Response to Fitzgerald et al. re: “Thyroid Stimulating Hormone and Thyroid Hormones (Triiodothyronine and Thyroxine): An American Thyroid Association-Commissioned Review of Current Clinical and Laboratory Status”

Dear editor,

The recent American Thyroid Association (ATA) commissioned review on clinical and laboratory status for thyroid laboratory testing ¹ prompted Fitzgerald et al. to retort by raising two pivotal points ² : a request for robust empirical evidence that TSH is superior to thyroid hormones for thyroid testing, and evidence for individual set points in thyroid tests. ²

The individual thyroid set point has been demonstrated over decades, using different assays, and in several populations, including euthyroid subjects, patients with thyroid dysfunction, and pregnant women. Such data were reviewed earlier in this journal ³ in a report on the importance of biological variation for interpretation of thyroid function tests, with the addition of a summary of the methods recommended for producing evidence to individual patient testing. The repeated additional finding of narrow individual variation in thyroid function tests ³ implies that humans regulate thyroid hormones about slightly different set points.

Consequently, individual reference ranges for thyroid hormones are only half the width of the population-based reference ranges, ³ and an individual results of T4 can leave the reference range for your patient, while still being within the reference range of your laboratory. ³ Further support for the individuality of the pituitary–thyroid axis is available from a study of healthy twins reporting that each individual had a genetically determined thyroid function set point. ⁴ Thus, ample evidence agrees on an individual set point in thyroid function tests, as is accounted for in the ATA update on status for thyroid laboratory testing. ¹

Robust empirical data have shown that TSH outperforms T4 for detecting minor differences in thyroid function in the individual patient. Such data were reported in a recent comparison of data collected in strict accordance with recommendations. ⁵ The narrow individual variation around an individual set point in mildly hypothyroid patients caused the restricted response in T4 to remain within the population-based reference range. This contrasted with the magnified TSH response causing TSH to depart from the population-based reference range when T4 did not. These physiologically explained dynamics influence the performance of thyroid function tests, and Table 1 lists sensitivity, specificity, and positive and negative predictive values of TSH, total T4, and free T4 based on the empirical data. ⁵

In conclusion, robust empirical data show that TSH outperforms T4 for detecting minor differences in thyroid function in the individual patient in accordance with the physiology of the hypothalamic–pituitary–thyroid axis, which provides an explanation and corroborates the empirical data to establish TSH as superior to thyroid hormones for thyroid testing in the individual patient.

While Fitzgerald raises the point that T4 is the main predictor of disease in population outcome assessment, TSH is the key in the assessment of thyroid aberration in the individual patient. Still, each of the thyroid function tests provides valuable information about different aspects of thyroid function. Hence, while TSH remains the primary screening test for thyroid dysfunction, assessing T4 and T3 levels may be appropriate in addition to patient history and physical signs related to thyroid disease.