The need to measure free thyroxine levels has diminished in the past 2 decades, but it's still a frequently ordered test and may be an important one for some patient populations, especially pregnant women.
There's no consensus, however, on how to measure free thyroxine (T4). Efforts are underway to standardize free T4 testing amid some controversy over whether this is scientifically feasible—or medically necessary.
Results of free T4 assays vary according to which immunoassay is used and which laboratory processes it.
The available immunoassays “all have problems,” says Dr. James D. Faix, director of clinical chemistry and immunology at Stanford (Calif.) University. “If your goal is to accurately measure the free T4, they probably all fail at that, but they may be clinically useful in terms of estimating the free T4.”
Improvements in measuring thyroid stimulating hormone (TSH) in the 1980s and early 1990s expanded the usefulness of TSH tests from detection of elevated TSH levels to include detection of suppressed TSH levels associated with hyperthyroidism. For many clinicians, TSH became a single-test screen of thyroid function, and the use of free T4 tests to detect or monitor hyperthyroidism declined.
The test frequently still gets ordered today to verify hyperthyroidism in a patient with low TSH results, even though its usefulness is questioned by some.
Primary care physicians may get along fine using just TSH tests, “but for a thyroid clinic, we routinely get both” tests, said Dr. Francis S. Greenspan, chief of the thyroid clinic at the University of California, San Francisco. “Most textbooks recommend this combination to evaluate patients who have thyroid disease.”
If a patient's TSH level is suppressed, for example, a treating clinician wants to know if that's due to a slight rise in circulating hormone or a significant rise, to help determine the need for aggressive therapy, he said.
Free T4 immunoassays aren't reliable in subgroups of patients with disturbances in binding proteins, Dr. Carole Spencer noted.
“These free T4 immunoassays will give spuriously misleading values in just those conditions where you'd like to have an accurate free T4, in conditions like pregnancy and nonthyroidal illnesses” that affect some hospitalized patients, said Dr. Spencer, who is professor of medicine at the University of Southern California, Los Angeles.
A recent study by Dr. Jerald Nelson and associates at Loma Linda (Calif.) University reported that a free T4 immunoassay measured total T4, not free T4 (Clin. Chem. 2007;53:911–5). Further studies found that the same is true of the other free T4 immunoassays, he said.
Dr. Nelson's findings call to mind previous unsuccessful efforts by the American Thyroid Association to get the immunoassay manufacturers to call them free T4 estimate tests, “to alert physicians that they weren't the real McCoy. They were merely estimate tests and could be erroneous at times,” Dr. Spencer said. “These new studies of Dr. Nelson shockingly show that these immunoassays are just glorified T4 tests and can be very misleading.”
Dr. Faix cautioned that physicians should “continue to be skeptical” about the tests. “People who rely on free T4 to confirm abnormal TSH results need to be aware that the problems with free T4 testing have not been solved.”
Dr. Faix is a member of the Working Group on Standardization of Thyroid Function Tests commissioned by the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC). The working group is one of several launched by the IFCC to respond to a European Union directive to standardize all laboratory testing.
So far, it's unclear to the group whether it's feasible to set a gold standard for measuring free T4 (see sidebar), he said. In addition, some U.S. physicians think that the European mandate for standardizing all tests is overreaching, and shouldn't include free T4.
The effort could pay off, however, especially for pregnant women, Dr. Faix said.
Women with subclinical hypothyroidism develop clinical thyroid disease during pregnancy, creating risks for the fetus if they go without therapy. Because pregnancy increases levels of thyroid binding proteins, a free T4 immunoassay to confirm an abnormal TSH level in pregnancy “is not as reliable because of these biological changes,” he said. “A better standardized method for ascertaining the thyroid function status during pregnancy would be helpful.”
Using the immunoassays, free T4 readings go lower and lower as gestation progresses and vary by the test used. Up to 65% of pregnant women will have low free T4 values by the third trimester according to one immunoassay, but only about 15% of pregnant women have that result using a different one, Dr. Spencer said.