Dramatic, although transient, increases in serum TSH levels have been observed in infants and young children during surgical hypothermia.[2] Also, a prompt and important secretion of TSH occurs in the newborn, in the first few hours after birth, accompanied by an increase in thyroid hormone secretion and clearance.[3,4] Since this TSH surge is partially prevented by maintaining infants in a warm environment, postnatal cooling appears to be responsible in part for the rise in TSH secretion. In most studies, exposure of adults to cold or even intensive hypothermia has produced no changes,[5,6] or at best minimal increases[7] in serum TSH. More prolonged exposure to cold generally results in maintenance of the total T4 (TT4) and free T4 (fT4) levels with maintenance of a normal or decreased total T3 (TT3) and free T3 (fT3) levels.[7a,7b], however, others have shown prolonged arctic residence leads the increase in TSH to be associates with an increase in, thyroglobulin and T3.[7c] These alterations may be partly the consequence of a direct effect of temperature on the rate and pathways of thyroid hormone metabolism with more rapid production and clearance of T3. Altered kinetics have been demonstrated in humans[7d], but have been more thoroughly studied in animals.[8,9,9a,9b] It has been more difficult to show a clear seasonal variation in serum hormone concentration. However, the variation demonstrated in several studies[10,11] has been that T4 and T3 values are higher during the colder months.

Cold exposure in animals leads to thyroid gland hyperplasia, enhanced hormonal secretion, degradation, and excretion, accompanied by an increased demand for dietary iodine. All of these effects are presumably due to an increased need for thyroid hormone by peripheral tissues. The prompt activation of pituitary TSH secretion after cold exposure of the rats[12,13] is possibly due in part to a direct effect on the hypothalamus.[14] Exposure to cold has also resulted in augmented TRH production, and serum levels,[16] and blunted responses of TSH to exogenous TRH.[17] These effects have not been reproduced by other laboratories[13,18] although an increase in thyroid hormone secretion has been clearly demonstrated.[6,19,20] In the rat, it is associated with augmented rates of T4 and T3 deiodination, increased conversion of T4 to T3, and enhanced hepatic binding and biliary and fecal clearance of the iodothyronines.[8,9,9a,21,22] Finally, thyroid hormone effects may be enhanced by alterations in co-activators which enhance the activity of thyroid hormone receptors on gene activation. [22a]

In general, an increase in ambient temperature has produced effects opposite to those observed during cold exposure, although the effects of heat have not been extensively investigated. As indicated above, thyroid hormone levels in serum tend to be lower during the summer months. A decrease in the serum T3 concentration, with reciprocal changes in the levels of rT3, have been observed in normal subjects acutely exposed to heat and during febrile illnesses.[23,24] In the latter condition, the contribution of the rise in body temperature relative to other effects of systemic illness cannot be dissociated. A decrease in the elevated serum TSH level associated with primary hypothyroidism has been induced by increases in body temperature.[25]

Multiple alterations in thyroid hormone regulation and metabolism have been noted during caloric restriction. The most dramatic effect is a decrease in the serum TT3 within 24-48 hours of the initiation of fasting.[36-40b] Because changes in the free T3 fraction are usually small, the absolute concentration of FT3 is also reduced, clearly into the hypothyroid range The marked reduction in serum T3 is caused by a reduction in its generation from T4 rather than by an acceleration in its metabolic clearance rate.[41,42] The decline in T3 concentration is accompanied by a concomitant and reciprocal change in the concentration of total and free rT3. The increase in the serum rT3 concentration tends to begin later and to return to normal at the time serum T3 is being maintained at a low level with continuous calorie deprivation.[38,39] Little change occurs in the concentrations of TT4 and FT4 and the production and metabolic clearance rates of T4.[38,39,41,42] When small changes have been observed, they were generally in the direction of an increase in the FT4 concentration. They are attributed to decreased concentration of the carrier proteins in serum, as well as to their diminished association with the hormone caused by the inhibitory effect of free fatty acids (FFA) the level of which increases during fasting.[40,43]

Decreased outer ring monodeiodination (5'-deiodinase activity) would explain both the decreased generation of T3 from T4 and the excess accumulation of rT3. This hypothesis seems to be fully supported by in vitro studies using liver tissue from fasted fats.[44] It is further supported by the finding of increased generation and serum concentration of 3',5'-T2 and 3'-T1 and decreased 3,5-T2 and 3,3'-T2.[44-47] However, a less important increase in the monodeiodination of the inner ring of T4 (5-deiodination)[42] explains the temporal dissociation of changes in serum T3 and rT3 concentration. A decrease in plasma T3 after fasting with an increase in hepatic type III deiodinase activity and mRNA has also been noted in chickens. [47a] An increase in the nondeiodinative pathway of T4 degradation with the formation of Tetrac has been also reported.[48]

Considerable controversy remains regarding the mechanisms responsible for the observed changes in the rates of the deiodinative pathways of iodothyronines. Decreased generation of nonprotein sulfhydryls (NP-SH) as a cause of the reduction in 5'-deiodinase activity was suggested on the basis of the observed enhancement in enzyme activity by the in vitro addition of dithiothreitol. Reduced glutathione and NADPH had a similar effect.[49] Although Chopra's[50] direct measurements of NP-SH in tissue during fasting seemed to confirm this hypothesis, the precise mechanism is likely more complex. Decreased tissue NP-SH content does not always correlate with the inhibition of T3 generation, which may be restored by glucose refeeding independently of changes in NP-SH content.[50,51]

Composition of the diet rather than reduction in the total calorie intake seems to determine the occurrence of decreased T3 generation in peripheral tissues during food deprivation. The dietary content of carbohydrate appears to be the key ingredient since as little as 50 g glucose reverses toward normal the fast-induced changes in T3 and rT3.[52] Replacement of dietary carbohydrate with fat results in changes typical of starvation.[39,53] Refeeding of protein may partially improve the rate of T3 generation, but the protein may be acting as a source of glucose through gluconeogenesis.[54] Yet, dietary glucose is not the sole agent responsible for all changes in iodothyronine metabolism associated with starvation. For example, the increase in serum rT3 concentration may not be solely dependent on carbohydrate deprivation since a pure protein diet partially restores the level of rT3 but not that of T3[39] (Fig. 5-1). The composition of the antecedent diet also has an effect on the magnitude of the serum T3 fall during fasting.[39,52] It is possible that the cytoplasmic redox state, measured in terms of the lactate/pyruvate ratio rather than glucose itself, regulates the rate of deiodinative pathways of iodothyronines.[55]

The basal serum TSH level during calorie deprivation is either normal or low, the response to TRH is blunted[37-39] and the normal nocturnal rise in TSH is blunted.[40a] These changes are quite surprising given the consistent and profound decrease in serum FT3 levels. Several hypothesis have been proposed to explain this paradox. Because the pituitary is able to continue to respond appropriately during fasting to both suppressive and stimulatory signals,[56] it has been suggested that starvation only "resets" the set point of feedback regulation. A more plausible hypothesis, supported by experimental data,[57,58] proposes that the pituitary is regulated by the intracellular concentration of T3, which may remain unaltered through factors ensuring its continuous local generation during starvation, whereas a decrease is typically found in other tissues. Further support for this hypothesis comes from a recent study demonstrating that fasting produces a marked increase in hypothalamic Type II Deiodinase mRNA[58a] which would enhance local T3 production. This hypothesis gives credence to the preservation of a closer inverse relationship between serum FT4 and TSH than between FT3 and TSH. Hypothalamic TRH content in starved rats has been reported to be normal,[59] low[60] or even elevated.[60a] The elevation of TRH was accompanied by normal levels of proTRH mRNA and decreased pituitary TSH; it was suggested that this represented decreased TRH release. [60a] In a different study of starved rats, the hypothalamic proTRH mRNA and the TRH content were both decreased,[60b] but these effects were reversed by adrenalectomy suggesting that they were secondary to increased glucocorticoid levels.[60b] Neonatal starvation in rats leads to diminished TRH and TSH production, with resultant hypothyroidism and growth retardation.[61]

Starvation produces a greater than 50% decrease in the maximal binding capacity of T3 to rat liver nuclear receptors within 48 hours.[62] Although accompanied by a diminution of almost equal magnitude in the nuclear T3 content, it is unlikely that the observed change represents an alteration of the receptor content by the hormone as the more profound diminution of nuclear T3 content associated with hypothyroidism does not produce changes in the maximal binding capacity of T3 in rat liver nuclei. The reduction in maximal binding capacity has been demonstrated to coincide with a reduction in the level of the thyroid hormone receptors.[62a] The affinity of the rat liver T3 receptor is not affected by starvation.[62,63] Studies in humans have used circulating mononuclear cells and, probably due to the limited choice of tissue, results have been either equivocal or negative.[64]

Other hormonal and metabolic changes during fasting may account for the observed alterations in the regulation and metabolism of thyroid hormones. Among them are the increase in plasma cortisol and suppression of adrenergic stimuli.[65] Both changes are known to induce independently a decrease in the serum T3 concentration by inhibition of T4 to T3 conversion in peripheral tissues (see below). Accordingly, they may be partly responsible for the decrease in T3 neogenesis during starvation. There is likely a highly complex interplay between the changes in thyroid hormone and the many metabolic changes of starvation. In addition to a direct effect of glucose, changes in FFA, ketosis, and the redox state may influence thyroid hormone metabolism, while T3 itself may impact hepatic glucose production.[40b]

Two major issues of theoretical and practical importance remain unresolved - do the observed changes in thyroid function produce some degree of hypothyroidism, and is this state beneficial to the energy-deprived organism? Although the suppressed serum TSH response to TRH suggests that the starving organism does not suffer from a significant deprivation in thyroid hormone, other observations indicate the contrary. The decreased pulse rate, systolic time interval, oxygen consumption, and decrease in activity of some liver enzymes are suggestive of hypothyroidism at the level of peripheral tissues.[66] Furthermore, administration of T3 to restore its serum level to normal during fasting increased the production and excretion of urea and 3-methylhistidine.[56,67] Larger doses of T3, given during fasting, had even more profound effects. These effects included dramatic increased in the excretion of urea and creatine, and increased plasma levels of ketones and FFA indicating an accelerated protein and fat breakdown.[68] Such evidence leaves little doubt that the decrease in T3 generation during calorie deprivation has an energy- and nitrogen-sparing effect. It is tempting to speculate that the result is beneficial in the adaptation to malnutrition through reduction in metabolic expenditure.

Fasting is not only a useful model for studying the effects of calorie deprivation on thyroid hormone but is also the prototype of the "low T3 syndrome".[69] The latter is produced by a number of chemical agents and drugs, and accompanies a variety of nonthyroidal illnesses. It is possible that malnutrition, concomitant in a number of acute and chronic illnesses, is in part responsible for some of the observed changes in thyroid physiology.

As in the case of starvation, PCM is associated with a low serum T3 concentration and increased rT3 levels, probably due to similar changes in iodothyronine monodeiodination. However, important differences exist between the abnormalities in thyroid function observed in PCM and acute calorie deprivation. Most reports indicate important decreases in TBG and TTR concentrations, and there are also indications of hormone binding abnormalities.[70,71] As a consequence, the free concentrations of both T4 and T3 are usually normal.[70,72,72a] Recovery is associated with restoration of the level of serum thyroid hormones and binding proteins. Despite an accelerated turnover time, the absolute amount of extrathyroidal T4 disposed each day is reduced. Refeeding restores the T4 kinetics to normal.[70] The thyroidal RAIU is reduced due to a defect in the iodine-concentrating mechanism.[73] The most striking difference between starvation and PCM is the finding the latter of an exaggerated and sustained TSH response to TRH, with basal TSH levels either elevated or normal.[70,72,72a,72b,74]

The experimental model of protein malnutrition in the rat yielded different results from those observed in humans. Serum T4 and T3 levels were found to be both elevated.[75] However, in the lamb, as in humans, chronic malnutrition leads to a lower rate of T4 utilization.[76]

Overfeeding produces an increase in the serum T3 concentration as a result of an increased conversion of T4 to T3. It is particularly marked when the excess calories are given in the form of carbohydrates.[77] Thus, it appears that the effect of overnutrition on iodothyronine metabolism is the opposite of that of starvation. This finding gives further credence to the speculation that changes in thyroid hormone may serve to modulate the homeostasis of energy expenditure.

Although it has been reported that serum T3 concentrations correlate with body weight,[78] it appears that this phenomenon reflects the effect of an increase in caloric intake on T3 production. Most studies find that obese subjects have normal thyroid function and hormone metabolism.[79] Furthermore, no abnormalities in the hypothalamic-pituitary-thyroid axis have been demonstrated in obese subjects.

Iodine. Of the many minerals that may affect thyroid function, iodine is the most important. It is an essential substrate for thyroid hormone synthesis and also interacts with the function of the thyroid gland at several levels.

Acute administration of increasing doses of iodide enhances total hormone synthesis until a critical level of intrathyroidal iodide is reached. Beyond this level, iodide organification and hormone synthesis are blocked (the acute Wolff-Chaikoff block). Chronic or repeated administration of moderate to large doses of iodine causes a decrease in iodide transport resulting in a decrease in its intrathyroidal concentration. The latter relieves the Wolff-Chaikoff block and is known as the escape or adaptation phenomenon. Although the exact mechanisms of the block and escape remain unknown, they appear to be autoregulatory in nature since they are independent of pituitary TSH secretion. Iodoloactones may play a role in the induction of the Wolff-Chaikoff block.[80] One mechanism through which iodide acts is via desensitization of the thyroid gland to TSH. In TSH stimulated glands, iodine rapidly reduces the level of the mRNA for thyroid peroxidase (TPO) and the Na/I symporter (NIS) but not for thyroglobulin (Tg) or the TSH receptor (TSHr).[80a] Iodine also antagonizes TSH stimulated thyrocyte proliferation.[80a] In FRTL-5 cells, iodine blocks the TSH stimulation of Tg synthesis but does not alter the level of the Tg mRNA.[80b] These actions occur without a change in TSH receptor number, and may, in part, be via an action on adenylyl cyclase.[80c] More detailed description is provided in Chapter 2.

Another effect of large doses of iodine, apparently independent of TSH and hormone synthesis, is the prompt inhibition of hormone release. It has been exploited to achieve rapid amelioration of thyrotoxicosis in Graves' disease and toxic nodular goiters (see Chapters 11 and 13). In normal persons, the inhibitory effect of large doses of iodine on thyroid hormone release produces a transient decrease in the serum concentration of T4 and T3. It causes, in turn, a compensatory increase in serum TSH, which stimulates hormone secretion and thus counteracts the effect of iodine.[81,82] The mechanisms of thyroidal autoregulation are believed to serve the purpose of accommodating wide and rapid fluctuations in iodine supply.

The most intriguing effects of iodine are the involution of hyperplasia and the decrease in vascularity that occur when the ion is administered to patients with diffuse toxic goiter. Iodine may be able to induce apoptosis in thyroid cells. [82a,82b] Under different circumstances, iodide may intensify the hyperplasia and produce a goiter (Chapter 20).

Iodine deficiency used to be the leading cause of goiter in the world and still remains so in certain regions. When severe, it can cause hypothyroidism and cretinism, described in detail in Chapter 20 . In the United States and the rest of the developed world, untoward effects from excess iodine supplementation or the use of iodine-containing compounds are more common than problems related to iodine deficiency.

Excess iodine can be responsible for the development of goiter, hypothyroidism, and thyrotoxicosis. However, it should be emphasized that these complications usually occur in persons with underlying defects of thyroid function who are unable to utilize the normal adaptive mechanisms. Iodide-induced goiter (iodide goiter), without or with hypothyroidism (iodide myxedema), is encountered with greater frequency in patients with Hashimoto's thyroiditis or previously treated Graves' disease.[83,84] Other predisposed persons include those who have undergone partial thyroid gland resection, patients with defects of hormonogenesis, and some with cystic fibrosis.[85] Drugs such as phenazone,[86,87] lithium,[88] sulfadiazine,[89] and cycloheximide[90] may act synergistically with iodide to induce goiter and/or hypothyroidism.

More rarely, ingestion of excess iodide may cause thyrotoxicosis (iodide-induced thyrotoxicosis or Jodbasedow).[90a] This was initially observed with the introduction of iodine prophylaxis in areas of endemic iodine deficiency.[91,92] It has also been observed after the administration of iodide in excess to patients with nodular thyroid disease residing in areas of moderate iodine deficiency or even iodine sufficiency.[93,94] Although the exact mechanism of induction of thyrotoxicosis remains obscure, it may be related to the stimulation of increased thyroid hormone synthesis in areas of the gland with autonomous nodular activity.

Ingestion of excess iodide by a gravid woman may cause an iodide goiter in the fetus, and if the gland is large enough it may result in asphyxia during the postnatal period (Chapter 20). Consumption of Kombu, the iodine-rich seaweed, is responsible for the occurrence of endemic goiter in the Japanese island of Hokkaido.[95] It has also been suggested that the increase in dietary iodine content in the United States during the last three decades is responsible for the higher recurrence rate of thyrotoxicosis in patients previously treated with antithyroid drugs.[96]

Calcium. Calcium is said to be goitrogenic when in the diet in excess. Administration of 2 g calcium per day was associated with decreased iodide clearance by the thyroid.[97] The action is unknown, but it may in some way make overt a borderline dietary iodine deficiency. Calcium also acutely and chronically reduces the absorption of thyroxine It has been recently shown that calcium reduces the absorption of thyroxine.[97a, 97b]

Nitrate. Nitrate in the diet (0.3 - 0.9%) can interfere with ¹³¹I uptake in the thyroid of rats and sheep.[98] This concentration is found in some types of hay and in silages.

Bromine. Bromine is concentrated by the thyroid and interferes with the thyroidal ¹³¹I uptake in animals[99, 99a] and humans, possibly by competitive inhibition of iodide transport into the gland. Bromine can also induce alterations in cellular architecture, blood supply and can lead to a reduction in T4 and T3 levels.[99b]

Rubidium. Rubidium is goitrogenic in rats.[100] However, the mechanism of action is unknown.

Florine. Fluorine is not concentrated by the thyroid but has a mild antithyroid effect, possibly by inhibiting the iodide transport process.[101] In large amounts, it is goitrogenic in animals. The amounts of fluorine consumed in areas with endemic fluorosis are not sufficient to interfere with thyroid function or to produce goiter.[102, 103] However, other data suggest that dietary fluorine may exacerbate an iodine deficiency and thus modulate the distribution of goiter in areas with low iodine intake.[104]

Cobalt. Cobalt inhibits iodide binding by the thyroid.[105] The mechanism is unknown. Cobalt deficiency is associated with a reduction in type I monodeiodinase activity and a fall in T3[105a] while cobalt excess may produce goiter and decreased thyroid hormone production. [105b] It is sufficiently active to have been used in the treatment of thyrotoxicosis.[106]

Cadmium. Administration of cadmium to rats or mice decreases serum levels of T4 and T3. [106a],[106b] It also decrease the activity of hepatic Type I - 5’Deiodinase.[106a, 106c]

Lithium Ion. Lithium ion is goitrogenic when used in the treatment of manic-depressive psychosis and can induce myxedema.[107] Experimentally, lithium increases thyroid weight and slows thyroid iodine release.[108] When lithium carbonate was given to human subjects in doses of 900 mg four times daily, there was a significant decrease in the rate of release of thyroidal iodine in euthyroid and hyperthyroid subjects.[109] Lithium also decreases the rate of degradation of T4 in both hyperthyroid and euthyroid subjects.[110] Inhibition of thyroid hormone release may be the dominant effect of the ion.[110a] Therefore, the decrease in serum T3 concentration is greater in hyperthyroid patients, and changes in the rT3 level, if any, are minimal.[111-113]

A number of mechanisms have been suggested for the effects of lithium. One well-documented phenomenon is a potentiation of an iodide-induced block of binding and hormone release,[88,114] perhaps because lithium is concentrated by the thyroid[115] and increases the intrathyroidal iodide concentration[109, 111] (Fig. 5-2). Although it has been shown that lithium inhibits the adenylate cyclase activity in the thyroid gland as well as in other tissues,[116] it also blocks the cAMP-mediated translocation of thyroid hormone. The latter effect, which is probably responsible for the inhibition of hormone release, appears to be due to the stabilization of thyroid microtubules promoted by lithium.[117] In rat brain, lithium administration decreased both the levels of the Type II 5’Deiodinase and the Type III 5 Deiodinase.[117a] In the rat, lithium may also lead to an alteration in the distribution of thyroid hormone receptors with the alpha 1 isoform being increased in the cortex and decreased in the hypothalamus while the beta isoform was also decreaseed in the hypothalamus. [117b]

An exaggerated response of TSH to TRH may be seen in a majority of lithium treated patients[110a] but an elevated basal TSH is usually absent. An increase in the basal serum TSH concentration and its response to TRH most likely represents an early manifestation of hypothyroidism rather than a direct effect of lithium on the hypothalamic-pituitary axis.[118] The prevalence of goiter has been reported to be as high as 60%.[110a] Based on studies in FRTL-5 cells, lithium may have direct mitogenic effects on the thyroid that are independent of TSH and cAMP. [110b] The occurrence of hypothyroidism during lithium therapy occurs in 10-40% of lithium treated patients and is far more frequent in women than men.[110a, 118a], [118b, 118c]

Although much less frequent, lithium therapy has been associated with the development of thyrotoxicosis.[110a] Lithium is also reported to produce exophthalamos during chronic therapy; the condition regresses when treatment is stopped. The phenomenon is a protrusion of the globe but does not involve the other changes of infiltrative ophthalmopathy of Graves' disease.[118, 119]

Selenium. Selenium is a component of the enzymes glutathione peroxidase (GSH-Px) and superoxide dismutase, both enzymes responsible for protection against free radicals. In addition, Type I 5’Deiodinase also contains selenium.[119a] Thus, a deficiency of selenium could predispose the thyroid to oxidative injury and lead to decreased peripheral T3 production. In the elderly, reduced selenium levels have been associated with a decreased T3/T4 ratio.[119b] It has been postulated that the combined deficiency of iodine and selenium in Zaire results in myxedematous rather than neurologic cretinism because the decrease in peripheral conversion to T3 results in greater delivery of T4 into the neonatal developing brain.[119c] In rats, selenium deficiency led to a decrease in renal but not hepatic Type I 5’ Deiodinase activity and serum T3 levels were unaffected.[119d] Selenium deficiency led to decrease GSH-Px activity in the liver, kidney and rbc’s but not the thyroid.[119d] Serum T4 was normal when both dietary iodine and selenium were both deficient, but was reduced when either was deficient alone.[119d] In other studies, brain GSH-Px and Type I deiodinase activity were normal in the presence of iodine or selenium deficiency while brain Type II Deiodinase activity was increased by iodine deficiency and unaffected by selenium deficiency.[119e] In contrast in brown adipose tissue (BAT), both selenium and iodine deficiency led to decreased deiodinase activity and decreased production of the uncoupling protein.[119e]

Treatment of goitrous children with combined seleium and iodine deficiency leads to a reduction in serum TSH and goiter size.[119f] The response, however, was correlated with the selenium level with both the goiter and TSH responses being correlated with the baseline selenium level. [119f]

Surgery has been used as a means to study the effect of stress on thyroid physiology in animals.[122] Studies in humans have been prompted by the suspicion that thyroid hormone may mediate the postoperative metabolic changes leading to increased oxygen consumption and protein wastage. Some discrepancies in available data stem from lack of uniformity in the groups of patients studied in terms of preoperative state or disease, type of surgery, types of anesthetic agents and other drugs used, and the postoperative course, including nutrition and the period of recovery.

The most striking change in thyroid function is a decrease in the serum TT3 and FT3 concentrations shortly after surgery; rT3 concentrations are elevated in the postoperative period.[123, 124] The combined findings suggest a diversion in the normal deiodinative pathways of T4. FT4 levels may also be depressed in the postoperative period, but to a lesser degree.[124] The TTR but not the TBG level is sharply reduced.[125] This clear reduction in the concentration of the active forms of thyroid hormone during the postoperative period is preceded by a small, short-term increase in FT4 and FT3 concentrations during surgery.[123, 124] The magnitude of the subsequent reduction in T3 level appears to correlate with the severity of trauma and the morbidity during the postoperative course.[123] The serum TSH concentration also tends to diminish,[124] except during surgery performed in children under the conditions of hypothermia.[2]

Because surgical trauma produces a prompt elevation in plasma cortisol levels and food intake is curtailed during the pre-, intra-, and postoperative periods, the possibility that glucocorticoids and starvation are the principal contributors to the observed changes in thyroid function has been given strong consideration. However, Brandt et al.[126] showed equally profound diminution in the serum T3 concentration when surgery was carried out with epidural anesthesia, which abolishes the plasma cortisol surge. Similarly, the almost routine use of glucose infusion should have been able to prevent the changes in serum T3 and rT3 levels if starvation played a major role in producing the changes observed during surgery.

Data on the effect of emotional stress on thyroid function in humans are principally derived from studies in patients with psychiatric disturbances. Thus, even if only patients with acute psychiatric decompensation are considered, the results are colored by the nature of the mental illness, its antecedent history, and the use of drugs. An early suggestion of enhanced hormonal secretion came from the observation of elevated protein-bound iodine (PBI) levels in the serum of psychiatric patients presumably under emotional stress and in medical students in the course of examinations.[127] In more recent studies, elevations of the FT4I have been consistently found during admission of acute psychiatric patients. The incidence ranged from 7 to 18%.[128-130] In one study, an equal number of patients (9%) had a low FT4I.[128] In most instances, values became normal with time and treatment of the psychiatric illness. The TSH response to TRH is blunted or even absent in most psychiatric patients with elevated FT4I.[130] Significant abnormalities in the serum T3 concentration are rare.

The discovery of natural and synthetic substances that impair the synthesis of thyroid hormone are landmarks in the history of pharmacology.[131] These substances are discussed in more detail in Chapter 20. Although iodide deficiency is, without doubt, the major cause of endemic goiter and cretinism throughout the world, dietary goitrogens may play a contributing role in some endemics, and may possibly be the dominant factor in certain areas. The dietary goitrogens fall into several categories, more than one of which may occur in the same food.

Certain foods contain cyanogenic glucosides,[132] compounds that, upon hydrolysis by glucosidase, release free cyanide. These foods include almond seeds and such important dietary items as cassava, sorghum, maize, and millet. Cassava contains enough cyanogenic glucoside to be lethal if large quantities are consumed raw. Ordinarily, the root is extensively soaked, then dried and powdered. Most of the cyanide is lost in this process; that left in the root is liberated after ingestion and converted to SCN-. Chronic poisoning due to cassava is responsible for a tropical neuropathy in Nigeria[133] and Tanzania, and is suspected of being a contributing cause of goiter in Central Africa.[134, 135]

Other important classes of antithyroid compounds arise from hydrolysis of the thioglucosides.[132, 136, 137] These compounds are metabolized in the body to goitrin or thiocyanates and isothiocyanates, and ultimately to other sulfur containing compounds, or are excreted as such. They are important in the goitrogenic activity of seeds of plants of the genus Brassica and the cruciferae, compositae, and unbelliferae. Among the plants containing these compounds are cabbage, kale, brussel sprouts, cauliflower, kohlrabi, turnip, rutabaga, mustard, and horseradish. Cattle may ingest these goitrogens and pass them to humans through milk, as observed in Australia,[138] Finland,[139, 140] and England.[141] . The isothiocynate, cheiroline, occurs in the leaves of choumoellier and may be related to a focal area of endemic goiter in Australia. The goitrogen is thought to be transmitted from forage to cows, to milk, and finally to children. Although there is considerable circumstantial evidence relating these compounds to endemic goiter, it has been difficult to prove their role with certainty.

Thiocyanate is a well-known inhibitor of iodide trapping when in high concentration in blood. The blood levels obtained by ingestion of dietary goitrogens are rarely of this degree. Inhibition of iodide trapping, and thyroid peroxidase activity, and augmentation of urinary iodide loss, as demonstrated by Delange and Ermans and co-workers, all my play a role in the goitrogenic activity.[132, 134, 135] Thiocyanate may also reduce the iodine content of breast milk or animal milk and thus indirectly impact the thyroid function of young children in areas of marginal iodine sufficiency. [141a]

Astwood et al. and Greer[142, 143] found that turnips contain progoitrin, which is a mustard oil thioglycoside. It undergoes rearrangement by enzymes in human enteric bacteria, or in the turnip, to be converted to goitrin, an active goitrogenic thioglycoside, L-5-vinyl-2-thio-oxazolidone.[144, 145] Goitrin inhibits oxidation of iodine and its binding to thyroid protein in the same way as do the thiocarbamides.

Several endemics of goiter have been attributed to dietary goitrogens, usually acting together with iodine deficiency. Goitrin is apparently present in cow's milk in Finland.[146] In the Pedgregoso region of Chile, pine nuts of the tree Araucaria americana are made into a flour and consumed in large amounts, and may be related to endemic goiter.[147, 148] In the Cauca river valley of Colombia, sulfur-containing compounds found in the water supply, derived from sedimentary rocks containing a large amount of organic matter, are believed to be responsible for endemic goiter.[149] At least, extracts from these waters are goitrogenic in rats. Pearl millet has been reportd to cause goiter development in goats. [149a]

Other mechanisms may also contribute to dietary goitrogenicity. Thus, diets high in soybean components or other materials increasing fecal bulk may cause excess fecal loss of T4 and increase the need for this hormone.[150-153] These diets are low in iodine content, and soybean has been thought but not proven to contain a goitrogen.

The goitrogens, by blocking hormone synthesis, deplete the thyroid of iodide; this reduction itself increases the sensitivity of the gland to TSH.[154] This sensitivity, in turn, further promotes goitrogenicity.

According to their principal mode of action on thyroidal iodine metabolism, antithyroid drugs are divided into two categories: (1) the monovalent anions, which inhibit iodide transport into the thyroid gland, and (2) a large number of compounds that act through inhibition of thyroidal iodide binding and iodotyrosine coupling. The most important representatives of this latter category of compounds are the group of thionamides. The effect of the drugs in the first category is counteracted by exposure to excess iodine, whereas iodine has no, and at times even potentiates, the action of drugs in the second category. Other drugs inhibit thyroid hormone secretion or act through yet unknown mechanism. A list of these agents is provided in Table 5-1.

Monovalent Anions. Certain monovalent anions (SCN-, Cl04-, NO3-) inhibit transport of iodide into the thyroid gland and thereby depress iodide uptake and hormone formation.[164-166] Thiocyanate stimulates efflux of iodide from the thyroid as well,[167] and also inhibits iodide binding and probably coupling.[168, 169] A large number of complex anions, such as monofluorosulfonate, difluorophosphate, and fluoroborate,[170] inhibit iodide transport. Of these, fluoroborate,[171] like perchlorate,[172] is concentrated by the thyroid gland. These ions have a molecular volume and charge similar to those of iodide, and may compete with iodide for transport.[170, 171] Perchlorate is sufficiently active to be useful clinically.[173] Perchlorate and thiocyanate also displace T4 from thyroid hormone-binding serum proteins in vivo and in vitro and cause a transient elevation of free T4.[174] In contrast to the pharmacologic effects of perchlorate, concerns have been raised about the potential health effects of environmental perchlorate exposure, especially in municipal water supplies. Several studies have been unable to detect an increase in hypothyroidism [174a], congenital hypothyroidism [174b], or thyroid cancer [174c] in exposed populations.

Thionamides. The thionamide and thiourylene drugs do not prevent transport of iodide into the thyroid gland, but rather impair covalent binding of iodide to TG.[175-177] They may be competitive substrates for thyroid iodide peroxidase, preventing the peroxidation of iodide by this enzyme. In small doses, the thiocarbamides inhibit formation of iodothyronines from iodotyrosyl precursors. When slightly larger amounts are present, iodination of MIT and tyrosine is prevented.[177, 178] Minute amounts (10^-8 M) have, paradoxically, a stimulatory effect on iodination in thyroid slices.[179]

The basic structure necessary for the antithyroid action of these drugs is

N

|

S

|

-N=C-X-

where X may be C, N, or O[180, 181] (Fig. 5-3). The thiocarbamides are metabolized in the thyroid gland by transsulfuration.[182] The enzyme responsible may also be involved in the iodide peroxidase enzyme system.[183] Glands under TSH stimulation metabolize the antithyroid drugs at an accelerated rate, as has been shown for thiourea.[184]

Iodide is released more rapidly from a gland blocked by PTU than from one blocked by perchlorate.[165, 185] This action occurs presumably because PTU prevents the utilization of all iodide available to the gland (transported from the blood or formed in the gland by deiodination of iodotyrosines), whereas potassium perchlorate prevents uptake of iodide but does not inhibit reutilization of iodide derived from within the gland. T4 disappears from the PTU-blocked rat thyroid at a faster rate than do iodotyrosines.[185]

In addition to the effects on the thyroid gland, PTU (and, to a much lesser extent, methimazole) partially inhibits the peripheral deiodination of T4[186-191] and its hormonal action.[188,192-194] PTU acts directly on body tissues to inhibit the normal formation of T3 from T4.[191, 195] Coincidentally, fecal excretion of T4 increases.[186] In order to inhibit goiter induced by antithyroid drugs in rats, one must maintain the T4 concentration in blood at a higher level that is normal for the species.[188, 192] Presumably, inhibition of T4 monodeiodination by the antithyroid drug leads to a buildup of T4 in blood and diminishes the availability of T3 in the tissues.[191] Higher doses of T4 or higher blood levels may be sufficient to push the reaction toward T3 and allow formation of quantities sufficient to prevent goiter.

Metabolism of the antithyroid drugs has been observed after administration of ³⁵S-labeled drugs. Methimazole is rapidly absorbed from the gastrointestinal tract in humans. It reaches a peak plasma level about an hour after administration, and then declines gradually to near zero levels at 24 hours. These drugs are accumulated and degraded in the thyroid, since they are substrates of the peroxidase.[196, 197] Carbimazole is accumulated as its metabolic product, methimazole. The concentration ratio between thyroid and plasma for unmetabolized methimazole in rats may approach 25, eight hours after administration of the drug. The metabolic products derived from the drug are excreted in the urine, largely during the first day.

A number of other drugs, including the aminoheterocyclic compounds and substituted phenols, act as goitrogens principally by impairing TG iodination (Fig. 5-3). They are in general far less potent in their goitrogenic effect than the thionamides. None are used therapeutically as antithyroid drugs; rather, goitrogenesis is an undesirable side effect of their use. Some the compounds have multiple effects and thus influence thyroid physiology at various levels. These compounds are individually discussed in greater detail. A comprehensive list is provided in Table 5-1.

Sulfonamides. Sulfonamides, particularly those containing an aminobenzene grouping, have antithyroid activity. Acetazoleamide (Diamox), the diuretic agent, has a strong effect on animals and humans.[198, 199] Its action, prevention of intrathyroidal iodide binding, is not related to carbonic anhydrase inhibition. Sulfadiazine and sulfisoxazole have a similar action, probably through a synergistic effect on iodide.[89]

Sulfonylureas. Sulfonylureas, derivatives of sulfonamides and used as hypoglycemic-antidiabetic agents, also inhibit the synthesis of thyroid hormone. They include carbutamide, tolbutamide, methahexamide, and possibly chlorpropamide, but not the phenylethyl biguanide (Fig. 5-3). They impair thyroidal RAIU and cause goiter in the rat.[200, 201] Carbutamide is much more potent than tolbutamide. Carbutamide, 2 g/day (but not 1 g/day), may reduce the thyroidal RAIU in humans to 20% of control values, but the uptake gradually rises as treatment is continued and is normal after 20 weeks. From 1 to 2 g tolbutamide per day does not affect RAIU in humans.[202] Thus, in the usual dose range, tolbutamide will not depress thyroid function.

Chlorpropamide in large doses (3-7 g) depresses the RAIU in humans; the common therapeutic doses (up to 1 g daily) usually have no effect on serum T4.[203] A mild antithyroid action is often reflected in a rise in RAIU, which may be found after the agents are withdrawn.

These drugs inhibit hormone synthesis by inhibition of iodide binding. In most instances, the pituitary compensates for the effect and maintains a euthyroid state by increased synthesis of TSH. Nevertheless, hypothyroidism is said to be more common in diabetic patients on sulfonylureas than in patients treated by other means.[204]

Sulfonylureas also block binding of T4 to the carrier proteins in serum and thus depress the T4 concentrations.[205] This effect is most pronounced after intravenous administration.

Polychlorinated Biphenyls Animal studies have suggested that polchlorinated bihenyls (PCBs) may reduce thyroid hormone levels by decreasing synthesis, increasing biliary excretion of conjugated metabolites and displacing T4 from binding proteins. [205a] A review of studies in humans, did not find significant or consistent changes. [205a]

A large number of substances may affect thyroid gland function and thyroid hormone metabolism and action. The list continues to grow with the introduction of new diagnostic agents, drugs, and food additives. Drugs affect the transport, metabolism, action and excretion of T4 and its derivatives as well as regulation at all levels of the hypothalamic-pituitary-thyroid axis. Some drugs may induce hypothyroidism or thyrotoxicosis, and if autoimmune mechanisms are involved, the thyroid dysfunction may not resolve with discontinuance of the drug. Some compounds may not have any direct effect on thyroid hormone economy or regulation, but have clinical relevance by interfering in specific diagnostic assays.

Compounds are discussed and listed below based on their major mechanisms of action. Many drugs have more than one mechanism of action and the explanation for observed abnormalities is not always known. Results of experiments conducted in animals or in vitro are not always applicable to human pathophysiology. Compounds which alter thyroid hormone secretion are generally goitrogens or anti-thyroid drugs and were discussed in the preceeding section. Selected compounds with significant effects on the thyroid, wide-spread use or that are of particular interest in understanding the mechanism of drug effects are described in greater detail.

Some hormones and drugs may affect thyroid hormone transport in blood by altering the concentration of the binding proteins in serum. Thyroid hormone transport may also be affected by substances that compete with the binding of thyroid hormone to its carrier proteins (Table 5-2). TBG synthesis is increased by estrogens[220-223] and decreased by androgens and anabolic steroids.[223, 224] Estrogen’s effect to increase TBG is blunted or reversed by tamoxifen and raloxifene.[224a] The most extensively studied compounds that interfere competitively with thyroid hormone binding to the carrier proteins in serum are salicylates, diphenylhydantoin, and heparin.[212,225-231,231a,b] A clinically significant effect of furosemide[211] may only be seen with very high doses and with accumulation with renal failure.

In general, the effect of increased hormone binding is an increase in the serum concentration of total (bound) T4 and of reduced binding is a decrease in the total (bound) T4, with T3 effected to a lesser extent. There is no significant effect on the absolute concentration of the metabolically active fractions of FT4 and FT3, or usually their free indices (FT4I and FT3I). In the steady state, the quantity of thyroid hormone reaching peripheral tissues and the pathways and amount of hormone degradation remain unaltered. However, before this steady state is reached, an acute perturbation in the equilibrium between free and bound hormone brings about transient changes in thyroid hormone secretion and degradation. The hypothalamic-pituitary-thyroid axis participates in the reestablishment of the new steady state. For example, as illustrated in Figure 5-4, an abrupt increase in the concentration of TBG shifts the equilibrium between total and bound hormone, causing a decrease in the concentration of free hormone. The consequences are fourfold. First, there is a shift in the exchangeable hormone from tissues to blood. Second, a decreased hormone content in tissues diminishes its absolute degradation rate. Third, a decline in hormone concentration in tissues activates the hypothalamic-pituitary axis, causing an increase in TSH secretion. Fourth, the latter acts on the thyroid gland to step up its hormonal secretion and reestablish an appropriate thyroid hormone/TBG ratio. Thus, a normal thyroid hormone concentration in serum and tissues and hormonal production and disposal rates are reestablished. TSH concentration returns to normal, and a new steady state is maintained at the expense of an increased intravascular pool and a decreased fractional turnover rate and total distribution space of thyroid hormone.[232, 233] The reverse sequence of events accompanies an acute decrease in TBG concentration or binding (Fig. 5-4).

Agents that may alter the extrathyroidal metabolism of thyroid hormone are listed in Table 5-3. Several drugs with wide use in clinical practice inhibit the conversion of T4 to T3 in peripheral tissues. Glucocorticoids,[239, 240] amiodarone,[241, 242] and propranolol[243-245] are a few examples. As expected, their most profound effect on thyroid function is a decrease in the serum concentration of T3,[239, 241, 243] usually with a concomitant increase in the rT3 level.[239, 241] An increase in the serum T4 concentration has also been observed on occasion.[241, 245] The serum TSH concentration may also occasionally rise,[241] provided the drug does not have a direct inhibitory effect on the hypothalamic-pituitary axis.[246] In the absence of inherent abnormalities in thyroid hormone secretion or in its regulation, TSH levels should return to normal and hypothyroidism should not ensue from the chronic administration of compounds the only effect of which is to interfere partially with T4 monodeiodination.

Other mechanisms by which some compounds affect the extrathyroidal metabolism of thyroid hormone are acceleration of the overall rates of deiodinative and nondeiodinative routes of hormone disposal. Examples of drugs acting principally through the former mechanism are diphenylhydantoin and phenobarbital,[247-249] and via the latter, colestipol[237], ferrous sulfate[238a], aluminum hydroxide[238b] and sucralfate[238c]. Patients receiving these drugs should increase the secretion of hormone from the thyroid gland in order to compensate for the enhanced hormonal loss through degradation or fecal excretion. Thyroid hormone concentration in blood should remain unaltered. However, hypothyroid patients receiving such drugs may require higher doses of exogenous hormone to maintain a eumetabolic state (Chapter 9). In patients on thyroid hormone therapy who are also taking drugs which bind thyroid hormone in the gastrointestinal tract, the administration of the two drugs at different times will markedly reduce or eliminate the effect on thyroid hormone absorption.

Acute increases in serum T4 and FT4 concentration after the injection of insulin or during halothane anesthesia have been attributed to an enhanced release of T4 normally stored in the liver.[250, 251]

The last two decades have seen a prodigious growth in the list of substances that can be shown to act on the hypothalamic-pituitary axis (Table 5-4). Although many of these compounds are used frequently, only a few have significant effects on thyroid function via this central mechanism. Furthermore, patients receiving these drugs rarely have any abnormality of serum TSH although the response of TSH to the administration of TRH may be altered. An effect of these drugs may be seen in patients with untreated or partially treated primary hypothyroidism. In patients with an elevated basal level of serum TSH, addition of these drugs may produce a further increase or a significant diminution.

Although the following paragraphs discuss the general mechanisms of action for these compounds, specific mechanisms are not always known. A major problem in interpretation is the variability of experimental designs. These variables include doses, routes of administration, duration and time of treatment, drug combinations, age and sex of subjects, hormonal status at the time of testing, and time of blood sampling. Furthermore, observed responses may be effected by the method of data analysis. For example, results of TSH responses to TRH have been expressed in terms of changes in the absolute value, increments or decrements from the basal level, and percent of the basal value at either the peak and nadir of the response or the integrated area over the duration of the response.

The most potent suppressors of pituitary TSH secretion are thyroid hormone and its analogs. They act on the pituitary gland by blocking TSH secretion through the mechanisms discussed in Chapter 4. Some TSH-inhibiting agents listed in Table 5-4, such as, fenclofenac and salicylates, may act solely by increasing the free thyroid hormone level through interference with its binding to serum proteins. Other agents appear to have a direct inhibitory effect on the pituitary and possibly on the hypothalamus. The most notable is dopamine and its agonists. They have been shown to suppress the basal TSH levels in euthyroid persons[284, 285] and in patients with primary hypothyroidism.[267,284-286] More uniformly, they suppress the TSH response to the administration of TRH.[268,285,287,288] In contrast, most dopamine antagonists increase TSH secretion.[150-155] Increases in the basal TSH and in its response to TRH have been observed in euthyroid persons,[252, 255] as well as in patients with primary hypothyroidism[250-256] who have been given these drugs. A notable exception to this rule, which casts some doubt on the assumed mechanism of action of dopamine antagonists, is neuroleptic dopamine blocker, pimozide, which has been reported to reduce the elevated serum TSH level in patients with primary hypothyroidism.[289]

Iodine and some iodide-containing organic compounds cause a rapid increase in the basal and TRH-stimulated levels of serum TSH. This effect is undoubtedly due to a decrease in the serum thyroid hormone concentration either by inhibition of hormone synthesis and secretion by the thyroid gland[81,82] or by a selective decrease in the concentration of T3.[290] The latter effect is mediated through the inhibition of T3 generation from T4. A more selective, intrapituitary inhibition of T4 to T3 conversion appears to be responsible for the TSH-stimulating effect of the radiographic contrast agent iopanoic acid[58] and amiodarone. Iodine does not stimulate TSH secretion in patients in whom it has produced hyperthyroidism. [94] A decrease in the free thyroid hormone concentration in serum, albeit minimal in magnitude, may also be responsible for the increase in TSH levels observed during treatment with clomifene.[260]

It has been postulated that some agents may act by modifying the effect of TSH on its target tissue. For example, theophylline may potentiate the action of TSH through its inhibitory effect on phosphodiesterase, which may lead to an increase in the intracellular concentration of cAMP.[291] In fact, the presence of the pituitary is required to demonstrate that methylxanthines augment the goitrogenic effect of a low-iodine diet in the rat.[292] One of the postulated effects of diethyl ether anesthesia in the rat is inhibition of the action of TSH on the thyroid gland,[293] although it has also been reported to induce a transient redistribution of T4 between serum and tissues.[294]

A handful of drugs seem to act by blocking some of the peripheral tissue effects of thyroid hormone. Others appear to mimic one or several manifestations of the thyroid hormone effect on tissues. Guanethidine releases catecholamines from tissues.[295] It has a beneficial effect in thyrotoxicosis, including a decrease in BMR, pulse rate, and tremulousness.[296, 297] This agent has little effect on the thyroid gland, but depresses manifestations of thyrotoxicosis that are mediated by sympathetic pathways. The sympatholytic agents phentolamine and dibenzyline have been reported both to depress and to stimulate thyroid function in animals. Their action is not clear, and it is of muinimal clinical significance.[298-300] Among several -adrenergic blocking agents tested, only phentolamine showed an inhibitory effect on the TSH response to TRH.[271]

Theoretically, thyroid hormone effects could be blocked by drugs which interfere with the tissue uptake of thyroid hormone or binding to its receptors. Inhibition of both cellular uptake and nuclear receptor binding has been demonstrated in vitro for amiodarone in hepatocytes and cultured pituitary cells. Inhibition of cellular thyroid hormone uptake has also been reported for calcium channel blockers and benzodiazapines. Furosemide and non-steroidal anti-inflammatory drugs reduce T3 binfding to cytosolic receptors. There is, however, no clear evidence that any of these drugs have a clinically significant effect on thyoid hormone action.

Among the multiple effects the ß-adrenergic blocker, propranolol, has on thyroid hormone economy, it appears to reduce the peripheral tissue responses to thyroid hormone (see also Chapters 3 and 11). Dinitrophenol enhances oxygen consumption by a direct effect on tissues and thus mimics one of the actions of thyroid hormone.[301]

Estrogens and selective estrogen receptor modulators (SERMs). Hyperestrogenism, either endogenous (caused by pregnancy, hydatidiform moles, or estrogen-producing tumors) or exogenous (due to the administration of estrogens), is accompanied by an increase in TBG and a decrease in TTR concentrations in serum.[220-222] Estrogens are the most common cause of TBG elevation, and this effect can be produced even after their topical application. The magnitude of TBG increase is in part dose related and occurs in women as well as in men. While tamoxifen blocks the estrogen induced increase of TBG[224a], tamoxifen alone in post-menopausal women increases TBG and T4 and 3 levels.[301a] . The selective estrogen receptor modulator (SERM) raloxifene, increases TBG, produces small increase in T4 and insignificant changes in free T4. [301b],[301c] In a single case report, raloxifene appeared to also alter thyroid hormone absorption. [301d] Estrogen increases the complexity of oligosaccharide side chains and, as a consequence, the number of sialic acids in the TBG molecule which in turn prolongs its survival in serum.[302] The concentrations of other serum proteins, including several that bind hormones, such as cortisol-binding globulin and sex-hormone binding globulin, are also increased.[303]

The consequences of increased TBG concentration in serum are higher serum levels of T4, T3 and rT3 and, to a lesser extent, other metabolites of T4 deiodination. The fractional turnover rate of T4 is depressed principally due to an increase in the intravascular T4 pool. On the other hand, the FT4 and FT3 concentrations and the absolute amount of hormone degraded each day remain normal.[232, 233] Transient changes in these parameters during the early changes in TBG concentration can be anticipated as described above. Some of the effects of pregnanc