The pituitary hormone, thyrotropin (TSH), is regarded as the primary biomarker for evaluating thyroid function and is useful in guiding treatment with levothyroxine for patients with hypothyroidism. The amplified response of TSH to slight changes in thyroid hormone levels provides a large and easily measured signal in the routine care setting. Laboratories provide reference ranges with upper and lower cutoffs for TSH to define normal thyroid function. The upper limit of the range, used to diagnose subclinical (mild) hypothyroidism, is itself a matter for debate, with authoritative guidelines recommending treatment to within the lower half of the range. Concomitant diseases, medications, supplements, age, gender, ethnicity, iodine status, time of day, time of year, autoantibodies, heterophilic ant ibodies, smoking, and other factors influence the level of TSH, or the performance of current TSH assays. The long-term prognostic implications of small deviations of TSH from the reference range are unclear. Correction of TSH to within the reference range does not always bring thyroid and other biomarkers into range and will not always resolve the patient's symptoms. Overt hypothyroidism requires intervention with levothyroxine. It remains important that physicians managing a patient with symptoms suggestive of thyroid disease consider all of the patient's relevant disease, lifestyle, and other factors before intervening on the basis of a marginally raised TSH level alone. Finally, these limitations of TSH testing mitigate against screening the population for the undoubtedly substantial prevalence of undiagnosed thyroid disease, until appropriately designed randomised trials have quantified the benefits and harms from this approach.
