The Thyroid and its Diseases
	Chapter 11                                           HOME Diagnosis and Treatment of Graves’ Disease Last updated- 5 Feb 2007 -by Leslie J. De Groot, MD TO OBTAIN A DOWNLOAD OF THIS CHAPTER IN PDF OR WORD,  CLICK HERE

The diagnosis of Graves' disease is usually easily made. The combination of eye signs, goiter, and any of the characteristic symptoms of hyperthyroidism forms a picture that can hardly escape recognition (Fig 11-1). It is only in the atypical cases, or with coexistence of some other disease, or in cases in which the disease is so mild or early as to be unconvincing, that the diagnosis may be in doubt.The symptoms and signs of Graves' disease have been described in detail in the preceding chapter. For convenience, the classic findings from the history and physical examination are grouped together in Table 11-1.These occur with sufficient regularity that clinical diagnosis can be reasonably accurate. Scoring the presence or absence and severity of particular symptoms and signs can provide a clinical diagnostic index almost as reliable a diagnostic measure as laboratory tests. The frequency of signs and symptoms, adapted from the index of Wayne and co-workers ¹ appears in Table 11-2.

Figure 11-1. Graves' disease patient with exophthalmos and vitiligo.

Occasionally diagnosis is not at all obvious- in patients severely ill with other disease, elderly patients with "apathetic hypothyroidism", or when the presenting symptom is unusual, such as muscle weakness, or psychosis, and diagnosis depends on clinical alertness and laboratory tests.

We should note that the diagnosis of Graves’ Disease does not depend on thyrotoxicosis. Ophthalmopathy, or peretibial myxedema are seen without goiter and thyrotoxicosis, or even with spontaneous hypothyroidism. While proper classification can be debated, these patients seem to represent one end of the spectrum of Graves’ Disease.

Table 11-1. Symptoms and Signs of Graves' Disease

Table 11-2. Incidence (Percentage) of Common Signs and Symptoms in Thyrotoxic Patients and Controls

Once the question of thyrotoxicosis has been raised, laboratory data are required to verify the diagnosis, help estimate the severity of the condition, and assist in planning therapy. The numerous techniques of laboratory assessment are critically reviewed in Chapter 6. A single test such as the TSH or estimate of fT4  may be enough, but in view of the sources of error in all determinations, most clinicians prefer to assess two more or less independent measures of thyroid function. For this purpose, an assessment of fT4 and sensitive TSH are suitable.

As an initial single test, a sensitive TSH assay may be most cost-effective and specific. TSH should be 0 - .1 µU/ml in significant thyrotoxicosis, although values of .1 - .3 are seen in patients with mild illness, especially with smoldering toxic multinodular goiter in older patients.  TSH can be low in some elderly patients without evidence of thyroid disease ^1.1  TSH can be normal -- or elevated -- only if there are spurious test results from antibodies, or the thyrotoxicosis is TSH-driven, as in a pituitary TSH-secreting adenoma or pituitary resistance to thyroid hormone. Measurement of FTI (or any measure of free T₄) is also useful, and the degree of elevation of the fT4 above normal provides an estimate of the severity of the disease. During replacement therapy with thyroxine the range of both T4 and fT4  values tend to be about 20% above the normal range, possibly because only T4, rather than T4 and T3 from the thyroid, is providing the initial supply of hormone. Thus many patients will have a T4 or FTI above normal when appropriately replaced and while TSH is in the normal range. Except for this, elevations of FTI not due to thyrotoxicosis are unusual, and causes are given in Table 11-3. Of course the T₄ level may normally be as high as 16 or 20 µg/dl in pregnancy, and can be elevated without thyrotoxicosis in patients with familial hyperthyroxinemia due to abnormal albumin, the presence of hereditary excess TBG, the presence of antibodies binding T₄, the thyroid hormone resistance syndrome, and conditions listed in Table 11-3. The T₄ level may be normal in thyrotoxic patients who have depressed serum levels of T₄-binding protein or because of severe illness, even though they are toxic. Thus, thyrotoxicosis may exist when the T₄ level is in the normal range; measurement of free T₄ or fT4  usually obviates this source of error and is the best test. In the presence of typical symptoms, one measurement of suppressed TSH or elevated FTI is sufficient to make a definite diagnosis, although it does not identify a cause. If the fT4  is normal, repetition is in order to rule out error, along with a second test such as serum T₃.

A variety of methods for free T4 determination have become available, including commercial kits. Although these methods are usually reliable, assays using different kits do not always agree among themselves or with the determination of free T₄ by dialysis.  Usually T₄ and T₃ levels are both elevated in thyrotoxicosis, as are the FTI and an index constructed using the serum T₃ and rT₃U levels, and the newer measures of "free T₄" or "free T₃".

The serum T₃ level determined by RIA is almost always elevated in thyrotoxicosis and is a useful but not commonly needed secondary test. Usually the serum T₃ test is interpreted directly without use of a correction for protein binding, since alterations of TBG affect T₃ to a lesser extent than T₄. Any confusion caused by alterations in binding proteins can be avoided by use of a "free T₃" assay or T₃ index calculated as for the FTI. In patients with severe illness and thyrotoxicosis,^4,5 especially those with liver disease or malnutrition or who are taking steroids or propranolol, the serum T₃ level is not elevated, since peripheral deiodination of T₄ to T₃ is suppressed ("T₄ toxicosis"). A normal T₃ level has also been observed in thyrotoxicosis combined with diabetic ketoacidosis.⁶ Whether or not these patients actually have tissue hypermetabolism at the time their serum T₃ is normal is not entirely certain. In these patients the rT₃ level may be elevated. If the complicating illness subsides, the normal pattern of elevated T₄, FTI, and T₃ levels may return. Elevated T₄ levels with normal serum T₃ levels are also found in patients with thyrotoxicosis produced by iodine ingestion. ⁷

Table 11-3. Conditions Associated with Transient  Elevations of the FTI

T₃ Toxicosis

Since 1957, when the first patient with T₃ thyrotoxicosis was identified, a number of patients have been detected who had clinical thyrotoxicosis, normal serum levels of T₄ and TBG, and elevated concentrations of T₃ and FT₃. ⁸ Hollander et al.⁹ found that approximately 4% of patients with thyrotoxicosis in the New York area fit this category. These patients often have mild disease but otherwise have been indistinguishable clinically from others with thyrotoxicosis. Some have had the diffuse thyroid hyperplasia of Graves' disease, others toxic nodular goiter, and still others thyrotoxicosis with hyperfunctioning adenomas. Interestingly, in Chile, a country with generalized iodine deficiency, 12.5% of thyrotoxic subjects fulfilled the criteria for T₃ thyrotoxicosis. ¹⁰ Asymptomatic hypertriiodothyronemia is an occasional finding several months before the development of thyrotoxicosis with elevated T₄ levels. ¹¹ Since T₄ is normally metabolized to T₃, and the latter hormone is predominantly the hormone bound to nuclear receptors, it makes sense that elevation of T₃ alone can produce thyrotoxicosis.

RAIU

In patients with thyrotoxicosis the RAIU at 24 hours is characteristically above normal. In the United States, which has had an increasing iodine supply in recent years, the upper limit of normal is now about 25% of the administered dose. This value is higher in areas of iodine deficiency and endemic goiter. The uptake value at a shorter time interval, for example 6 hours, is as valid a test and may be more useful in the infrequent cases having such a rapid isotope turnover that "uptake" has fallen to normal by 24 hours. If there is reason to suspect that thyroid isotope turnover is rapid, it is wise to do both a 6- and a 24-hour RAIU determination during the initial laboratory study. As noted below,rapid turnover of 131-I can seriously reduce the effectiveness of 131-I therapy. Because of convenience, and since serum assays of thyroid hormones and TSH are reliable and readily available, the RAIU is infrequently determined unless ¹³¹I therapy is planned. A drawback of this approach is that cases of transient thyrotoxicosis (described below) may be missed unless the typical low RAIU is recognized. To avoid errors, we recommend that the RAIU test be done in patients who are believed to be thyrotoxic with suppressed TSH, but who do not have typical symptoms and/or signs. This may include patients with brief symptom duration, small goiter, or lacking eye signs, absent family history, or negative antibody test results ¹². Obviously other causes of a low RAIU test need to be considered and excluded.

Thyroid Scanning

Scanning of the thyroid has a limited role in the diagnosis of thyrotoxicosis, except in those patients in whom the thyroid is difficult to feel or in whom nodules are present that require evaluation, or rarely to prove the function of ectopic thyroid tissue. Nodules may be incidental, or may be the source of thyrotoxicosis (toxic adenoma), or may contribute to the thyrotoxicosis that also arises from the rest of the gland. Scanning should usually be done with ¹²³I in this situation, in order to combine it with an RAIU measurement.

Iodide fluorescence scanning is available in some institutions and can be used to delineate the anatomy of the thyroid when the use of radioactive isotopes is contraindicated (pregnancy, lactation) or when the uptake is suppressed by excess iodides. This technique is only used as a research tool. Fluorescence scanning has also been used to quantitate the thyroidal iodine content, which is not usually elevated in Graves' disease, at least in the United States.

Determination of antibody titers provides supporting evidence for Graves' disease. More than 95% of patients have positive assays for TPO-microsomal antigen, and about 50% have positive anti-thyroglobulin antibody assays. In thyroiditis the prevalence of positive TG antibody assays is higher. Positive assays prove that autoimmunity is present, but they do not prove thyrotoxicosis. However, patients with causes of thyrotoxicosis other than Graves' disease usually have negative assays. During therapy with antithyroid drugs the titers characteristically go down, and this change persists during remission. Titers tend to become more elevated after RAI treatment.

ANTIBODIES TO TSH-RECEPTOR

Thyroid stimulating antibody assays have become  readily available, and a positive result supports the diagnosis ^10.1. The assay is valuable as another supporting fact in establishing the cause of exophthalmos, in the absence of thyrotoxicosis, and high levels may predict neonatal thyrotoxicosis. However, the test is expensive, and is only rarely needed. Measurement of TBII (TSH-R binding antibodies) is generally available, and while not as specific as TSAb, also can be used to support the diagnosis of Graves. Using current tests, both are positive in about 90% of patients with Graves disease who are thyrotoxic. "Second generation" assays becoming available use monoclonal anti-TSH-R antibodies and biosynthetic TSH-R in coated tube assays, which reach 99% specificity and sensitivity(10.11). Although rarely required, serial assays are of interest in following a patient’s course during antithyroid drug therapy, and a decrease predicts probable remission ^11.1,11.2.

Determination of the BMR might be of theoretical interest, but is not a good diagnostic test and is often unreliable when done for the first time. It requires a skilled technician and appropriate equipment, both of which are not generally available.

Other Assays

General availability of assays that can reliably measure suppressed TSH has made this the gold standard to which other tests must be compared, and has effectively eliminated the need for most previously used ancillary tests. There are only rare causes of confusion in the sTSH assay. Severe illness, dopamine and steroids, and hypopituitarism, can cause low sTSH, but suppression below 0.1 µ/ml is uncommon and below 0.05 µ/ml is exceptional, except in thyrotoxicosis. Thyrotoxicosis is associated with normal or high TSH in patients with TSH producing pituitary tumors and selective pituitary resistance to thyroid hormone.

If these procedures do not establish the diagnosis, it may be wise to do nothing further except to observe the course of events. In patients with significant thyroid hyperfunction, the symptoms and signs will become clearer, and the laboratory measurements will fall into line.

In past years it was common to try to resolve confusion by use of a T₃ suppression test or TRH test. It must be remembered that the T₃ suppression test may be positive in Graves' disease, in the absence of thyrotoxicosis, since it measures "thyroid autonomy" and not hyperthyroidism per se. The test result is also positive in hyperfunctioning adenomas and in some glands having the histologic picture of Hashimoto's disease. The cumbersome and occasionally dangerous T₃ suppression test was largely discarded in favor of the simpler TRH test. An increase in TSH level after TRH administration is regularly absent in patients with hyperthyroid Graves' disease; the response in those with euthyroid Graves' disease is often but not always absent. Lack of response is strong presumptive evidence of chemical, if not clinical, thyrotoxicosis. Ormston et al. ¹³ found the TRH test to correlate fairly closely with the T₃ suppression test in patients with exophthalmos. The response was usually exaggerated when the patient was borderline hypothyroid and low or absent when the patient was borderline thyrotoxic. Others have found a poor correlation between suppressibility, the TRH response, and the course of exophthalmos.¹⁴ Unresponsiveness to TRH may also be found in patients with treated Graves'. ¹⁵

The vast majority of patients are diagnosed by elevated fT4  or T₃ levels and suppressed TSH, and TRH testing is rarely indicated. Since there is very high correlation between basal TSH levels and response to TRH, TRH testing provides no additional information.

In the past, some clinicians placed diagnostic reliance on the striking response of patients with Graves' disease to the administration of iodine. If 6 mg iodine or more is given daily to a person who has Graves' disease and is not already receiving iodine, within the succeeding 7 or 10 days there will be an amelioration in symptoms, and the FTI level will fall in parallel. Often the patient may reach a euthyroid state, at least temporarily. If iodine administration is then stopped, the signs and symptoms quickly return to their previous state. Since iodine administration interferes with treatment by antithyroid drugs and with ¹³¹I therapy, this may prove to be a difficulty. The therapeutic trial with iodine is thus  of historical interest.