Changes of the Setpoint
	In the hypothalamus-pituitary-thyroid axis the thyroid gland is the servo-generator producing thyroid hormones. The response - i.e. ambient serum thyroid hormone concentrations - is sensed by the pituitary, which compares it with the command signal (TRH) from the hypothalamus. The pituitary as a comparator generates an error signal, which delivers the control signal (TSH) to the thyroid gland. The setpoint of the pituitary-thyroid axis is thus determined by the hypothalamus and/or the pituitary. The setpoint may change under the influence of various conditions. Age is presumably one of the physiological determinants of the setpoint. In newborns and during early infancy serum TSH values are relatively high for the concomitant FT4 concentrations as compared to adults. In old age the mean 24h TSH concentration is decreased with a dampening of the circadian variation in TSH secretion, although serum FT4 levels usually do not change; it suggests a decrease in the command signal TRH from the hypothalmus. Changes in setpoint as a function of age are likely visualized as parallel movements of the setpoint line without alteration of its slope. Pathologic changes of the setpoint can be due to structural damage of the hypothalamus-pituitary area, e.g. by tumors. More intriguing are functional changes of the setpoint in the absence of anatomical lesions. Many infants with Down's syndrome have high TSH values which are inappropriate in view of their serum T4 concentrations that do not differ from age matched controls (14); thyroid autoantibodies are not detected in their serum, and the phenomenon remains unexplained, but an altered setting of the pituitary-thyroid axis might well be involved. A similar combination of increased TSH but normal T3 and T4 serum concentrations can be found in families with loss of function mutations of the TSH receptor (16). Such cases are likely to be found in subjects labeled as having subclinical hypothyroidism, especially when autoimmune markers and inborn errors of thyroid hormone biosynthesis are absent. Another example of setpoint modulation might be the sick euthyroid syndrome in which a decrease of serum FT4 is not followed by an increase of serum TSH. Recent studies have shown that the expression of TRH in the human paraventricular nucleus is decreased in nonthyroidal illness, in direct relationship to the decrease of serum T3 (15). In this case the altered setpoint is likely characterized by a change in the slope of the line, possibly mediated by a change in release of cytokines affecting the command signal TRH. A final intriguing matter in this respect is the occurrence of perfectly normal FT4 and FT3 concentrations despite suppressed TSH levels in clinically euthyroid Graves' patients without any medication.This condition may persist for a rather long time in some patients after appropriate treatment of the hyperthyroid state, and has been incompletely understood. The recent description of the presence of the TSH receptor in the human pituitary might explain this phenomenon (16). It is hypothesized that the pituitary TSH receptor senses the output of the control signal TSH, which by virtue of a short-loop feedback can be advantageous for fast adjustments of the pituitary release of TSH. Circulating antibodies directed against the TSH receptor might also be recognized by the pituitary TSH receptor causing down- regulation of TSH secretion.