Metabolismo de las hormonas tiroideas y el yodo en el embarazo / Thyroid hormone and iodine metabolism in pregnancy

El yodo es un micronutriente esencial necesario para que la glándula tiroides sintetice 2 hormonas yodadas: la tetrayodotironina (tiroxina, T4) y la 3’,3,5-triyodotironina (T3), con 4 y 3 átomos de yodo respectivamente. Son necesarias durante toda la vida, especialmente la T4 para el desarrollo de la corteza cerebral, desde el primer trimestre del embarazo. La necesidad de un aporte adecuado de yodo se reconoce entre los Derechos de la Infancia, ya que su deficiencia es, después de la inanición extrema, la causa nutricional más frecuente de retraso mental prevenible en el mundo. Aquí desarrollamos varios puntos: ¿son equivalentes la T4 y la T3 para el cerebro en desarrollo?; ¿qué ocurre con la T4 en condiciones de yodo deficiencia?; ¿qué cambios impone el feto mismo a la función tiroidea de la madre?; ¿qué ocurre cuando hay yodo deficiencia durante el embarazo?; ¿y la lactancia? Contestarlos explica por qué se duplican las necesidades de yodo desde el comienzo mismo del embarazo. Incluso en situaciones de yodo deficiencia leve-moderada, prevalentes todavía en España, se requiere la suplementación diaria con al menos (..) (AU)

Iodine is an essential micronutrient without which the thyroid is unable to synthesize and secrete its two iodine-containing hormones, tetra-iodo-thyronine or thyroxine (T4) and 3’, 3, 5-tri-iodothyronine (T3),containing, respectively, 4 and 3 iodine atoms per molecule. Both hormones are needed throughout life, with T4 being especially important for the development of the cerebral cortex as early as during the first trimester of pregnancy. The need for adequate iodine intake is recognized among the Rights of the Child, since, after starvation, iodine deficiency is the most frequent nutritional cause worldwide of preventable mental retardation. The present article discusses several questions: are T4and T3 equivalent for the developing brain? What happens to T4 during iodine deficiency? What changes are imposed on maternal thyroid function by the fetus? What happens when a pregnant woman is iodine deficient? What effect does breastfeeding have on iodine status? The answers to the above questions explain why iodine requirements are doubled from the very onset of pregnancy. Even in conditions of mild-moderate iodine deficiency, which still prevail throughout Spain, daily supplementation of at least (..) (AU)

Ontogenesis of thyroid function and interactions with maternal function.

Fetal and neonatal development of thyroid function involves the embryogenesis, differentiation and maturation of the thyroid gland, of the hypothalamic-pituitary-thyroid axis and of the systems controlling thyroid hormone metabolism. We focus here on aspects related to neurodevelopment. Throughout gestation, thyroxine (T4) transferred from the mother, present in embryonic fluids by 4 weeks, protects the fetal brain. Free T4 (FT4) in fetal fluids increases rapidly, approaching adult levels by midgestation, in concentrations that are determined by the maternal serum T4. T3 remains very low throughout pregnancy. In the cerebral cortex T3, generated from T4, reaches adult values by midgestation and is partly bound to specific nuclear receptor isoforms. The iodothyronine deiodinases are important for the spatial and temporal presence of T3 in different fetal brain areas. After onset of fetal thyroid secretion at midgestation, maternal transfer of T4 continues to contribute importantly to fetal serum T4, protecting neurodevelopment until birth. In rats, even a transient period of maternal hypothyroxinemia disrupts neurodevelopment irreversibly, supporting epidemiological evidence for its negative role in human neurodevelopment. The prompt treatment of maternal hypothyroidism or hypothyroxinemia should mitigate negative effects on neurodevelopment. Neurodevelopmental deficits of preterm infants might also result from an untimely interruption of the maternal transfer of T4 [Morreale de Escobar et al: J Clin Endocrinol Metab 2000;85:3975-3987; Best Pract Res Clin Endocrinol Metab 2004;18:225-248; Eur J Endocrinol 2004;151(suppl 3):U25-U37].

El yodo durante la gestación, lactancia y primera infancia. Cantidades mínimas y máximas: de microgramos a gramos / Iodine during pregnancy, lactation and infancy. Minimum and maximum doses: From micrograms to grams

Una deficiencia de yodo durante el período fetal y posnatal puede dar lugar a déficit de desarrollo mental y psicomotor, tanto más graves cuanto mayor sea la deficiencia de yodo y cuantos antes se haya padecido. Muchos déficit se han hecho irreversibles antes de la mitad de la gestación, por lo que hay que asegurar una ingesta adecuada de yodo (250-300 μg/día) mediante medidas de suplementación desde el comienzo del embarazo. Preferiblemente desde antes de su comienzo.Por elevada que sea la ingesta basal de yodo de la mujer la suplementación diaria con 250-300 μg de yodo no resulta nociva para la madre, el feto ni para el lactante.Hay un amplio margen de seguridad de dos a tres órdenes de magnitud entre las cantidades de yodo beneficiosas durante la gestación y primera infancia y las que pueden ser nocivas. Estas últimas son 500 a 3.000 veces superiores a las que se ingieren habitualmente, incluso cuando se toman suplementos. Las cantidades nocivas de yodo se relacionan invariablente con el uso de medicamentos, de antisépticos yodados (p. ej., povidona yodada) o de medios de contraste radiológicos. No hay actualmente excusa alguna para seguir usando antisépticos que, aunque normalmente inocuos en el adulto, son muy peligrosos durante el embarazo, parto y postparto. Su uso debe quedar terminantemente prohibido, sobre todo considerando que pueden ser sustituidos con eficacia por otros preparados no yodados, como clorhexidina al 0,05%. No debe olvidarse que el feto y el neonato, sobre todo si prematuro, carecen aún de los mecanismos de autorregulación tiroidea del adulto. Como consecuencia, ante un exceso de yodo se produce un un bloqueo del tiroides del feto y del neonato, sobre todo cuando éste es prematuro. Este hipotiroidismo iatrogénico puede dar lugar a defectos irreversibles de maduración cerebral, ya las consecuencias serían las mismas que las que se observan en niños con hipotiroidismo congénito, que no se han tratado a tiempo con tiroxina

An insufficient iodine intake during fetal and postnatal development often results in disorders of mental and psicomotor development, which are the more severe, the greater the degree of iodine deficiency and the earlier it is suffered during development. Many of these disorders have become irreversible by midgestation, and it is necessary to ensure the mother with an intake of 250- 300 μg I/day, from the beginning of the pregnancy or, preferably, before its onset, and throughout lactation. When the pregnancy is planned, supplementation ought to start before conception.Even if the basal iodine intake of the pregnant woman is more than adequate, supplementation with 250- 300 μg I/day is safe for her, the fetus and the newborn.There is a great margin of safety between the amounts of iodine which are beneficia, and those which are dangerous, as the latter are 2 to 3 orders of magnitude higher than the former. The amounts of iodine which may be harmful are 500-3,000 times higher than those we receive through food, even when dietary supplements are added. Harmful doses of iodine are invariably associated with the use of some medicines and drugs, mainly of iodinated disinfectants, or of radiological contrast agents. The damage results from the lack of maturation in fetuses and neonates, of thyroid autorregulatory mechanisms that protect the adult thyroid from the blocking effects of an iodine excess. There is at present no excuse for not totally banning, completely and permanently, the use of such disinfectants, especially during delivery, and the newborn period. Other disinfectants, such a clorhexidine, are equally effective without increasing the risk of inducing iatrogenic hypothyroidism during important phases of human brain development. It should continuously be born in mind that thyroid failure of the fetus and newborn, especially if preterm, may result in irreversible brain damage, whether its cause, is congenital or iatrogenic

Is neuropsychological development related to maternal hypothyroidism or to maternal hypothyroxinemia?

Several recent publications have drawn attention to the role of the thyroid hormone status of the mother on the future neuropsychological development of the child. The screening of pregnant women for clinical or subclinical hypothyroidism based on second trimester elevated maternal TSH values has been proposed. Here, we have summarized present epidemiological and experimental evidence strongly suggesting that conditions resulting in first trimester hypothyroxinemia (a low for gestational age circulating maternal free T4, whether or not TSH is increased) pose an increased risk for poor neuropsychological development of the fetus. This would be a consequence of decreased availability of maternal T4 to the developing brain, its only source of thyroid hormone during the first trimester; T4 is the required substrate for the ontogenically regulated generation of T3 in the amounts needed for optimal development in different brain structures, both temporally and spatially. Normal maternal T3 concentrations do not seem to prevent the potential damage of a low supply of T4, although they might prevent an increase in circulating TSH and detection of the hypothyroxinemia if only TSH is measured. Hypothyroxinemia seems to be much more frequent in pregnant women than either clinical or subclinical hypothyroidism and autoimmune thyroid disease, especially in regions where the iodine intake of the pregnant woman is inadequate to meet her increased needs for T4. It is proposed that the screening of pregnant women for thyroid disorders should include the determination of free T4 as soon as possible during the first trimester as a major test, because hypothyroxinemia has been related to poor developmental outcome, irrespective of the presence of high titers of thyroid autoantibodies or elevated serum TSH. The frequency with which this may occur is probably 150 times or more that of congenital hypothyroidism, for which successful screening programs have been instituted in many countries.

El yodo durante la gestación, lactancia y primera infancia. Cantidades mínimas y máximas: de microgramos a gramos / Iodine during pregnancy, lactation and infancy. Minimun and maximum doses: from micrograms

No disponible

Tissue-specific patterns of changes in 3,5,3'-triiodo-L-thyronine concentrations in thyroidectomized rats infused with increasing doses of the hormone. Which are the regulatory mechanisms?

We have measured 3,5,3'triiodothyronine (T3) in 12 tissues from thyroidectomized (Tx) rats infused with increasing doses of T3, and related them to their corresponding plasma levels. Young adult Wistar rats were surgically Tx. After 4 weeks, the animals were infused with placebo or T3 (0.25, 0.50, 0.75, 1.00 or 2.00 microg/100 g body weight/day). Placebo-infused intact rats served as euthyroid controls. Plasma and samples of cerebral cortex, cerebellum, brown adipose tissue (BAT), pituitary, liver, heart, lung, kidney, spleen, skeletal muscle, ovary and adrenal were obtained after 12-13 days of infusion. We determined plasma T3 and thyrotropin (TSH), and tissue T3 and thyroxine (T4), the latter being virtually undetectable. Results were compared with the relationships between tissue and plasma T3 in Tx rats on T4 infusions. Most tissues presented changes which paralleled those in plasma T3, irrespective of its source (infusion of T3, or generation from infused T4). However, at similar plasma T3 concentrations, cerebral cortex, cerebellum and BAT (containing type II 5' iodothyronine deiodinase (DII) activity), reached much lower T3 levels in the T3-infused Tx rats, than in Tx rats on T4, and required elevated plasma T3 levels for normal tissue T3. In these tissues, and in the pituitary, T3 concentrations were always lower than expected from plasma T3 levels. On the contrary, the lung and ovary of the T3-infused Tx rats contained more T3 than expected from plasma T3. Unexpectedly, both the ovary and adrenal attained higher tissue T3 concentrations in Tx rats on T3 than on T4 at comparable plasma T3 levels. In conclusion, the patterns of changes of the concentrations of T3 as a function of increasing plasma T3 are not only tissue-specific when T4 is provided, but also when circulating T3 is the only source of this iodothyronine. Further studies are needed to identify the mechanisms involved in the regulation of tissue T3 concentrations.

Immunohistochemical and morphometric studies of the fetal pancreas in diabetic pregnant rats. Effects of insulin administration.

BACKGROUND: Maternal diabetes influences fetal pancreas development. As there are some controversial reports, we studied the morphometric changes of the fetal insular pancreas and insulin immunostain of beta cells as well as the proliferative activity of insular cells in 21-day-old fetuses from control, diabetic, and insulin-treated diabetic pregnant rats. METHODS: Streptozotocin was injected into 7-day-pregnant rats (controls were not injected). Some rats were either left untreated (diabetic) or injected with insulin. Animals were killed at 21 days of gestation. Fetal pancreas were fixed in toto for the morphometry and immunohistochemistry studies using anti-insulin, anti-Ki-67 and anti-proliferating cell nuclear antigen (PCNA) antibodies. RESULTS: Diabetic status was determined by measuring maternal and fetal serum glucose and insulin levels. The morphometric studies showed hyperplasia of the diabetic fetal insular tissue which had not been normalized by insulin therapy. Diabetes caused an increase of both insulin-positive and insulin-negative cells. The increase in insulin-positive cells was not corrected by insulin treatment, although the number of non-beta cells became normal. The nuclear area in beta cells increased in diabetic rats but was not corrected by insulin. The cytoplasmic area decreased in diabetic rats and was normalized by insulin administration. Diabetes increased the expression of the nuclear antigen Ki-67 in fetal insular pancreas, and insulin treatment returned it to the normal state. CONCLUSIONS: Maternal diabetes leads to hyperstimulation of fetal beta cells, with increased proliferative activity. Insulin administration to the dams corrects some of the changes observed.

[Current situation of goiter endemic and iodine intake in the population of the Pyrenees and the Segrià region of Lleida]. / Situación actual de la endemia de bocio y del consumo de yodo en la población del Pirineo y de la comarca del Segrià de Lleida.

OBJECTIVE: To assess the prevalence of goiter in five areas of the Pyrineans and in the region of Segrià in Lleida. DESIGN: Transversal descriptive study. SETTING: Five Pyrinean's regions and Segrià. PATIENTS: Randomised selection of a sample of 601 subjects from the population over 6 years old. MEASUREMENTS: The field work, which was preceded by an informative campaign in the media, was carried out from October of 1994 through February of 1995. Survey with a personal interview, blood pressure, weight, height, goiter palpation, blood analysis with thyroidal hormones and urine analysis with the iodine/creatine ratio determination. RESULTS: The prevalence of goiter was 18.3% which was higher among women, the ratio being 3.7/1 (women/men). No significant differences were found in regard to geographic distribution. Mean iodinuria was 120 micrograms/l, though it was below 50 in 11.1% subjects. The prevalence of goiter has been founded to be related to age, increasing from the age of 45 onwards. Higher percentage of goiter was found among individuals with a family history of the disease and women who have had children. The prevalence of hypothyroidism was 3.4%. CONCLUSIONS: We have found a medium degree of goiter's endemia in the study area, the mean iodinuria in the population is in normal range. The women with children have a higher prevalence of goiter probably due to a lack of sufficient iodine intake being a subgroup at risk.

Thyroid hormones influence serum leptin concentrations in the rat.

Leptin, the product of the ob gene, is secreted by adipocytes and has been shown to decrease appetite and increase energy expenditure. Leptin mRNA in adipocytes correlates with body wt, and serum leptin levels correlate with body fat. Alterations in thyroid status are frequently associated with changes in body wt. To evaluate the possible influence of thyroid status on the leptin system, we have measured serum leptin concentrations in thyroidectomized rats infused either with placebo, or with different doses of T4 (0.8 to 8.0 microg/100 g body wt per day) or T3 (0.25 to 2.0 microg/100 g body wt per day), covering a wide range of thyroid hormone concentrations, from overt hypothyroidism to hyperthyroidism. Intact animals infused with placebo were used as euthyroid controls. Infusion of T4 or T3 into thyroidectomized rats resulted in a decrease in serum leptin levels with respect to the thyroidectomized animals infused with placebo. When compared to the control group, serum leptin levels were decreased in the groups infused with the higher T4 and T3 doses, and tended to be elevated in the thyroidectomized animals infused with placebo. The leptin/body wt ratio was markedly increased in thyroidectomized rats infused with placebo, and decreased in the animals infused with the higher thyroid hormone doses. In conclusion, thyroid hormones exert a negative influence on serum leptin concentrations, which is greater than expected by the changes in body wt The precise mechanism of this influence remains to be elucidated.

Myelin basic protein immunoreactivity in the internal capsule of neonates from rats on a low iodine intake or on methylmercaptoimidazole (MMI).

Rats fed on low iodine diets (LIDs) result in a normal circulating level of triiodothyronine (T3), a low level of thyroxine (T4) and an elevated thyroid-stimulating hormone (TSH). These changes are similar to those observed in habitants who live in iodine-deficient areas and different from those observed when the hypothyroidism is produced by goitrogens. To study the effects of LID or goitrogens on the myelin basic protein (MBP) immunoreactivity (MBP-ir) during the myelination of the internal capsule, one group of experimental female rats was fed on an LID, and another group received a standard laboratory diet with methylmercaptoimidazole (MMI) added in the drinking water. Animals fed on a standard laboratory diet and animals fed on an LID supplemented with KI were used as controls. At P10, the MMI treatment has produced a more marked decrease in the surface density of MBP-ir processes with respect to controls than that produced in the LID animals. This decrease was correlated with the cerebral concentrations of triiodothyronine (T3) we found. During the postnatal development, a recovery in the levels of the surface density with respect to controls was observed in both experimental groups. The recovery occurred by P20 in the LID group and by P32 in the MMI rats.

Regulation of iodothyronine deiodinase activity as studied in thyroidectomized rats infused with thyroxine or triiodothyronine.

To provide new insights into the in vivo regulation of iodothyronine deiodinases in the different tissues of the rat, we have evaluated the effects on these enzymatic activities of T4 or T3 infusions into thyroidectomized rats. Thyroidectomized rats were infused with placebo, T4, or T3. Placebo-infused intact rats served as euthyroid controls. Plasma and samples of cerebral cortex, brown adipose tissue, pituitary, liver, and lung were obtained after 12-13 days of infusion. Plasma TSH, plasma and tissue T4 and T3, and iodothyronine deiodinase activities were determined. Type II 5'-deiodinase (DII) was increased in cortex, brown adipose tissue, and pituitary from animals infused with placebo. DII activity returned to normal only with T4 infusion, remaining elevated in the animals infused with T3 alone despite normal tissue T3 concentrations. Cortex type III 5-deiodinase was only increased when hyperthyroidism was induced by infusion of T3. Liver type I 5'-deiodinase (DI) paralleled the changes in plasma and tissue T3 regardless of whether T4 or T3 was infused. On the contrary, the increase in lung DI, proportional to the increases in plasma and tissue T3, was higher when T4 was infused. As a rule, the tissues with DII presented a tighter homeostasis in their T3 concentrations than the tissues with DI. In conclusion, the regulation of iodothyronine deiodinases depends on the hormone infused into the thyroidectomized animals and on the tissue in which the deiodinase is studied, demonstrating the existence of tissue-specific regulation of its thyroid hormone concentrations.

Maternal nonthyroidal illness and fetal thyroid hormone status, as studied in the streptozotocin-induced diabetes mellitus rat model.

We have used the streptozotocin-induced diabetes mellitus pregnant rat as a model of maternal nonthyroidal illness. We measured the effects of different degrees of diabetes mellitus on maternal body weight, the outcome of pregnancy, circulating glucose, insulin, T4, T3, rT3, and TSH in mother and fetus, T4 and T3 in maternal and fetal tissues, and iodothyronine deiodinases in liver, lung, and brain. All of the changes in thyroid hormone status typical of nonthyroidal illnesses were observed in the mothers and were related to the degree of the metabolic imbalances. Most were controlled with a daily insulin dose of 0.5 U/100 g BW. Normalization of maternal placental T4, however, required higher insulin doses than in other maternal tissues. The number and body weight of the fetuses, their pituitary GH contents, and their thyroid hormone status were severely affected. The total extrathyroidal T4 and T3 pools decreased to one third of normal fetal values. T4 and T3 concentrations in the fetal brain were lower than normal, and the expected increase in type II 5'deiodinase activity was not observed. The low cerebral T3 only improved with adequate insulin treatment of the dams. It is concluded that maternal diabetes mellitus, and possibly other nonthyroidal illnesses that impair the availability of intracellular energy stores, may affect fetal brain T3 when thyroid hormones are essential for normal development.

Maternal diabetes mellitus, a rat model for nonthyroidal illness: correction of hypothyroxinemia with thyroxine treatment does not improve fetal thyroid hormone status.

Maintenance of normal maternal thyroxinemia prevents severe triiodothyronine (T3) deficiency of the fetus with primary thyroid failure (1). We have studied whether thyroxine (T4) would also protect the fetal brain when maternal hypothyroxinemia is caused by nonthyroidal illnesses. We have used the streptozotocin-induced diabetes mellitus pregnant rat as a model of maternal nonthyroidal illness. We measured the effects of diabetes mellitus, and of correction of the ensuing maternal hypothyroxinemia with T4 as compared to insulin, on maternal body weight, the outcome of pregnancy, glucose, insulin, T4, T3, reverse T3, and thyrotropin levels in the maternal and fetal circulation, as well as T4 and T3 concentrations in tissues, and iodothyronine deiodinases in liver, lung, and brain. The diabetic mothers showed changes in thyroid hormone status typical of nonthyroidal illnesses. Thyroid hormone status of the fetuses was severely affected: the total T4 and T3 pools decreased to one-third of normal values. T4 and T3 concentrations in the fetal brain were lower than normal and the expected increase in 5'-deiodinase activity was not observed. Although insulin treatment avoided or mitigated these changes, the low cerebral T3 did not improve with T4 treatment of the maternal hypothyroxinemia. Several findings indicated that treatment of the severely ill dams with T4 was actually harmful for the outcome of pregnancy. These negative effects were observed without the expected increase in the maternal or fetal T3 pools.

Regulation of uncoupling protein messenger ribonucleic acid and 5'-deiodinase activity by thyroid hormones in fetal brown adipose tissue.

We studied the regulation of type II 5'-deiodinase (5'D-II) activity and uncoupling protein (UCP) messenger RNA (mRNA) by thyroid hormones in fetal brown adipose tissue (BAT). Fetuses were obtained from hypothyroid pregnant rats infused with increasing doses of T4 or T3. Infusion of T4 into hypothyroid pregnant rats increased T4 and T3 concentrations and inhibited 5'D-II activity in fetal BAT. In contrast, infusion of T3 increased BAT 5'D-II activities at low, normal, or high BAT T3 concentrations. The relationship between thyroid hormone concentrations in fetal BAT and plasma showed that BAT T3 concentrations are relatively stable, increasing less than 2-fold over a wide range of circulating T4 (3-fold) or T3 (8-fold) concentrations. Most T3 in fetal BAT are locally derived from T4 and not from plasma T3. UCP mRNA expression decreased to 30% of control values in hypothyroid fetuses. UCP mRNA levels were restored to normal in parallel with BAT T3 concentrations after the infusion of either T4 or T3. UCP mRNA levels correlate well with BAT T3 concentrations. Supraphysiological doses of T4 did not further increase either BAT T3 or UCP mRNA levels. T3 might regulate basal UCP mRNA expression during fetal life.

Displacement of T4 from transthyretin by the synthetic flavonoid EMD 21388 results in increased production of T3 from T4 in rat dams and fetuses.

EMD 21388 displaces T4 from circulating transthyretin, is a potent in vitro inhibitor of outer-ring deiodination (5'D) of T4 and affects thyroid hormone secretion. To study its extrathyroidal effects on the thyroid hormone status of pregnant dams and their fetuses, we treated the dams with methimazole and infused them with T4 and with either 2.5 mg EMD 21388/rat per day [(EMD(+)], or placebo solution [EMD(-)]. EMD reduced total T4 and T3 in the maternal circulation, but free T4 increased and free T3 decreased. The total amount of T3 generated from T4 in the maternal compartment increased. Placental T3 also increased in EMD(+) animals, T4 remaining the same. EMD also reached the fetal circulation. The total fetal extrathyroidal T4 pool decreased to half that of EMD(-) fetuses, whereas T3 increased 1.8-fold, thus mitigating fetal T3 deficiency, especially in the lung. Thus, if the maternal supply of T4 is kept constant, EMD mitigates the T3 deficiency of many tissues of the hypothyroid fetus. Most of the effects of this dose of EMD could result from the displacement of T4 from circulating transthyretin. Liver 5'D-I activity did not decrease, but actually increased by 40% in dams and fetuses. The enhanced transfer of T4 into tissues would also increase the amount of substrate available to 5'D-I, leading to an increased amount of T3 in maternal and fetal pools. This had not been anticipated from the changes in circulating T4 and T3, whether maternal or fetal, total or free.

Replacement therapy for hypothyroidism with thyroxine alone does not ensure euthyroidism in all tissues, as studied in thyroidectomized rats.

We have studied whether, or not, tissue-specific regulatory mechanisms provide normal 3,5,3'-triiodothyronine (T3) concentrations simultaneously in all tissues of a hypothyroid animal receiving thyroxine (T4), an assumption implicit in the replacement therapy of hypothyroid patients with T4 alone. Thyroidectomized rats were infused with placebo or 1 of 10 T4 doses (0.2-8.0 micrograms per 100 grams of body weight per day). Placebo-infused intact rats served as controls. Plasma and 10 tissues were obtained after 12-13 d of infusion. Plasma thyrotropin and plasma and tissue T4 and T3 were determined by RIA. Iodothyronine-deiodinase activities were assayed using cerebral cortex, liver, and lung. No single dose of T4 was able to restore normal plasma thyrotropin, T4 and T3, as well as T4 and T3 in all tissues, or at least to restore T3 simultaneously in plasma and all tissues. Moreover, in most tissues, the dose of T4 needed to ensure normal T3 levels resulted in supraphysiological T4 concentrations. Notable exceptions were the cortex, brown adipose tissue, and cerebellum, which maintained T3 homeostasis over a wide range of plasma T4 and T3 levels. Deiodinase activities explained some, but not all, of the tissue-specific and dose related changes in tissue T3 concentrations. In conclusion, euthyroidism is not restored in plasma and all tissues of thyroidectomized rats on T4 alone. These results may well be pertinent to patients on T4 replacement therapy.