Mistaking the population reference range of thyroid-stimulating hormone for an individual's “normal” range can lead to suboptimal diagnosis and treatment of thyroid disease, according to Carole Spencer, Ph.D., professor of medicine and director of the Endocrine Services Laboratory at the University of Southern California in Los Angeles.
Population reference ranges for thyroid function tests are based on statistical averages, not on standards of biologic activity at different levels of thyroid hormones, Dr. Spencer stressed in an audioconference on the challenges of thyroid testing sponsored by the American Association of Clinical Chemistry. “Because individuals have different pathophysiologic factors that influence TSH [thyroid stimulating hormone], the combination of these variables in cohorts of individuals widen the TSH population reference range,” she said.
For example, “the typical [National Health and Nutrition Examination Survey] U.S. population reference range for TSH is 0.4–4.1 mU/L, which is much wider than the annual fluctuations in TSH typical of individuals,” said Dr. Spencer. “In fact, a TSH change in an individual that exceeds 0.75 mU/L, which is well within the population range, is biologically significant.”
Although the TSH population reference range is a crude parameter for detecting disease in individual patients, its accuracy is considered critical “because TSH is the first abnormality to appear as disease develops,” said Dr. Spencer. Additionally, “there is growing data suggesting that mild subclinical hypothyroidism can exacerbate the risk for cardiovascular disease in susceptible individuals,” she said. As such, identifying patients at risk for developing this condition is important.
There has been much debate among endocrinologists over the upper reference limit for TSH in particular. Over the past 3 decades, “the reference ranges have contracted as a result of improvements in TSH assay sensitivity and specificity, coupled with more sensitive thyroid antibody tests that are used to eliminate individuals with thyroid autoimmunities from reference range calculations,” said Dr. Spencer. “While the lower reference has been established as being around 0.3 mU/L, the upper reference limit has steadily declined from approximately 10 mU/L to between 4.0 and 4.5 mU/L.”
Additionally, there is strong clinical rationale for adopting the even narrower TSH reference range of 0.3–3.0 mU/L recommended by the American Association of Clinical Endocrinologists in 2003, Dr. Spencer noted. “We know from a 20-year follow-up survey of the Wickham thyroid survey [a historical cohort study that provided incidence data for thyroid disease for a representative cross-sectional sample of the population] that a TSH above approximately 2.0 mU/L is a risk factor for future development of hypothyroidism, especially when thyroid peroxidase [TPO] antibody is detected.” This is true “even in the absence of thyroid antibodies,” she said.
The empiric lowering of the upper reference limit is further justified in light of the growing recognition that even mild subclinical thyroxine deficiency in early pregnancy is detrimental to the mother and fetus, Dr. Spencer added.
Despite the data supporting the narrower reference ranges, many U.S. laboratories have not revised their “normal range” criteria, nor have all clinicians adopted the new range for diagnostic and treatment purposes, according to Dr. Spencer. One perceived roadblock has been the lack of data on the patient management implications of making the change, she said.
“Patients with TSH outside the narrower reference range do not necessarily require treatment,” said Dr. Spencer. “Rather, the degree of TSH abnormality outside the range should be viewed as a risk factor for current or future development of hyper- or hypothyroidism.”
The threshold for treatment has to be adjusted for patient-specific factors, such as the degree of TSH abnormality, family history of cardiovascular disease, diabetes, insulin resistance, hypertension, smoking, age, and presenting symptoms, Dr. Spencer noted. Additionally, the presence and concentration of thyroid autoantibodies is an important factor. “The higher the thyroid peroxidase autoantibody concentration, the more rapid the progression of disease,” she said.
Ultimately, the diagnosis and efficacy of treating subclinical hypothyroidism should not be based on the TSH reference range alone, “but should integrate the degree of TSH elevation with patient-specific risk factors and the concentration of TPO antibodies,” Dr. Spencer said.