Abstract
BACKGROUND: Recent attention has been given to subclinical hypothyroidism, defined as an elevation of TSH (4.5-10 uIU/L) with T4 and T3 levels still within the normal range. Controversy exists about the proper lower limit of TSH that defines patients in the subclinical hypothyroidism range and about if/when subclinical hypothyroidism should be treated. Additional data are needed to examine the relationship between markers of thyroid function in the subclinical hypothyroidism range, biomarkers of health and ultimately health outcomes. OBJECTIVE: We aimed to assess the relationship between serum TSH levels in the 0.5-10 uIU/L range and serum cortisol in a cohort of healthy young men and women without clinical evidence of hypothyroidism. Based on data in frank hypothyroidism, we hypothesized that serum TSH levels would be positively correlated with serum cortisol levels, suggesting derangement of the cortisol axis even in subclinical hypothyroidism. METHODS: We conducted a cross sectional study in 54 healthy, young (mean 20.98 +/- 0.37 yrs) men (19) and women (35). Lab sessions took place at 1300 hrs where blood was drawn via indwelling catheter for later assessment of basal serum TSH, free T3, free T4, and cortisol levels. RESULTS: All but 1 participant had free T3 levels within the normal reference intervals; free T4 levels for all participants were within the normal reference intervals. Linear regression modeling revealed that TSH levels in the 0.5-10 uIU/L were significantly and positively correlated with cortisol levels. This positive TSH-cortisol relationship was maintained below the accepted 4.5 uIU/L subclinical hypothyroid cutoff. Separate regression analyses conducted by systematically dropping the TSH cutoff by 0.50 uIU/L revealed that the TSH-cortisol relationship was maintained for TSH levels (uIU/L) ≤4.0, ≤3.5, ≤3.0, and ≤2.5 but not ≤2.0. Linear regression modeling did not reveal a relationship between free T3 or free T4 levels and cortisol levels. CONCLUSIONS: Results suggest a positive relationship between TSH and cortisol in apparently healthy young individuals. In as much as this relationship may herald a pathologic disorder, these preliminary results suggest that TSH levels > 2.0 uIU/L may be abnormal. Future research should address this hypothesis further, for instance through an intervention study.