The pituitary-thyroid axis in patients with pituitary disorders

The pituitary-thyroid axis of 12 patients, exposed to transsphenoidal pituitary microsurgery because of nonfunctioning adenomas (6), prolactinomas (3) and craniopharyngioma (1), or to major pituitary injury (1 apoplexy, 1 accidental injury), was controlled more than 6 months following the incidents. The patients did not receive thyroid replacement therapy and were evaluated by measurement of the serum concentration of thyroxine (T4), 3,5,3'-triiodothyronine (T3), 3,3',5'-triiodothyronine (rT3), T3-resin uptake test and thyrotropin (TSH, IRMA method) before and after 200 micrograms thyrotropin releasing hormone (TRH) iv. The examination also included measurement of prolactin (PRL) and cortisol (C) in serum. Apart from 1 patient with pituitary apoplexy all had normal basal TSH levels and 9 showed a significant TSH response to TRH. Compared to 40 normal control subjects the 12 patients had significantly decreased levels of T4, T3 and rT3 (expressed in free indices), while the TSH levels showed no change. Five of the patients, studied before and following surgery, had all decreased and subnormal FT4I (free T4 index) after surgery, but unchanged FT3I and TSH. The levels of FT4I were positively correlated to both those of FT3I and FrT3I, but not to TSH. The TSH and thyroid hormone values showed no relationship to the levels of PRL or C of the patients exposed to surgery. It is concluded that the risk of hypothyroidism in patients exposed to pituitary microsurgery is not appearing from the TSH response to TRH, but from the thyroid hormone levels.     

The thyroid reserve in children with chronic lymphocytic thyroiditis revealed by the thyrotropin-releasing hormone and thyroid-stimulating hormone tests

Serum thyroid hormone and TSH concentrations were measured before and after the administration of TRH (10 micrograms/kg body weight) and bovine TSH (10 IU) in 14 children with chronic lymphocytic thyroiditis. The TRH test showed that the responsiveness of TSH was positively correlated with the basal TSH (P less than 0.001) and inversely with the increase in serum thyroid hormones, for delta T3 (P less than 0.05) and for delta T4 (P less than 0.001). Overall, the patients had significantly lower mean values for basal T4, but not for T3. The TSH test revealed that the delta T3 was positively correlated with delta T4 (P less than 0.05). delta T3 after TSH administration was positively correlated with it after TRH (P less than 0.05). The patients were divided into three groups on the basis of their peak TSH values after TRH administration. In Group 1 (peak value below 40 microU/ml; N = 5); T3 increased significantly after TRH and TSH administrations (P less than 0.05 and P less than 0.025, respectively). In addition, delta T4 was significant after TSH administration. In Group 2 (peak TSH above 40 and less than 100 microU/ml; N = 6); only delta T3 after TRH was significant (P less than 0.05). In Group 3 (peak TSH above 100 microU/ml; N = 3); the response of thyroid hormones was blunted. Thus, the thyroid hormone responses to endogenous TSH coincided with that to exogenous TSH, and the exaggerated TSH response to TRH indicates decreased thyroid reserve.     

Maturation of the pituitary-thyroid axis during the perinatal period.

To clarify the maturation process of the pituitary-thyroid axis during the perinatal period, thyrotropin (TSH) response to thyrotropin releasing hormone (TRH) and serum thyroid hormone levels were examined in 26 healthy infants of 30 to 40 weeks gestation. A TRH stimulation test was performed on 10 to 20 postnatal days. Basal concentrations of serum thyroxine (T4), free thyroxine (free T4) and triiodothyronine (T3) were positively correlated to gestational age and birth weight (p less than 0.001-0.01). Seven infants of 30 to 35 gestational weeks demonstrated an exaggerated TSH response to TRH (49.7 +/- 6.7 microU/ml versus 22.1 +/- 4.8 microU/ml, p less than 0.001), which was gradually reduced with gestational age and normalized after 37 weeks gestation. A similar decrease in TSH responsiveness to TRH was also observed longitudinally in all of 5 high responders repeatedly examined. There was a negative correlation between basal or peak TSH concentrations and postconceptional age in high responders (r = -0.59 p less than 0.05, r = -0.66 p less than 0.01), whereas in the normal responders TSH response, remained at a constant level during 31 to 43 postconceptional weeks. On the other hand, there was no correlation between basal or peak TSH levels and serum thyroid hormones. These results indicate that (1) maturation of the pituitary-thyroid axis is intrinsically controlled by gestational age rather than by serum thyroid hormone levels, (2) hypersecretion of TSH in preterm infants induces a progressive increase in serum thyroid hormones, and (3) although there is individual variation in the maturation process, the feedback regulation of the pituitary-thyroid axis matures by approximately the 37th gestational week.     

Effect of non aromatizable androgens on LHRH and TRH responses in primary testicular failure

We have assessed the gonadotropin, TSH and PRL responses to the non aromatizable androgens, mesterolone and fluoxymestrone, in 27 patients with primary testicular failure. All patients were given a bolus of LHRH (100 micrograms) and TRH (200 micrograms) at zero time. Nine subjects received a further bolus of TRH at 30 mins. The latter were then given mesterolone 150 mg daily for 6 weeks. The remaining subjects received fluoxymesterone 5 mg daily for 4 weeks and 10 mg daily for 2 weeks. On the last day of the androgen administration, the subjects were re-challenged with LHRH and TRH according to the identical protocol. When compared to controls, the patients had normal circulating levels of testosterone, estradiol, PRL and thyroid hormones. However, basal LH, FSH and TSH levels, as well as gonadotropin responses to LHRH and TSH and PRL responses to TRH, were increased. Mesterolone administration produced no changes in steroids, thyroid hormones, gonadotropins nor PRL. There was, however, a reduction in the integrated and incremental TSH secretion after TRH. Fluoxymesterone administration was accompanied by a reduction in thyroid binding globulin (with associated decreases in T3 and increases in T3 resin uptake). The free T4 index was unaltered, which implies that thyroid function was unchanged. In addition, during fluoxymesterone administration, there was a reduction in testosterone, gonadotropins and LH response to LHRH. Basal TSH did not vary, but there was a reduction in the peak and integrated TSH response to TRH. PRL levels were unaltered during fluoxymesterone treatment.(ABSTRACT TRUNCATED AT 250 WORDS)     

Effect of thyroid hormone therapy on plasma insulin-like growth factor I levels in normal subjects, hypothyroid patients and endemic cretins

The effect of thyroid hormone therapy (L-T4 or L-T3) on plasma immunoreactive insulin-like growth factor I (somatomedin C, Sm-C) concentrations was studied in 8 normal controls, 14 primary hypothyroid subjects and in 7 patients with endemic cretinism. In normals basal levels of Sm-C (1.56 +/- 0.77 U/ml) increased to (2.46 +/- 1.0 U/ml; L-T4) and to (2.9 +/- 0.95 U/ml; L-T3). Plasma Sm-C basal levels were significantly lower in primary hypothyroid subjects (0.81 +/- 0.48 U/ml) and increased to 2.54 +/- 1.43 U/ml (L-T4) and to 2.16 +/- 0.83 U/ml (L-T3). A significant and positive correlation (r = 0.56) was found between Sm-C and serum T4 and T3 concentrations. Plasma Sm-C concentrations in endemic cretinism were initially normal in 4 patients, but low in the remaining 3 (mean +/- SD: 1.18 +/- 0.63 U/ml) and did not increase after 12 months (1.34 +/- 0.61 U/ml) or 18 months (1.01 +/- 0.43 U/ml) of L-T4 and L-T3 therapy. Plasma T4 levels and free T4 increased considerably in EC after therapy with a significant decrease in the previously elevated plasma TSH concentrations. The subnormal response of plasma Sm-C during effective thyroid thyroid hormone therapy could be an additional factor involved in growth failure of endemic cretins.     

The role of estrogen in the TSH and prolactin responses to thyrotropin-releasing hormone in postmenopausal as compared to premenopausal women.

The basal and TRH (Thyrotropin-Releasing Hormone) stimulated TSH (Thyrotropin) and PRL (Prolactin) responses (incremental area; IA) to 200 micrograms TRH was studied in 13 pre- and 13 postmenopausal women of 60 years of age. Both groups consisted of healthy women, none had goiter and all were negative for thyroid autoantibodies. The serum levels of TSH, T3, T4 and SHBG (sex hormone-binding globuline) were in the normal range and did not differ significantly between the groups. There were no differences in basal TSH (1.3 +/- 0.5 vs 1.4 +/- 0.5 mIU/l) or PRL (6.4 +/- 2.7 vs 6.6 +/- 2.5 micrograms/l) or for PRL IA (498 +/- 126 vs 584 +/- 165) between pre- and postmenopausal women. However, for TSH IA there was a slight decrease (15%), but not significant, in the postmenopausal group compared to the premenopausal group (1630 +/- 598 vs 2067 +/- 893). In conclusion, a weak but not significant decrease in the TSH response to TRH in postmenopausal women may be explained by the lower endogenous estradiol level.     

Thyroid hormone modulates insulin-like growth factor-I(IGF-I) and IGF-binding protein-3, without mediation by growth hormone, in patients with autoimmune thyroid diseases.

The expression and synthesis of insulin-like growth factor-1 (IGF-I) and IGF-binding protein-3 (IGFBP-3) are regulated by various hormones and nutritional conditions. We evaluated the effects of thyroid hormones on serum levels of IGF-I and IGFBP-3 levels in patients with autoimmune thyroid diseases including 54 patients with Graves' disease and 17 patients with Hashimoto's thyroiditis, and in 32 healthy age-matched control subjects. Patients were subdivided into hyperthyroid, euthyroid and hypothyroid groups that were untreated, or were treated with methylmercaptoimidazole (MMI) or L-thyroxine (L-T4). Serum levels of growth hormone (GH), IGF-I and IGFBP-3 were determined by radioimmunoassay. Serum GH levels did not differ significantly between the hyperthyroid and the age-matched euthyroid patients with Graves' disease. The serum levels of IGF-I and IGFBP-3 showed a significant positive correlation in the patients (R=0.616, P<0.001). The levels of both IGF-I and IFGBP-3 were significantly higher in the hyperthyroid patients with Graves' disease or in those with Hashimoto's thyroiditis induced by excess L-T4 administration than in control subjects. Patients with hypothyroid Graves' disease induced by the excess administration of MMI showed significantly lower IGFBP-3 levels as compared to those in healthy controls (P<0.05). Levels of IGFBP-3, but not IGF-I levels, showed a significant positive correlation with the levels of free T4 and free T3. In Graves' disease, levels of TPOAb, but not of TRAb, showed a significant positive correlation with IGFBP-3. We conclude that in patients with autoimmune thyroid diseases, thyroid hormone modulates the synthesis and/or the secretion of IGF-I and IGFBP-3, and this function is not mediated by GH.     

Thyroid function tests in children with congenital hypothyroidism on L-thyroxine treatment

The plasma levels of thyroxine (T4), triiodothyronine (T3), free T4 (FT4), free T3 (FT3), reverse T3 (rT3) and immunoradiometrically assayed thyrotropin (IRMA TSH) have been measured in 28 L-T4-treated children with congenital hypothyroidism as well as in a control group (group C). The patients were subdivided into 2 groups according to the nonsuppressed (group A) or suppressed (group B) TSH response to TSH-releasing hormone (TRH). Basal IRMA TSH correlated with the TSH increment after TRH and it was significantly lower in group B vs. groups A and C, while no difference was present between groups A and B in regard to T4, FT4 and rT3, all higher than in group C. FT3 levels were similar in the 3 groups. In children, as in adults, basal IRMA TSH seems to be a reliable index in monitoring overtreatment.     

Significant fluctuation of thyroid function in children with chronic lymphocytic thyroiditis

In order to investigate the degree of pituitary reserve of TSH secretion and the fluctuation of thyroid function in children with chronic lymphocytic thyroiditis, TSH response to TRH was examined in 42 patients, and the thyroid function was carefully followed up in two patients retrospectively and in four prospectively. Increased basal TSH levels were revealed in seven patients (16.8%), and an exaggerated response of TSH to TRH loading in 15 (35.8%). We retrospectively observed spontaneous recovery of thyroid function in two cases. In one of them, two episodes of a transient decrease in thyroid function over a period of several years were noted. Prospectively, low normal T4, elevated TSH and normal T3 were detected in two cases at the first visit. Thereafter, TSH levels decreased to the normal range and the exaggerated response of TSH to TRH became normal. In two other cases, typical transient hypothyroidism occurred during the observation period. These fluctuations lasted for only a few months, and concomitant changes in the size of the thyroid gland were observed. No signs or symptoms suggesting viral infection were noted during the study period. Nor were changes in titers of thyroid auto-antibodies detected. These results show that the secretion of TSH is exaggerated and the thyroid function is decreased in adolescents with chronic lymphocytic thyroiditis, but the thyroid function may fluctuate from euthyroid to hypothyroid within a short period. The causes of these changes, especially of the transient hypothyroidism remain to be classified.     

Diurnal variations of medial basal and anterior hypothalamic thyroliberin [TRH] and serum thyrotropin [TSH] concentrations in male rats.

The diurnal variation of TRH concentrations in different parts of hypothalamus was studied in 80 male rats, which were killed in groups of 5 at 3 h intervals. The hypothalamus was dissected into three parts: I) medial basal hypothalamus (MBH), II) anterior hypothalamus, and III) dorsal hypothalamus. Anterior pituitary and serum TSH concentrations were also measured. TRH concentrations were higher in MBH than in the other parts of the hypothalamus: at night 300–450 pg/mg of wet weight of tissue. When the lights were turned on, MBH-TRH levels began to decrease, reaching a nadir of 210 pg/mg at 12 noon. After 15 h, MBH-TRH levels began to increase again. The changes in TRH levels in anterior hypothalamus were usually opposite to those in MBH (r = ?0.6185). Serum TSH levels were about 800 ng/ml during the day and were decreased to about one half of these levels when the lights were turned off. Serum TSH levels were positively correlated with anterior hypothalamic TRH levels (r = 0.6457) and inversely correlated with MBH-TRH levels (r = ?0.7747). Anterior pituitary TSH levels showed small but statistically insignificant variations. In conclusion, there were statistically interrelated diurnal rhythms in anterior hypothalamic and MBH-TRH levels and serum TSH concentrations.     

Prolactin as a marker of dopaminergic activity at different levels of thyroid function in man

To investigate the hypothesis of an altered dopaminergic activity in hypothyroidism, seven patients without thyroid tissue were studied by means of three consecutive tests: an iv bolus of TRH (200 micrograms); a continuous iv infusion (5 mg during 30 min) of metoclopramide (MCP); and a second, post-MCP, iv bolus of TRH (200 micrograms). The study was performed three times: (A) without treatment; (B) on the 15th day while on L-T4 (150 micrograms i.d.); and (C) on the 30th day with the same treatment. Each time was a different situation of thyroid function; on the basis of basal serum TSH (P less than 0.001, A vs B vs C). The response of PRL to the first (non-primed) TRH, expressed as the sum of increments in ng/ml (mean +/- SE), was significantly higher in A (659 +/- 155) than in C (185 +/- 61). Individual PRL responses correlated with circulating T3 (P less than 0.02), but not with T4. A significant increase of PRL occurred after MCP in the three situations, but there were no differences among them. Likewise, the responses to the second (MCP-primed) TRH showed no differences. Although there was an expected high correlation (P less than 0.001) between basal TSH and circulating thyroid hormones, the maximal response of TSH to both non-primed and MCP-primed TRH was in B. After MCP, no measurable increase of TSH could be demonstrated at any of the three levels of thyroid function. These results do not support the hypothesis of an altered dopaminergic activity in hypothyroidism.(ABSTRACT TRUNCATED AT 250 WORDS)     

术前血清促甲状腺激素水平与甲状腺结节相关性的研究

目的:探讨术前血清促甲状腺激素(TSH)水平与甲状腺结节良恶性的关系。方法:回顾性分析了1499例甲状腺结节手术切除患者术前血清TSH、甲状腺B超,手术记录、术后病理诊断报告。根据术后病理报告判定甲状腺结节良恶性,分析术前血清TSH水平在甲状腺良恶性结节中的不同分布。结果:分化型甲状腺癌(DTC)患者术前血清TSH水平明显高于甲状腺良性结节组(2.179±2.017vs1.259±0.884μIU/mL),P<0.001;在DTC患者中,有淋巴结转移较无淋巴结转移、TNM分期III、IV期较I、II期以及肿瘤直径≥1cm较<1cm的患者术前血清TSH明显升高(均P<0.001)。结论:术前血清TSH水平是预测甲状腺结节良恶性的重要指标。     

Unchanged response to stimulation of pituitary hormone release after serial UV irradiation in men

A group of 24 healthy young men were evaluated before and after serial suberythematous ultraviolet (UV) radiation: group I, control (no irradiation); groups II and III, 12 radiations in 4 weeks with two different spectra (both containing UV-B). Before the first and 2 days after the last exposure all the volunteers were given an intravenous injection of thyrotropin releasing hormone (TRH, protirelin 0.2 mg) and luteinizing hormone releasing hormone (LH-RH, gonadorelin 0.1 mg). The serum concentrations of TSH, follicle stimulating hormone, LH and prolactin were measured at 0, 20, 30, 45 and 60 min by radioimmunoassay. Neither basal nor stimulated levels of the pituitary hormones showed significant changes after UV radiation. The results showed that exposure to suberythematous doses of UV did not influence the regulation of pituitary hormones in these healthy individuals. Accepted: 24 October 1996     

A case of Graves' disease with false hyperthyrotropinemia who developed silent thyroiditis.

We encountered a patient who developed silent thyroiditis during the course of Graves' disease. The diagnosis of silent thyroiditis was made on the basis of a low thyroidal 131I uptake, no response to the thyrotropin releasing hormone (TRH) test, and subsequent hypothyroidism despite the presence of high titers of thyrotropin (TSH) receptor antibody (TRAb) and thyroid stimulating antibody (TSAb). The patient, in addition, had a discrepancy between serum TSH and thyroid hormone values. This was due to the presence of interfering substances that react to mouse IgG in the sera since serum TSH levels were decreased in a dose dependent manner by the addition of increasing amounts of mouse IgG to the sera. It should therefore be noted that silent thyroiditis can develop in patients with Graves' disease. Furthermore, clinicians should be aware that two-site immunoassay kits that use mouse monoclonal antibodies are subject to interference by some substances, possibly antibodies which react to mouse IgG.     

Assessment of the clinically significant TSH response to TRH in patients with nodular goitre

Twenty-five patients with nodular goitre who had thyroid hormone levels within normal ranges and an absent thyrotropin (TSH) response to TSH releasing hormone (TRH) as measured by a conventional radioimmunoassay with a lower detection limit of 0.6 mU/l were studied. Based on these data, and the clinical evaluation patients were divided into a hyperthyroid group (n = 12) and a euthyroid group (n = 13). The samples from the TRH test were reanalyzed by an immunoradiometric TSH assay with a detection limit of 0.05 mU/l. Basal serum TSH showed a considerable overlap between the two groups, but values above 0.10 mU/l were always associated with euthyroidism. Using this level of discrimination 76% of the patients were correctly classified. A TSH response to TRH of 0.10 mU/l provided a better discrimination allowing a correct diagnosis in 92% of the patients. It is concluded that serum TSH as measured by a sensitive assay is suitable as a first line test in patients with nodular goitre. However, patients with basal serum TSH levels below 0.10 mU/l need further investigation with a TRH-test. A TSH response to TRH above 0.10 mU/l seems to secure euthyroidism, whereas lower responses almost always are associated with hyperthyroidism.     

Thyreotrophic hypophysial function after surgery for euthyroid goiter or autonomous adenoma

44 euthyroid patients with nodular goiter and 23 patients with autonomous adenomas were treated by hemithyrectomy or subtotal thyrectomy. Thyroid function was followed over 6 weeks post-operation by TRH tests, which were performed before and at the 5th, 14th, 28th and 42nd day after operation. Bilateral subtotal thyrectomized patients with euthyroid goiter showed a continous increase of basal and TRH stimulated TSH level into the hypothyroid range. 19 of 25 patients were hypothyroid 6 weeks after operation. In contrast, 14 of 19 hemithyrectomized patients with euthyroid goiters remained euthyroid during the time investigated; 5 patients showed a transient TSH increase into the hypothyroid range but were euthyroid again after 6 weeks. TSH levels obtained from patients operated for autonomous adenoma may not yet reflect thyroid function during the time interval investigated here. We conclude that all patients with euthyroid goiter after bilateral subtotal thyrectomy should receive hormone substitution because they are at high risk to develop recurrency. However, we propose that in patients hemithyrectomized for euthyroid goiters the decision of long term hormone substitution should be cased on the result of a TRH-test 3--4 month after operation. Substitution with thyroid hormone should be preferred to iodide because it is unclear yet how far a failure in iodide organification and hormone synthesis is the reason for goiter recurrency.     

Anterior pituitary cell antibodies detected in Hashimoto's thyroiditis and Graves' disease

An immunofluorescence study using unfixed cryostat sections of rat pituitary glands was carried out on sera from 34 patients with Hashimoto's thyroiditis, 28 patients with Graves' disease, 10 patients with thyroid adenoma and 50 healthy subjects. After absorption of sera with rat liver tissues, 19 of 34 patients retained reactivity to anterior pituitary cell antibodies (PCA, 55.8%). On the other hand, immunofluorescence in anterior pituitary cells was faint and detected in only 2 of 28 patients with Graves' disease (7.1%) after absorption of their sera with rat liver aceton powder. A similar result was also obtained when PCA were compared in the sera of Hashimoto's thyroiditis and Graves' disease with high titers of thyroid microsomal autoantibodies. PCA were detected neither in the sera of patients with thyroid adenoma nor in the healthy subjects. The present study suggests that PCA were considerably more prevalent in Hashimoto's thyroiditis than in Graves' disease.     

The value of plasma TSH estimation and TRH stimulation test in the course of treatment of differentiated thyroid cancers.

Plasma TSH levels were measured on 114 occasions in 96 patients treated for differentiated thyroid cancer. Prior to thyroid surgery, plasma TSH levels were within the range of normal. Plasma TSH levels increased slightly following partial thyroid resection and definitely after total thyroid ablation. In patients where the removal of normal thyroid induced hormonogenesis in thyroid tumours, plasma TSH levels were dependent on the hormonal secretion of the tumour as shown by inverse relationship between TSH and both PBI and 131I uptake. The increase of radioiodide uptake following stimulation by exogenous bovine TSH was inversely related to the plasma thyrotropin levels. The suppressibility of enhanced thyrotropin levels was complete with individually adjusted doses of synthetic thyroid hormones. With the exception of patients on suppressive treatment, TRH administration induced increase in plasma TSH levels. The findings are discussed with regard to the role played by TSH in the induction of hormonogenesis in thyroid tumours. The practical values of TSH estimation and TRH stimulation seem to be low; the measurements of thyrotropin levels may be important for the estimation of the suppressive effect in the course of and following withdrawal of treatment with thyroid hormones.     

Relationship of iodide-induced increase in TSH response to TRH to changes in serum thyroid hormones.

Large doses of iodide (500 mg three times a day) administered to normal men for 10--12 days caused a rise in basal serum TSH and a concomitant rise in the peak TSH response to TRH. The basal and peak levels of TSH were highly correlated (p less than 0.001). However, the iodide-induced rise in the peak TSH after TRH was poorly correlated with concomitant changes in serum thyroid hormones. Serum T3 wa not lower after iodide and, while serum T4 was somewhat lower, the fall in serum T4 was unexpectedly inversely rather than directly correlated with the rise in the peak TSH response to TRH. Thus, increased TSH secretion after iodide need not always be directly correlated with decreased concentrations of circulating thyroid hormones even when large doses of iodide are used. Clinically, a patient taking iodide may have an increased TSH response in a TRH stimulation test even though there is little or no change in the serum level of T3 or T4.     

Thyroid function and antimicrosomal antibody during the course of silent thyroiditis

The thyroid function and antithyroidal antibody were studied in 17 patients with silent thyroiditis unrelated to pregnancy. The antimicrosomal hemagglutination antibody (MCHA) was negative in ten of them (group I) and was positive in seven (group II). At one month after the thyrotoxicosis, thyroid function became normal in both groups. At two months after the onset of thyrotoxicosis, in group I T4 (8.1 +/- 1.8 micrograms/dl, Mean +/- SD), T3 (113 +/- 25 ng/dl) and TSH were normal. At that time T4 (2.8 +/- 2.2 micrograms/dl) was significantly decreased (p less than 0.001) compared with those of group I and the levels of TSH were strikingly increased in 6 patients in group II. The level of T3 (96 +/- 29 ng/dl) in group II was not different from that of group I. Therefore MCHA was negative in patients who did not develop hypothyroidism and MCHA was positive in patients who developed hypothyroidism. The development of hypothyroidism two months after thyrotoxicosis and positive MCHA are correlated. The Tg was elevated in 7 out of 13 patients (54%) with negative antithyroglobulin hemagglutination antibody and in the remainder was normal during thyrotoxicosis. The discrepancy between the level of Tg and thyroid hormones was discussed.