T4 can be measured directly by radioassay (or, more recently, by enzyme-linked assay). Radioassay involves competition of patient hormone and radioactive hormone for a hormone binder. In T4by radioimmunoassay (T4-RIA) the binder is anti-T4antibody; in T4 competitive binding (T4-CPB) or T4 displacement methods (T4-D) the binder is T4-binding serum protein. After patient T4 and radioactive T4 have reacted with the T4 binder, the bound complex is separated from the free (unbound) hormone and the radioactivity in either the bound or free fraction is counted. The amount of patient T4 determines the amount of radioactive T4 allowed to attach to the binder. There is general (but not unanimous) agreement that T4 values are not greatly changed in old age.

Results in thyroid disease. The sensitivity of T4 assay is approximately 95% (literature range, 90%-100%) for hyperthyroidism, and approximately 92% (literature range, 54%-100%) for hypothyroidism. There is some difficulty in evaluating T4 data for hypothyroidism from the literature since patients with a temporary T4 decrease from thyroiditis or severe non thyroid illness may be included with patients having true primary hypothyroidism in some reports.

Drawbacks. As noted previously, more than 99% of total serum T4 is protein bound, so that the contribution of free T4 to the test results is negligible. Therefore, serum T4 by any current method essentially measures protein-bound T4 and will be affected by alterations in T4-binding proteins. These alterations may be congenital, drug induced, or secondary to severe non thyroid illness. As a rule of thumb, estrogens from pregnancy or birth control medication increase T4 levels by increasing TBG levels, and amphetamine abuse is reported to increase T4 levels in some patients by increasing pituitary secretion of TSH. Most other drugs that affect T4 produce decreased T4 levels by decreasing TBG levels or T4 binding to TBG (or albumin). Medications that may affect T4 assay are listed in Table 37-14. The most commonly encountered medication problems, in my experience, are associated with administration of levothyroxine (Synthroid), estrogens, or phenytoin (Dilantin). Like phenytoin, valproic acid, furosemide, and some of the nonsteroidal anti inflammatory drugs can decrease T4 levels by displacing T4 from its binding proteins. However, these medications, especially furosemide, are often administered to patients who already have severe non thyroid illness, so that it may be difficult to separate the effect of the drug from the effect of the illness.

Another condition producing false T4 increase is familial dysalbuminemic hyperthyroxinemia (FDH), which is uncommon but not rare. This is a congenital condition transmitted as an autosomal dominant trait, which results in production of an albumin variant that has an increased degree of T4 binding (albumin normally has only a weak affinity for T4). Therefore, in FDH more T4 binds to albumin, which raises serum total T4 values. Thyroid function is not affected and there apparently is no association with disease. Yet another cause for false T4 elevation is the very uncommon syndrome of peripheral resistance to T4. Peripheral tissue utilization of T4 is decreased to variable degree; whether symptoms develop depends to some extent on the degree of “resistance.” The pituitary response to T4 feedback may be decreased in some patients. Total T4, free T4, and total T3 are elevated; serum THBR (T3U), TSH, and TSH response in the TRH test are normal or elevated.

With the possible exception of TBG alterations, the most frequent false abnormalities in T4 are produced by severe non thyroidal illness (see the box). Severe non thyroid illnesses or malnutrition of various etiologies may decrease T4 below the lower limits of the T4 reference range.

Some Nonthyroid Illnesses that Can Affect Thyroid Tests

Cirrhosis or severe hepatitis
Renal failure
Cancer
Chemotherapy for cancer
Severe infection or inflammation
Trauma
Postsurgical condition
Extensive burns
Starvation or malnutrition
Acute psychiatric illness

The incidence of decreased T4 is roughly proportional to the severity of illness. Therefore, the overall incidence of decreased T4 in non thyroid illness is low but in severe illness reports indicate decreased T4 levels in approximately 20%-30% of cases (literature range, 9%-59%). In fewer cases the free T4 index (TI) or even free T4 assay (especially analog methods) is reduced below reference limits. On the other hand, in the majority of patients, and in patients without alterations in albumin or TBG, the T4 remains within reference range (although possibly decreased from baseline values). In occasional patients (for poorly understood reasons) the T4 value may be mildly elevated. This has been called “sick euthyroid syndrome” by some investigators. The TSH value may decrease in severe non thyroid illness somewhat parallel to T4 but to a lesser extent, usually (but not always) remaining within reference range. In the recovery phase, TSH levels increase before or with T4 and could temporarily even become elevated. The THBR value in severe non thyroid illness is most often normal but may be mildly increased, reflecting decreased TBG levels. Occasionally the THBR is decreased, mainly in acute hepatitis virus hepatitis

Cushing’s syndrome produces a mild or moderate T4 decrease below lower limit of reference range in two thirds or more of patients. Free T4 (FT4) levels are usually normal.

Conditions producing a decrease in T4 are listed in the box and conditions producing an increase in T4 are listed in the box.

One group of conditions involves certain iodine-containing substances. For many years it was thought that neither organic iodine nor inorganic iodide would interfere with T4 by immunoassay methods. However, it was subsequently found that a few x-ray contrast media such as ipodate (Oragrafin) and iopanoic acid (Telepaque) can elevate

Causes for Decreased Thyroxine or Free Thyroxine Values

Lab error
Primary hypothyroidism
Severe non thyroid illness;* many patients
Lithium therapy; some patients
Severe TBG decrease (congenital, disease, or drug-induced) or severe albumin decrease*
Dilantin, Depakene, or high-dose salicylate drugs*
Pituitary insufficiency
Large doses of inorganic iodide (e.g., SSKI)
Moderate or severe iodine deficiency
Cushing’s syndrome
High-dose glucocorticoid drugs; some patients
Pregnancy, third trimester (low-normal or small decrease)
Addison’s disease; some patients (30%)
Heparin effect (a few FT4 kits)
Desipramine or amiodarone drugs; some patients
Acute psychiatric illness; a few patients
___________________________________________________________________
*FT4 less affected than T4; two-step FT4 method affected less than analog FT4 method.

T4values above reference range in some patients and that the iodinated antiarrhythmic drug amiodarone can produce temporary hyperthyroidism in some patients and hypothyroidism in others. These compounds do not affect the T4 assay directly but can increase T4 values by certain metabolic actions (e.g., blocking of T4 deiodination) that are still being investigated. Effects of iodine on the thyroid will be discussed in detail later