Serum thyrotropin receptor antibodies concentrations in patients with Graves' disease before, at the end of methimazole treatment, and after drug withdrawal: evidence that the activity of thyrotropin receptor antibody and/or thyroid response modify during the observation period.

AIM AND METHODS: We performed a quantitative retrospective analysis of serum thyrotropin receptor antibody (TRAb) concentrations measured by a second-generation radioreceptor assay in 58 patients with Graves' disease (GD) at the onset of the disease, at the end of 18 month methimazole (MMI) treatment, and after MMI withdrawal in order to evaluate the correlation between the presence of these antibodies and the relapse of hyperthyroidism. Sixty healthy subjects were enrolled as a control group. RESULTS: Before MMI treatment the best cutoff TRAb value for identifying patients with GD was 1.45 UI/L (specificity, 100%; sensitivity, 98.3%). At the end of MMI treatment, serum TRAb concentrations were significantly lower (p < 0.001) than those measured at baseline, but they were still significantly higher (p < 0.001) than those found in the control subjects. At the end of MMI treatment, 44 patients (75.9%) had positive TRAb values (>1.45 UI/L). After MMI withdrawal (median, 15 months), 34 patients (58.6%) became hyperthyroid, 4 patients (6.9%) became hypothyroid, and 20 patients (34.5%) remained euthyroid. There was a significant correlation between serum TRAb concentrations at the end of MMI treatment and the percentage of patients who became hyperthyroid (r: 0.56; p < 0.001) and the time of appearance of hyperthyroidism (r: -0.38; p = 0.03). All 4 patients with TRAb values below 0.9 UI/L at the end of MMI treatment remained euthyroid throughout the follow-up period. Among the 27 patients who had serum TRAb values higher than 4.4 UI/L, 23 developed hyperthyroidism and 4 hypothyroidism. The TRAb values between 0.9 and 4.4 UI/L did not discriminate between the 27 patients (46.6%) who remained euthyroid from those who had relapse of hyperthyroidism. Thus a different TRAb end of treatment cutoff was calculated to identify patients who became again hyperthyroid. This TRAb cutoff value was 3.85 UI/L (sensitivity, 85.3%; specificity, 96.5%). All but 1 patient who had serum TRAb values above 3.85 UI/L became hyperthyroid after MMI was withdrawn (positive predictive value, 96.7%). In these patients, relapse of hyperthyroidism was independent of the changes in serum TRAb concentrations (r: 0.27; p = 0.15) and occurred after a median period of 8 weeks (range, 4-48). Hyperthyroidism also developed in 5 of 24 patients who had serum TRAb concentrations lower than 3.85 UI/L at the end of MMI treatment. In these 5 patients the relapse of hyperthyroidism occurred after a median period of 56 weeks (range, 24-120) and was always accompanied by an increase in serum TRAb concentrations. CONCLUSIONS: TRAb persist in the blood of most patients with GD after 18 months of MMI treatment. Both the frequency and the time of appearance of hyperthyroidism are closely correlated with serum TRAb concentrations at the end of MMI treatment. Our data would suggest that TRAb maintain stimulating activity after a full course of MMI treatment in the large majority of patients with GD. However, it is likely that the potency of these antibodies and/or the thyroid response to them change during treatment, as suggested by the different values measured in euthyroid control subjects and in euthyroid patients after MMI treatment.

Sarcoidosis of the thyroid gland associated with hyperthyroidism: review of the literature and report of two peculiar cases.

Sarcoidosis is a systemic disease characterized by non-caseating granulomas that rarely involve the thyroid gland. Thyroid sarcoidosis has seldom been documented, and few cases have so far been described in association with hyperthyroidism. Here, we review the literature on this association, report two patients presenting with hyperthyroidism and histologically-proven sarcoidosis, and discuss related clinical, biochemical, pathological and genetic findings.

Thyrotoxicosis followed by hypothyroidism in patients treated with amiodarone. A possible consequence of a destructive process in the thyroid.

Amiodarone is an iodine-rich drug used in the treatment of resistant cardiac arrhythmias. Amiodarone-induced thyrotoxicosis (AIT) is well recognized and is generally believed to be due to the excess iodine released from the metabolism of the drug, although amiodarone-associated thyroiditis has occasionally been observed. We report the clinical, laboratory, and therapeutic course of nine patients with AIT. The thyrotoxic phase was often followed by mild hypothyroidism during and after antithyroid drug or corticosteroid therapy. The thyroid was tender to palpation in two patients and a fine-needle aspiration biopsy of the thyroid revealed changes consistent with thyroiditis in three patients who underwent biopsy. These findings strongly suggest that the etiology of AIT, at least in some patients, is due, in part, to drug-induced inflammatory changes in the thyroid, as has been reported to occur in lung, bone marrow, and testes. Thus, the AIT may be due to excess iodine, drug-induced thyroiditis, or a combination of the two. The favorable response of AIT to corticosteroids in occasional patients previously reported also suggests that acute thyroiditis was probably present.

Graves' disease and thyroxine-binding globulin deficiency.

Thyroxine-binding globulin (TBG) deficiency has been frequently described in single patients and in many families. Most people with abnormal TBG concentrations are euthyroid. Cases of Graves' disease and TBG deficit have rarely been reported. We describe the case of a person with Graves' disease and TBG deficiency. Because of this condition, the patient had a misdiagnosis during part of his clinical history, and therefore underwent unnecessary therapy.

The effect of diltiazem, a calcium channel-blocking drug, on cardiac rate and rhythm in hyperthyroid patients.

Tachycardia and tachyarrhythmias are frequent in patients with thyrotoxicosis, especially in the elderly. Since myocardial calcium uptake is increased in thyrotoxic rats, the efficacy of the calcium channel-blocking drug diltiazem in decreasing heart rate and the incidence of arrhythmias was evaluated in 11 hyperthyroid patients. All patients were studied with a 24-hour Holter monitor prior to the beginning of sole diltiazem therapy (120 mg given every eight hours), on the tenth day of therapy, and five days after therapy was discontinued. Heart rate significantly decreased by 17% during diltiazem treatment (96.5 +/- 3.7 systoles/min vs 79.9 +/- 3.2 systoles/min [mean +/- SE]) and returned to baseline values five days after the therapy was discontinued (100.7 +/- 3.4 systoles/min). Similarly, the number of premature ventricular extrasystoles per hour was significantly decreased (18 +/- 7 vs 2 +/- 1). In three patients, asymptomatic bouts of supraventricular tachycardia, paroxysmal atrial fibrillation, or ventricular tachycardia disappeared during diltiazem therapy. These findings suggest that calcium-blocking drugs may be extremely useful as adjunctive therapy for thyrotoxicosis in the presence of angina, congestive failure, and tachyarrhythmias.

Rat placenta is an active site of inner ring deiodination of thyroxine and 3,3',5-triiodothyronine.

Inner ring deiodination of L-T4 and L-T3 by rat placental homogenates resulted in the generation of rT3 from T4 and 3,3'-diiodothyronine and 3'-monoiodothyronine from T3. Dithiothreitol is required in the incubation medium. There was little or no detectable outer ring deiodination of T4 and T3. These findings suggest that placenta enzymatic inner ring monodeiodination of T4 and T3 could prevent the transplacental passage of T4 and T3 from dam to fetus. They also provide an explanation for our previous observations that fetal serum rT3 is partially dependent on maternal thyroid function.

The sex-related difference in serum thyrotropin concentration is androgen mediated.

Studies were carried out on various aspects of hypothalamic-pituitary-thyroid function in normal and gonadectomized adult male and female rats. Consistent increases in the serum TSH concentration and the serum TSH response to TRH were observed in the male rat compared to values in the female. Orchiectomy induced a decrease in the serum TSH concentration and the serum TSH response to TRH, and these functions were equal in gonadectomized male and female rats. Oophorectomy did not affect basal and TRH-stimulated serum TSH concentrations. Replacement doses of testosterone (0.33 mg/day) to orchiectomized rats increased and restored these values to those observed in the normal male rat, while replacement doses of estradiol (0.33 microgram/day) to the oophorectomized rat had no effect on basal or TRH-stimulated TSH concentrations. No sex-related differences in pituitary TSH and hypothalamic TRH contents or in serum T4 and T3 concentrations were observed. The present studies strongly suggest that the increased TSH responsiveness observed in male compared to female rats is due to the presence of testosterone. (Endocrinology 108: 529, 1981)

The effect of thyroid dysfunction and fasting on placenta inner ring deiodinase activity in the rat.

The placenta contains iodothyronine 5-deiodinase activity (P5-Dase) that probably acts on iodothyronines in the fetal circulation to convert T4 to rT3 and T3 to 3,3'-T2. Since thyroid status and fasting have profound effects on iodothyronine deiodinases in other tissues, the present studies were performed to determine if these perturbations affected P5-Dase. Control and treated rats were mated and killed near term on the 20th day of gestation. P5-Dase was determined in placenta homogenates enriched with dithiothreitol by measuring the conversion of T4 to rT3. In four of five studies, P5-Dase was similar in dams that underwent thyroidectomy (Tx) on day 7 of gestation and sham Tx dams. P5-Dase was not altered in dams that were treated with methimazole (MMI) to induce maternal and fetal hypothyroidism. Treatment of dams with supraphysiological doses of T4, beginning on the seventh day of gestation, did not significantly affect P5-Dase. In three of four studies, P5-Dase was similar in fed dams to values in dams fasted for the last 5 days of pregnancy. Placenta iodothyronine 5'-deiodinase activity (P5'-Dase) was also measured in some studies. P5'-Dase was not decreased in Tx rats and was modestly decreased in MMI-treated rats. However, the effect of MMI was not reversed by the administration of supraphysiological doses of T4, Tx, MMI treatment, and fasting all decreased hepatic T4 5'-deiodinase activity in pregnant rats. These results strongly suggest that thyroid status and fasting do not alter P5-Dase activity.

"Short" loop feedback regulation of hypothalamic and brain thyrotropin-releasing hormone content in the rat and dwarf mouse.

In order to determine whether T4, TSH, or both affect hypothalamic TRH content, primary or secondary hypothyroidism was induced in the rat by thyroidectomy (Tx) or hypophysectomy (Hx), respectively. Two weeks later, rats were treated with T4, TSH, or both for 14--16 days. Tx or Hx significantly decreased hypothalamic TRH content, and T4 treatment restored hypothalamic TRH to normal in the Tx but not in the Hx rats. When TSH was administered simultaneously with T4 to Hx rats, hypothalamic TRH content was restored to normal. Whole brain TRH content was not affected by Tx, Hx, or by the various treatment regimens. Similar experiments were performed in Snell dwarf (dw/dw) mice. Hypothalamic TRH content was significantly decreased when compared to normal litter mates (dw/+). As in Hx rats, T4 administration to dw/dw mice for 16 days did not restore hypothalamic TRH content to normal. Brain TRH content was similar in both groups and was not altered by T4 treatment. It is concluded that T4 affects hypothalamic TRH content, but that TSH is required for this T4 effect. Whole brain TRH, on the other hand, is unaffected by any of these alterations in thyroid-pituitary function.

Ontogenesis of placental inner ring thyroxine deiodinase and amniotic fluid 3,3',5'-triiodothyronine concentration in the rat.

Human and rat placentae contain enzymatic activity which converts T4 to rT3 and T3 to 3,3'-diiodothyronine and 3'-monoiodothyronine. This study presents data on the ontogeny of this inner ring iodothyronine deiodinase activity (P-T4ase) in rat placenta. P-T4ase was measured by quantitating the conversion of T4 to rT3 in 700 x g supernatants of placental homogenates. Groups of rats were mated to permit the dams to be killed on the same day, on days 12, 14, 16, 18, and 20 of gestation. Sufficient placental tissue was obtained to measure P-T4ase on all but the 12th day of gestation. The highest level of P-T4ase was observed on day 16. P-T4ase on days 14, 18, and 20 was 52%, 77%, and 41%, respectively, of that observed on day 16 (P less than 0.01, day 16 vs. all other days). Amniotic fluid rT3 concentrations were highest on day 18 and were 61% and 64%, respectively, of that observed on day 18 (P less than 0.01, days 16 and 20 vs, day 18). At 20 days, maternal serum T4 concentrations were significantly lower (P less than 0.01) than on days 14, 16, or 18. A brief period of maternal hypothyroidism (4 or 9 days before the time that the animals were killed on day 20 of gestation) did not significantly alter P-T4ase. These studies indicate that there are age-dependent changes in placental inner ring deiodinase activity in the rat. Amniotic fluid rT3 concentrations may reflect these changes. Brief reductions in maternal serum T4 concentrations do not account for changes in placental inner ring deiodinase activity. These studies emphasize the importance of gestational age in studies of placental inner ring iodothyronine deiodinase.

Iodine-induced hypothyroidism in euthyroid subjects with a previous episode of subacute thyroiditis.

The effects of the administration of pharmacological quantities of iodide on thyroid function in 18 euthyroid patients with a previous history of painful subacute thyroiditis (SAT) were evaluated, and the results compared to those of iodide administration to 12 euthyroid patients with a previous history of thyroid surgery (TX) for benign nodular thyroid disease. After baseline thyroid function tests, saturated solution of potassium iodide (SSKI; 10 drops; 300 mg iodide) was administered daily for 120 days, and serum T4, T3, and TSH concentrations were assessed 15, 30, 60, 90, 120 days and 2-4 months after SSKI was discontinued. Iodide perchlorate discharge tests were carried out before SSKI administration, and TRH tests were performed on the last day of iodide administration. Two SAT subjects developed clinical evidence of hypothyroidism with markedly elevated serum TSH concentrations, and SSKI was discontinued on days 60 and 90, respectively. Thirteen of 18 SAT patients had at least 1 abnormal thyroid function test (iodide perchlorate discharge test, elevated serum TSH concentration, and abnormal TSH response to TRH) compared to 2 of 12 TX patients. These findings strongly suggest that euthyroid subjects with a previous history of SAT are prone to develop iodide-induced hypothyroidism, suggesting that subtle abnormalities in the thyroid organification of iodide and subsequent thyroid hormone synthesis persist years after the episode of SAT.

The effects of propylthiouracil, iodothyronines, and other agents on thyroid hormone metabolism in human placenta.

Human and rat placental homogenates contain inner ring deiodinase activity (PT4ase) towards T4 and T3. This activity may decrease the transfer of T4 and T3 across the placenta and influence thyroid hormone disposal in the fetal circulation. Data are now presented on human PT4ase in subcellular fractions, the Km of human PT4ase, and the effects of drugs and other compounds on human and rat PT4ase. The specific activity (nanograms of rT3 produced per min/mg protein) of each fraction of human placenta was as follows: nuclear, 0.07; mitochondrial, 0.15; lysosomal, 0.19; microsomal, 1.30; and cytosol, 0.01. The apparent Michaelis-Menton (Km) for PT4ase in human placental microsomes was 1.2 X 10(-7) M. T3, 3,3'-diiodothyronine, iopanoic acid, iodoacetic acid, diamide, and propranolol all exhibited dose-dependent inhibition of human and rat PT4ase when tested in the presence of 10 mM dithiothreitol (DTT). Propylthiouracil did not inhibit PT4ase at 10 mM DTT, but when the DTT concentration was lowered to 0.25 mM, up to 71% inhibition was noted. Many drugs, as noted in other organs with respect to outer and inner ring iodothyronine deiodinases, inhibited human PT4ase. These studies may be relevant to the practice of administering propylthiouracil, propranolol, and iopanoic acid to pregnant women.

Sinus node function in hyperthyroid patients.

We have studied the electrophysiology of the sinus node and the role of the autonomic nervous system on sinus node function in 8 thyrotoxic patients of both sexes, 37.5 +/- 4.3 (mean +/- SE) yr old. The resting heart rate (RHR), the sino-atrial conduction time (SACT), and the sinus node recovery time (SNRT) were measured in the untreated condition (basal), after sympathetic blockade with propranolol 0.2 mg/kg body weight (BW) i.v. infusion, and after complete autonomic blockade with the additional administration of atropine 0.04 mg/kg BW i.v. bolus. 1) In the thyrotoxic patients the RHR was higher [117 +/- 6 beats per min (bpm)] than in 20 normal subjects (73 +/- 1 bpm, P less than 0.001), whereas the SACT and SNRT values were not different. 2) After sympathetic blockade with propranolol, the RHR decrement and SACT increase were greater in the hyperthyroid patients than in normal subjects, whereas there was no difference in SNRT values between the two groups. 3) In the thyrotoxic patients the complete autonomic blockade reestablished the electrophysiological parameters to values similar to those observed in basal condition. In conclusion, in thyrotoxic patients the intrinsic activity of the sinus node is increased. It appears that this is a direct consequence of thyroid hormone excess, rather than an effect of extrinsic influences exerted by the autonomic nervous system on sinus node activity.

Response of growth hormone to thyrotropin-releasing hormone during fetal life.

The effect of TRH administration to the term pregnant women on the GH response in cord blood (CB) was evaluated in 138 subjects. Previous studies have demonstrated that TRH readily crosses the placenta. TRH (400 microgram) was administered iv to 59 pregnant women just before delivery. CB samples were obtained at delivery and assigned to 6 groups, depending upon the duration of time between TRH injection and CB sampling. The control group comprised 79 pregnant women who received saline. A progressive rise and then a fall in the CB GH concentration were observed after TRH administration. Values were significantly elevated 61-90 min after TRH administration compared to values in saline-treated subjects (19.3 +/- 3.1 vs. 13.1 +/- 0.9 ng/ml; P less than 0.05). The present study is the first report of the effect of TRH on the GH concentration in CB and suggests that TRH stimulates GH release in the fetus.

Human placenta is an active site of thyroxine and 3,3'5-triiodothyronine tyrosyl ring deiodination.

Human placental homogenate deiodinates the tyrosyl ring of T4 and T3, converting these active thyroid hormones to the inactive iodothyronines, rT3 from T4, and 3,3'-diiodothyronine and 3'-monoiodothyronine from T3. The conversion of T4 or rT3 is time, temperature, pH, and protein content dependent and does not occur in the absence of the thiol-regenerating agent dithiothreitol. Phenolic ring deiodination of T4, T3, and rT3 was not detected. Failure of the transplacental passage of T4 and T3 from mother to fetus may be secondary to the placental tyrosyl ring deiodination of these iodothyronines.

Effects of chronic iodine administration on thyroid status in euthyroid subjects previously treated with antithyroid drugs for Graves' hyperthyroidism.

In view of the adverse effects of the administration of pharmacological quantities of iodine to euthyroid patients with a history of a wide variety of thyroid disorders, it has been suggested that iodine-containing medications and radioopaque dyes containing iodine should be avoided, if possible, in patients with underlying thyroid disease. We have now prospectively studied the effects of pharmacological doses of a saturated solution of potassium iodide (SSKI) on thyroid function in euthyroid patients with a previous history of hyperthyroid Graves' disease successfully treated with antithyroid drugs. Ten euthyroid women (mean age, 56 yr) who had hyperthyroid Graves' disease successfully treated with methimazole 36.4 +/- 4.7 months earlier were evaluated before, during, and after the administration of 10 drops SSKI daily for 90 days. The following thyroid function tests were obtained: serum T4, T3, TSH, TSH receptor antibody (TSH-RAb), and antithyroid peroxidase antibody (AbTPO) concentrations; TRH tests; and iodine perchlorate discharge tests. Serum T4, T3, basal and TRH-stimulated TSH, and TSH-RAb values were normal before SSKI administration, but serum AbTPO levels were markedly positive in 7 and iodine perchlorate discharge tests were positive in 4 of these 10 women. During SSKI administration, basal and TRH-stimulated serum TSH values increased above normal in 2 women with normal serum T4 and T3 concentrations; thyroid hormone values and TRH tests were normal in the other 8 patients and similar to values observed in 4 euthyroid women without a history of thyroid disease given SSKI. Serum AbTPO increased slightly, but significantly, during SSKI administration in the 7 women with positive values at baseline (P < 0.05). TSH-RAb remained undetectable. After SSKI withdrawal, the 10 women were reevaluated 60 and 120 days later. Two women developed a blunted TSH response to TRH, but normal serum T4 and T3 concentrations, and 2 women developed overt hyperthyroidism, with undetectable basal and TRH-stimulated serum TSH and elevated serum T4 and T3 concentrations, requiring methimazole therapy. All values in the remaining 6 women were similar to those present before SSKI administration. These results suggest that some euthyroid patients with a history of antithyroid drug therapy for Graves' disease may develop thyroid dysfunction during and after excess iodine administration. The development of subclinical hypothyroidism during SSKI administration was not clinically important, but the occurrence of overt hyperthyroidism after SSKI was discontinued did require antithyroid drug therapy. It is advisable, therefore, to avoid iodine-containing substances in euthyroid patients with a history of antithyroid drug therapy for Graves' disease.

Goiter size and thyroid function in an endemic goiter area in northern Italy.

Thyroid function was studied in a large number of subjects residing in Varsi, a town in the province of Parma, Italy. In this area, endemic goiter associated with moderate iodine deficiency [59 +/- 3 (+/- SE) microgram iodine/g creatinine], as defined by WHO criteria, affects 65% of the population. Serum T4, T3, thyroglobulin (Tg), and TSH concentrations were measured by RIA in 1218 subjects. The TSH response to TRH was determined in 108 subjects selected randomly from the groups with different grades of goiter. No significant change in serum T4 concentrations was found in subjects with different grades of goiter. Serum T3 concentrations were higher in subjects with the larger goiters. Serum Tg concentration progressively increased, and serum TSH progressively decreased with increasing goiter size. The TSH response to TRH was diminished in subjects with larger goiters. The findings of decreasing serum TSH concentrations and blunted TSH responses to TRH as goiter size increased suggest the possibility of autonomous thyroid function in the larger goiters in subjects residing in this area of moderate iodine deficiency.

Failure of metoclopramide to affect thyrotropin concentration in the term human fetus.

Metoclopramide (MET), a potent dopamine receptor-blocking drug, or saline was administered to 125 term pregnant women at various time intervals (5-412 min) before delivery. Maternal serum was obtained before and after MET injection. Cord blood was obtained at delivery in MET-treated and saline-treated (control group) women. No significant changes in serum TSH, T4, T3, or rT3 concentrations were observed in maternal or cord blood after MET administration. These results suggest that, in contrast to euthyroid nonpregnant women and men, MET administration does not induce a rise in serum TSH concentration in term pregnant women or in the term fetus. Thus, the dopaminergic inhibitory effect on anterior pituitary TSH secretion may not be an important factor in TSH regulation during pregnancy or in the fetus, or the dose of MET employed may be unable to overcome the dopamine inhibitory effect.

Iodine-induced subclinical hypothyroidism in euthyroid subjects with a previous episode of amiodarone-induced thyrotoxicosis.

Amiodarone-induced thyrotoxicosis (AIT) occurs most frequently in patients with underlying thyroid disease and is generally believed to be due to the iodine contamination of amiodarone and iodine released by the metabolism of the drug. We and others have suggested that the thyrotoxicosis may also be secondary to amiodarone-induced thyroiditis. To further determine the etiology of AIT, we administered large doses of iodides [10 drops saturated solution of potassium iodide (SSKI) daily] to 10 euthyroid patients long after an episode of AIT believed to be due at least in part to amiodarone-induced thyroiditis. Six of these 10 patients had an abnormal iodide-perchlorate discharge test before SSKI administration, indicating a subtle defect in the thyroidal organification of iodide. During SSKI administration, 6 patients developed marked iodine-induced basal and/or TRH-stimulated serum TSH elevations, 2 had suppressed basal and TRH-stimulated TSH values, and 2 had normal TSH responses compared to SSKI-treated euthyroid subjects with no history of amiodarone ingestion or thyroid disease. Serum T4 and T3 concentrations remained normal and unchanged during SSKI administration in both the AIT patients and control subjects. These results strongly suggest that excess iodine may not be the cause of the hyperthyroidism associated with amiodarone therapy, especially in those patients with probable amiodarone-induced thyroiditis. Furthermore, like patients with a previous history of subacute thyroiditis and postpartum thyroiditis, the present results suggest that some patients with a previous history of AIT may be at risk to develop hypothyroidism when given excess iodine.

Impaired intrathyroidal iodine organification and iodine-induced hypothyroidism in euthyroid women with a previous episode of postpartum thyroiditis.

Postpartum thyroiditis (PPT) is common and occurs in 1.7 to 16.7% of pregnant women, depending upon the study population. Most of these women develop transient hypothyroidism and thyroid function usually returns to normal. We have studied 11 euthyroid women with a previous history of PPT to determine the incidence of subtle defects in thyroid function measured by iodide-perchlorate (I-ClO4) discharge tests and TRH tests and to determine whether these women would develop iodide-induced hypothyroidism. Seven (64%) had positive I-ClO4 discharge tests and 5 (46%) had an abnormally high TSH response to TRH. Thyroid antimicrosomal and antithyroid peroxidase were positive in 8 women (73%) with a previous episode of PPT. The administration of pharmacological amounts of iodide (10 drops of saturated solution of potassium iodide daily) for 90 days to these 11 women resulted in elevated basal and TRH stimulated serum TSH concentrations in 8 (72.7%) compared to TSH values during iodide administration to women who had never been pregnant. Antimicrosomal and antithyroid peroxidase concentrations did not change during iodide administration. These findings strongly suggest that euthyroid women with a previous episode of PPT have permanent subtle defects in thyroid hormone synthesis and are inordinately prone to develop iodide-induced hypothyroidism, similar to findings previously reported in euthyroid subjects with Hashimoto's thyroiditis, with a previous episode of painful subacute thyroiditis, or previously treated with radioactive iodine or surgery for Graves' disease.