance index, triglycerides and cholesterol were assessed. RESULTS Incidence of TD among 435 diabetics, 286 prediabetics, and 989 healthy controls at baseline was 14, 18, and 21 per 1000 patients per year, respectively, being significantly lower in diabetics than that in healthy controls, a difference however that was not significant after adjusting for the variables mentioned (OR:0.64, 95% CI: 0.39-1.01). The incidence of TD in subjects with baseline serum TSH>1.94 mU/L or TPOAb≥40 IU/ml in all three groups was higher than that in patients with TSH≤1.94 mU/L or TPOAb<40 IU/ml, and remained significant after variable adjustment. Baseline TSH>1.94 mU/L was predictive of TD with 70% sensitivity and specificity. Baseline serum TSH (ROC area: 0.73, 95% CI: 0.68-0.77) had better predictive value than TPOAb (ROC area: 0.65, 95% CI: 0.61-0.69) for developing TD. CONCLUSION Incidence of TD in type 2 diabetics or prediabetics is not higher than healthy controls. It is however necessary to conduct thyroid tests in patients with TPOAb≥40 IU/ml or TSH>1.94 mU/L.