Chapter 9 The thyroid gland
The thyroid gland secretes three hormones: thyroxine (T4) and triiodothyronine (T3), both of which are iodinated derivatives of tyrosine (Fig. 9.1), and calcitonin, a polypeptide hormone. T4 and T3 are produced by the follicular cells but calcitonin is secreted by the C cells, which are of separate embryological origin. Calcitonin is functionally unrelated to the other thyroid hormones. It has a minor role in calcium homoeostasis and disorders of its secretion are rare (see Chapter 12). Thyroid disorders in which there is either over- or under-secretion of T4 and T3 are, however, common.
Thyroxine synthesis and release are stimulated by the pituitary trophic hormone, thyroid-stimulating hormone (TSH). The secretion of TSH is controlled by negative feedback by the thyroid hormones (see p. 119), which modulate the response of the pituitary to the hypothalamic hormone, thyrotrophin-releasing hormone (TRH; Fig. 9.2). This feedback is mediated primarily by T3 produced by the action of iodothyronine deiodinase on T4 in the thyrotroph cells of the anterior pituitary. Glucocorticoids, dopamine and somatostatin inhibit TSH secretion. The physiological significance of this is not known, but it may be relevant to the disturbances of thyroid hormones that can occur in non-thyroidal illness (see p. 160). The feedback mechanisms result in the maintenance of steady plasma concentrations of thyroid hormones.
Figure 9.2 Control of thyroid hormone secretion. TSH is released from the pituitary in response to the hypothalamic hormone, TRH, and stimulates the synthesis and release of thyroid hormones. TSH release is inhibited by thyroid hormones, which decrease the sensitivity of the pituitary to TRH. They may also inhibit TRH release by the hypothalamus.
The major product of the thyroid gland is T4. Ten times less T3 is produced (the proportion may be greater in thyroid disease), most (approximately 80%) T3 being derived from T4 by deiodination in peripheral tissues, particularly the liver, kidneys and muscle, catalysed by selenium-containing iodothyronine deiodinases. T3 is 3–4 times more potent than T4. In tissues, most of the effect of T4 results from this conversion to T3, so that T4 itself is essentially a prohormone. Deiodination can also produce reverse triiodothyronine (rT3; see Fig. 9.1), which is physiologically inactive. It is produced instead of T3 in starvation and many non-thyroidal illnesses, and the formation of either the active or inactive metabolite of T4 appears to play an important part in the control of energy metabolism.
Thyroid hormones are essential for normal growth and development and have many effects on metabolic processes. They act by entering cells and binding to specific receptors in the nuclei, where they stimulate the synthesis of a variety of species of mRNA, thus stimulating the synthesis of polypeptides, including hormones and enzymes. Among the latter are key enzymes involved in energy metabolism, including cytochrome oxidase. Their most obvious overall effect on metabolism is to stimulate the basal metabolic rate, oxygen consumption and heat production, through actions that include stimulating Na+,K+-ATPases involved in ion transport and increasing the availability of energy substrates. Overall, the effect of thyroid hormones is to increase net catabolism: weight loss and muscle wasting are typical features of excessive secretion of thyroid hormones. Thyroid hormones also increase the sensitivity of the cardiovascular and nervous systems to catecholamines, the former leading to increases in heart rate and cardiac output, and the latter to increased arousal.
Thyroid hormone synthesis involves a number of specific enzyme-catalysed reactions, beginning with the uptake of iodide by the gland and culminating in the iodination of tyrosine residues in the protein thyroglobulin (Fig. 9.3); these reactions are all stimulated by TSH. Rare, congenital forms of hypothyroidism caused by inherited deficiencies of each of the various enzymes concerned have been described.
Figure 9.3 Biosynthesis of the thyroid hormones. The iodination and condensation reactions involve tyrosine residues that are an integral part of the thyroglobulin polypeptide. The thyroid hormones remain protein bound until they are released from the cell. Iodide is actively absorbed into thyroid cells and oxidized to iodine, which is immediately incorporated into tyrosine residues to form monoiodotyrosine (MIT) and diiodotyrosine (DIT). These undergo coupling to form T3 and T4. Anti-thyroid thionamide drugs, such as carbimazole, act by inhibiting the oxidation of iodide and the coupling reaction.
Thyroglobulin is stored within the thyroid gland in colloid follicles. These are accumulations of thyroglobulin-containing colloid surrounded by thyroid follicular cells. Release of thyroid hormones (stimulated by TSH) involves pinocytosis of colloid by follicular cells, fusion with lysosomes to form phagocytic vacuoles, and proteolysis (Fig. 9.4). Thyroid hormones are thence released into the bloodstream. Proteolysis also results in the liberation of mono- and diiodotyrosines (MIT and DIT); these are usually degraded within thyroid follicular cells and their iodine is retained and re-utilized. A small amount of thyroglobulin also reaches the bloodstream.
The normal plasma concentrations of T4 and T3 are 60–150 nmol/L and 1.0–2.9 nmol/L, respectively. Both hormones are extensively protein bound: some 99.98% of T4 and 99.66% of T3 are bound, principally to a specific thyroxine-binding globulin (TBG) and, to a lesser extent, to prealbumin and albumin. TBG is approximately one-third saturated at normal concentrations of thyroid hormones (Fig. 9.5). It is generally accepted that only the free, non-protein-bound, thyroid hormones are physiologically active. Although the total T4 concentration is normally 50 times that of T3, the different extents to which these hormones are bound to protein mean that the free T4 concentration is only 2–3 times that of free T3 (typical reference ranges are 9–26 pmol/L for free T4 and 3.0–9.0 pmol/L for free T3).
The precise physiological function of TBG is unknown; individuals who have a genetically determined deficiency of the protein show no clinical abnormality. It has, however, been suggested that the extensive binding of thyroid hormones to TBG provides a buffer that maintains the free hormone concentrations constant in the face of any tendency to change. Protein binding also reduces the amount of thyroid hormones that would otherwise be lost by glomerular filtration and subsequent renal excretion.
Total (free + bound) thyroid hormone concentrations in plasma are dependent not only on thyroid function but also on the concentrations of binding proteins. If these were to increase (Fig. 9.6), the temporary fall in free hormone concentration caused by increased protein binding would stimulate TSH release and this would restore the free hormone concentrations to normal: if binding protein concentrations were to fall, the reverse would occur. In either situation, there would be a change in the concentrations of total hormones, but the free hormone concentrations would remain normal.
Figure 9.6 Effect of an increase in TBG concentration on plasma T4 concentration. (A) In the initial steady state, TBG is one-third saturated with T4. (B) TBG concentration increases, causing more T4 to be bound, thus reducing the free T4 concentration. This stimulates TSH secretion, which leads to an increase in the release of T4 from the thyroid. (C) T4 becomes redistributed between the bound and the free states, leading to a new steady state with the same free T4 concentration but an increased total T4.
This is a matter of considerable practical importance, as changes in the concentrations of the binding proteins occur in many circumstances (Fig. 9.7), causing changes in total hormone concentrations but not necessarily in those of the free (physiologically available) hormones. Furthermore, certain drugs, for example salicylates and phenytoin, displace thyroid hormones from their binding proteins, thus reducing total, but not free, hormone concentrations once a new steady state is attained. If an attempt is made to assess thyroid status in a patient who is not in a steady state, the results may be bizarre and misleading.
Only small amounts of T4 and T3 are excreted by the kidneys owing to the extensive protein binding. The major route of thyroid hormone degradation is by deiodination and metabolism in tissues, but they are also conjugated in the liver and excreted in bile.
Laboratory tests of thyroid function are required to assist in the diagnosis and monitoring of thyroid disease. Most laboratories offer a standard ‘profile’ of thyroid function tests (e.g. TSH and free T4), and perform additional tests only if these results are equivocal or the clinical circumstances require it.
Measurement of plasma total T4 (tT4) concentration was formerly widely used as a test of thyroid function, but this test has the major disadvantage in that it is dependent on binding protein concentration as well as thyroid activity. For example, a slightly elevated plasma tT4 concentration, compatible with mild hyperthyroidism, can occur with normal thyroid function if there is an increase in plasma binding protein concentrations. With the introduction of reliable assays for free T4 (fT4), there is now little, if any, justification for laboratories continuing to measure tT4 as a test of thyroid function.
Plasma total T3 (tT3) concentration is almost always raised in hyperthyroidism (usually to a proportionately greater extent than tT4, and hence it is the more sensitive test for this condition) but may be normal in hypothyroidism owing to preferential production of T3 in the thyroid and increased peripheral formation from T4. However, tT3 concentrations, like those of tT4, are dependent on the concentration of binding proteins in plasma, and their measurement has been largely superseded by measurements of free T3 (fT3).
The measurement of free hormone concentrations poses major technical problems because the binding of free hormones in an assay, usually by an antibody, will disturb the equilibrium between bound and free hormone and cause release of hormone from binding proteins. Various techniques have been developed that allow the estimation of fT4 and fT3 concentrations in plasma. Such measurements, in theory, circumvent the problems associated with protein binding, and have rendered obsolete techniques for the indirect assessment of free hormone concentrations, such as the resin uptake test, calculation of the free thyroxine index or measurement of the T4/TBG ratio. However, with gross abnormalities of binding protein concentrations, the results of measurements of free hormones may be misleading owing to technical limitations of the assays.
In pregnancy, TBG concentration increases, owing to increased synthesis stimulated by oestrogens, and leads to an increase in tT4. The fT4 concentration may rise slightly in early pregnancy, probably as a result of the weak thyroid-stimulating properties of chorionic gonadotrophin, but returns to normal values by 20 weeks; it may fall somewhat in the third trimester, but in most women fT4 remains within the non-pregnant reference range. Measured values should be compared with trimester-specific (and method-specific) reference ranges.
Just as the tT3 concentration can be normal in hypothyroidism (especially in mild cases), so too can the fT3 concentration, and its measurement is of no value in the diagnosis of this condition. Free T3 is, however, a sensitive test for hyperthyroidism. In hyperthyroid patients, both fT4 and fT3 are usually elevated (fT3 to a proportionately greater extent), but there are exceptions to this. In a small number of patients with hyperthyroidism, the fT3 concentration is elevated but fT4 is not (although it is usually high–normal)—a condition called ‘T3-toxicosis’. Occasionally, fT4 is elevated but not fT3. This is usually due to concomitant non-thyroidal illness resulting in decreased conversion of T4 to T3, and the fT3 concentration increases when this illness resolves.
If primary thyroid disease is suspected and the plasma TSH concentration is normal, it can be safely inferred that the patient is euthyroid. In overt primary hypothyroidism, TSH concentrations are greatly increased, often to ten or more times the upper limit of normal. Smaller increases are seen in borderline cases, but TSH measurement is more sensitive than T4 under these circumstances: TSH concentrations rise above the reference range before those of T4 fall below it. TSH can also increase transiently during recovery from non-thyroidal illness (see below). Plasma TSH concentrations are suppressed to very low values in hyperthyroidism, but low concentrations can also occur in individuals with subclinical hyperthyroidism and in euthyroid patients with non-thyroidal illness (’sick euthyroidism’, see below). Indeed, in hospital patients, a low plasma TSH concentration is more often due to non-thyroidal illness than to hyperthyroidism, while a slightly elevated concentration is as frequently due to recovery from such illness as to mild or incipient hypothyroidism.
Clinical biochemistry laboratories undertake large numbers of tests of thyroid function. To simplify their procedures, many adopt the approach of measuring TSH as a first-line test of thyroid function, adding other tests as required, for example if the concentration of TSH is found to be outside the euthyroid reference range or if there is a strong suspicion that thyroid dysfunction is secondary to pituitary disease (although this is far less common than primary thyroid dysfunction). A combination of tests may also be required to assess patients being treated for thyroid disease, particularly in the early stages.
It should be noted that immunometric assays (such as are used for TSH) are subject to interference by naturally occurring heterophilic antibodies against the monoclonal antibodies used in the assay; such interference occurs only infrequently, but can give rise to apparently high results. When the results of assays do not accord with those expected from the patient’s clinical condition, it may be prudent to repeat them using an alternative method.
In this test, plasma TSH is measured immediately before, and 20 and 60 min after, giving the patient 200 µg of TRH i.v. (Fig. 9.9). The normal response is an increase in TSH concentration of 2–20 mU/L in 20 min, with reversion towards the basal value at 60 min.