Each of the three categories of true laboratory thyrotoxicosis has a counterpart in which the apparent pattern does not reflect true thyroid hormone status. I would like to call the resulting test patterns “deceptive laboratory hyperthyroidism.” These patterns are misleading because of non thyroidal alteration of one or both thyroid hormone levels. Deceptive laboratory hyperthyroidism represents a significant (although relatively small) percentage of hyperthyroid patients. Therefore it is important to recognize these patients and to anticipate a potential problem when situations associated with deceptive test results arise. The categories of deceptive test results are the following:

1. Pseudo–T4/T3 hyperthyroidism (both T4 and T3-RIA test results are elevated, not due to Graves’ disease or Plummer’s disease)
2. Pseudo–T3 hyperthyroidism (T3-RIA value elevated; T4 test result not elevated)
3. Pseudo–T4 hyperthyroidism (T4 test result elevated; T3-RIA value not elevated)

Laboratory error may produce apparent abnormality in a euthyroid person or may reduce one or both of the hormone levels in true thyrotoxicosis. Therefore, unexpected test patterns or results may require repetition of one or more of the tests before a definitive diagnosis is made. Abnormality of the same type on two tests (i.e., both test results elevated) is more helpful for diagnosis than abnormality of only one.

Pseudo–T4/T3 hyperthyroidism (T4 and T3-RIA both elevated)

Causes of pseudo–T4/T3 Graves’ disease or Plummer’s disease are listed in the box. Pseudohyperthyroidism is most commonly produced by increase of thyroid binding proteins, principally TBG. The most common etiology is increased estrogens, either in pregnancy or through use of birth control pills. Both T4 and T3-RIA values are elevated in many of these patients. However, in some, the T3-RIA value may remain in upper reference range while only the T4 value is elevated.

Effects of TBG alterations on T4 levels can be counteracted in most instances by using the FT4I or by measuring FT4 rather than total T4 values.

Pseudo–T4/T3 Hyperthyroidism

Increased TBG values
Painless (silent) thyroiditis
Some patients with Hashimoto’s thyroiditis
Peripheral resistance to thyroid hormones
Factitious (self-medication with thyroid hormone)

The FT4I or FT4 value will usually be normal in TBG abnormalities. If the FT4I or FT4 and the T3-RIA values are measured in a patient with increased TBG, the T3-RIA will appear to be elevated since TBG alterations affect T3-RIA as well as T4, and the combination of elevated T3-RIA plus normal. FT4I values or FT4 value would suggest T3 toxicosis. Therefore, “correction” of TBG effect on T4 may prevent pseudo-T4/T3 toxicosis but produce pseudo-T3 toxicosis. This hazard can be prevented by either applying the same basic FT4I formula to T3-RIA (thus generating an FT3I) or simply inspecting the two separate components of the FT4I, the T4 and THBR (T3U). If T4 and TBHR are at the opposite ends of their respective reference ranges, this suggests artifact due to TBG alteration. Unfortunately, many laboratories that generate FT4I report only the single FT4I result without separate T4 and THBR values. The FT4I value alone has no feature that could lead anyone to suspect TBG abnormality.

Occasionally patients have been discovered with the syndrome of peripheral tissue resistance to thyroid hormone. These patients are usually euthyroid but have elevated T4 and T3-RIA values. TSH is normal or elevated.

Factitious (self-administered) ingestion of T4 compounds by a patient may be deliberate, may be due to prior therapy that is not mentioned by the patient, or may represent T4 included in diet-control pills unknown to the patient. In both factitious T4 ingestion and subacute thyroiditis the RAIU value is typically low. Spurious causes for a low RAIU value must be excluded, such as iodine ingestion (e.g., SSKI or amiodarone) or x-ray contrast medium administration within the past 3-4 weeks (see the box).

Thyrotoxicosis in thyroiditis is usually temporary and is produced by release of thyroid hormone from damaged thyroid tissue rather than by hypersecretion. Subacute thyroiditis typically has pain in the thyroid area and is accompanied by a low RAIU value. The erythrocyte sedimentation rate (ESR) is usually elevated (>50 mm/hour, Westergren method). Occasional cases of thyroiditis (“painless thyroiditis”) may present with the clinical picture of subacute thyroiditis, including hyperthyroidism, but without a painful thyroid gland and with a normal ESR. Chronic lymphocytic thyroiditis (Hashimoto’s thyroiditis) typically is associated with normal thyroid hormone blood levels, normal or decreased 24-hour RAIU value, normal ESR, and increased thyroid autoantibody levels. However, in occasional patients, chronic lymphocytic thyroiditis presents with hyperthyroidism that is clinically similar to nonpainful thyroiditis. The 24-hour RAIU value is typically decreased (although a few patients are reported to have normal values, and one study included a few patients with elevated values). Thyroid hormone levels are elevated, and the ESR is normal. Therefore, an elevated ESR favors subacute thyroiditis rather than factitious hyperthyroidism, painless thyroiditis, or thyrotoxic chronic lymphocytic thyroiditis.

Other entities with elevated T4 or elevated T3-RIA values associated with low RAIU values include T4/T3 toxicosis with artifactual RAIU suppression by exogenous iodine, iodine-induced hyperthyroidism (Jod-Basedow disease), radiation-induced active thyroid disease (caused by RAI therapy or external radiotherapy), and ectopic T4 production (struma ovarii). In severe (not mild or moderate) iodine deficiency, T3-RIA values may be increased, T4 values may be decreased, and TSH and RAIU values may be increased.

Pseudo–T3 hyperthyroidism (T3-RIA elevated; T4 not elevated)

Pseudo–T3 toxicosis may be of two types: (1) true hyperthyroid type, T4/T3 hyperthyroidism with normal-range T4 test result; and (2) false hyperthyroid type, elevated T3-RIA test result without hyperthyroidism (see the box).

True hyperthyroid type. Such cases are uncommon. It is said that T3 values may rise before T4 values in early thyrotoxicosis, and T3-RIA values are usually elevated to a greater degree than T4 values. Examples of isolated T3 elevation that eventually was joined by T4 elevation have been reported. An early or mild T4 abnormality may be masked in the upper area of population reference range (if a person’s normal T4 value were in the lower part of population reference range, the T4 value could double and still remain within reference range).

False hyperthyroid type. This type of pseudo–T3 toxicosis may be produced by measurement of T3-RIA plus either the FT4I or the FT4 in

Pseudo–T3 Hyperthyroidism

True hyperthyroidism
Rise of T3 level before T4 level in early T4/T3 hyperthyroidism
False Hyperthyroidism
Increased TBG with TI (or FT4) and T3-RIA results
1-2 hr after dose of T3 (liothyronine [Cytomel])
For several hours after dose of desiccated thyroid Severe iodine deficiency

patients with increased TBG values. Since FT4I and FT4 values usually remain normal when TBG values are altered, an increased TBG value would be associated with normal FT4I or FT4 values plus artifactual increase in T3-RIA value. The T3-RIA value may be increased alone for 1-2 hours after T3 (liothyronine) administration. It may also be temporarily increased for several hours after desiccated thyroid intake. Iodine deficiency of moderate degree usually is associated with normal T3-RIA and T4 levels, although mean T3-RIA values are higher than in normal persons. In severe iodine deficiency, T3-RIA values are sometimes increased, T4 values may be decreased, and TSH values may be increased. The RAIU value is increased in iodine deficiency, providing additional potential for misdiagnosis. As noted previously, one report indicates that occasional free T3 index elevations can be found in amphetamine abusers.

Pseudo–T4 hyperthyroidism (T4 elevated; T3-RIA not elevated)

Pseudo–T4 toxicosis may be of two types: (1) true hyperthyroidism, T4/T3 toxicosis with (temporarily) reduced T3-RIA test result; and (2) false hyperthyroidism, elevated T4 test result in euthyroid patient (see the box).

True hyperthyroidism. Ordinarily, both T4 and T3-RIA values are elevated in T4/T3 toxicosis or

Pseudo–T4 Hyperthyroidism

True hyperthyroidism (patient is hyperthyroid)
Factitious ingestion of levothyroxine
T4/T3 hyperthyroidism plus decrease in T3-RIA result due to:
Severe non thyroid illness
Advanced age
Certain medications (e.g., dexamethasone, propranolol)
False hyperthyroidism (patient is euthyroid)
Increased TBG value plus decrease in T3-RIA result
Severe non thyroid illness occasionally producing falsely elevated T4 levels
Increased TBG value with disproportionate T4 increase relative to T3-RIA
Acute psychiatric illness (some patients)
Amphetamine abuse
Certain x-ray contrast media
Certain medications (e.g., propranolol, amiodarone)
Specimen obtained 1-4 hr after levothyroxine dose rather than just before the dose
Patient taking therapeutic levothyroxine

when the TBG value is increased. Pseudo–T4 toxicosis may be produced in patients who have T4/T3 toxicosis if the T3-RIA level becomes decreased for some reason while the T4 level remains elevated. The most common causes for T3-RIA decrease in T4/T3 toxicosis are severe non thyroid illness and effect on T3-RIA of old age. Many severe non thyroid illnesses, particularly when chronic (see Table 28-3), depress T3-RIA levels, often to very low levels. The free T3 index also decreases but to a lesser extent. The effect of severe illness persists for variable periods of time and usually involves a shift from production of T3 toward reverse T3. Another factor that depresses T3-RIA levels but not T4 levels is the effect of advanced age. For patients over age 60 years, most T3-RIA kits have demonstrated a progressive decrease with time of approximately 10%-30% (literature range, 0%-52%). The degree of effect differs with individual manufacturers’ kits. Unfortunately, very few laboratories determine age-related values for the particular kit that they use. There is general but not unanimous agreement that T4 values are not greatly changed in old age. Certain medications (propranolol, dexamethasone) have been reported to decrease T3-RIA levels, although not severely.

False hyperthyroidism. Artifactual T4 elevation may result when TBG levels are increased, artifactually elevating both T4 and T3-RIA results, but some condition is superimposed that decreases T3-RIA results, leaving only the T4 value elevated. As noted previously, the most common reason for artifactual T3-RIA decrease is severe non thyroid illness. Another possibility is the effect of advanced age. In patients with normal TBG levels, severe non thyroidal illness may be associated with T4 values that are increased, decreased, or that remain within normal population range. Thyroxine levels most commonly display slight or mild decreases but still remain within normal limits. In a significant minority of patients (depending on the severity of illness), the T4 level is decreased below its reference range to varying degrees, producing pseudo hypothyroidism (clinical euthyroidism with laboratory test results falsely suggesting hypothyroidism). A small minority of patients exhibit an increase in T4 results for poorly understood reasons. In these patients, Pseudo–T4 toxicosis would be produced without clinical hyperthyroidism or increased TBG levels (designated “T4 euthyroidism” by some investigators). The TSH value in severe non thyroidal illness is most often normal but may be mildly increased. The THBR level may be normal but is sometimes mildly increased, reflecting decreased TBG levels. Occasionally the THBR level is decreased, mainly in acute hepatitis.

Certain conditions produce artifactual elevation of T4 values but not T3-RIA values. In some patients with increased TBG values without severe non thyroid illness, T3-RIA values remain within upper reference range while the T4 levels are elevated. One explanation is that increase in binding proteins affects T4 levels somewhat more than T3-RIA values. However, an increased TBG level frequently produces an elevated T3-RIA value as well as a T4 value. Elevated T4 values with normal T3-RIA values have been reported in some patients with acute psychiatric illness who were clinically euthyroid and where T4 values returned to the reference range after treatment of the psychiatric problem. However, the possibility of true thyrotoxicosis should not be ignored.

Amphetamine abuse has been reported to increase serum T4 values without affecting T3-RIA values. Both the TI and FT3I are elevated in some of these patients. Some of these patients had mildly elevated serum TSH values and some did not. Increased FT3I was also found in some cases without increase of FT4 index.

Certain x-ray contrast media such as ipodate and iopanoic acid gallbladder visualization agents decrease T3-RIA values and may increase T4 values somewhat. Dexamethasone and propranolol are reported to decrease T3-RIA values, as noted previously. Propranolol has been reported to increase T4 values, but there is some disagreement as to whether this occurs.

If a patient is taking therapeutic levothyroxine (Synthroid, Levothroid), and a blood specimen happens to be drawn 1-4 hours after a dose has been administered, a result above steady-state level will often be obtained that might be above the reference range. Peak values after an oral dose are reached in 2-4 hours and average 1-3 µg above steady-state level. Even at steady-state levels and drawn just before the scheduled dose, patients who are clinically normal and whose TSH and T3-RIA values are within reference range may have a steady state T4 level as much as 2 µg above the upper limit of the T4 reference range (discussed later). This is a problem because the dose is not always given at the scheduled time, the laboratory usually does not know what medications the patient is receiving to schedule the time of venipuncture, and sometimes the physician is unaware that a new patient is taking levothyroxine.

It has been reported that some hyperthyroid patients with elevated T4 levels but normal-range T3-RIA values have an elevated FT3I.

Hyperthyroidism with false laboratory euthyroidism. One final category of deceptive hyperthyroidism may be added, clinical hyperthyroidism with falsely normal T4 and T3-RIA values (see the box). Patients with decreased TBG

Hyperthyroidism With False Laboratory Euthyroidism
Hyperthyroidism plus decreased TBG value (see the box)
Hyperthyroidism plus severe non thyroid illness

levels may have falsely decreased T4 and T3-RIA levels that could convert elevated values to normal-range assay results. Severe non thyroidal illness decreases T3-RIA values and may decrease T4 values. This could mask expected T3-RIA elevation in T3 toxicosis and T3-RIA plus some patients with T4 elevation in some cases of T4/T3 or T4 toxicosis.

Isolated Graves’ ophthalmopathy

Besides the two classic types of clinical hyperthyroidism, there is one additional form known as isolated Graves’ ophthalmopathy, or “euthyroid Graves’ disease.” This consists of eye signs associated with hyperthyroidism but without other clinical evidence of thyrotoxicosis and with normal RAIU, T4, and T3-RIA values. Evidence of true hyperthyroidism consists of reports that about 50%-70% of these patients fail to demonstrate thyroid suppression on the T3 suppression test (literature range, 50%-100% in several small series of patients). About two thirds have a flat or blunted TSH response on the TRH test, suggestive of thyroid autonomy. A considerable number of these patients have detectable thyroid-stimulating immunoglobulins (TSI test;). However, published reports of the latest versions of this test show considerable differences in sensitivity between laboratories. Similar differences were reported in detection rates for Graves’ disease.