Introduction

The pituitary gland consists of two parts, the anterior pituitary, or adenohypophysis, and the posterior pituitary, or neurohypophysis. Although closely related anatomically, they are embryologically and functionally quite distinct. The anterior pituitary comprises primarily glandular tissue, while the posterior pituitary is of neural origin. The pituitary gland is situated at the base of the brain, in close relation to the hypothalamus (Fig. 7.1), which has an essential role in the regulation of pituitary function.

Figure 7.1 Diagrammatic sagittal section through part of the brain to show the anatomical relationship of the pituitary gland and hypothalamus. The portal blood vessels, through which hypothalamic hormones reach the anterior pituitary, and nerve fibres, which transport hypothalamic hormones to the posterior pituitary, are shown.

Anterior pituitary hormones

The anterior pituitary secretes several hormones, some of which are trophic; that is, they stimulate the activity of other endocrine glands (Fig. 7.2). The secretion of hormones by the anterior pituitary is controlled by hormones secreted by the hypothalamus, which reach the pituitary through a system of portal blood vessels. The secretion of hypothalamic hormones is influenced by higher centres in the brain, and the secretion of both hypothalamic and pituitary hormones is regulated by feedback from the hormones whose production they stimulate in target organs. Most of the blood supply of the anterior pituitary gland is derived from the hypothalamus, ensuring that it is exposed to high concentrations of the hypothalamic hormones. However, the concentrations of these hormones in the systemic circulation are low.

Figure 7.2 Anterior pituitary hormones and their actions. All the actions shown are stimulatory; trophic hormones stimulate both synthesis and release of hormones by their target organs.

Growth hormone

Growth hormone (GH, somatotrophin) is a 191 amino acid polypeptide hormone. It is essential for normal growth, although in the main it acts indirectly by stimulating the liver to produce insulin-like growth factor-1 (IGF-1), also known as somatomedin-C. IGF-1 has considerable amino acid sequence homology with insulin, and shares some of the actions of this hormone. GH also has a number of metabolic effects, which are summarized in Figure 7.3. The release of GH is controlled by two hypothalamic hormones: growth-hormone-releasing hormone (GHRH) and somatostatin (also known as somatotrophin release-inhibiting factor, SRIF). IGF-1 exerts negative feedback at the level of the pituitary, where it modulates the actions of GHRH, and at the level of the hypothalamus where, together with GH itself, it stimulates the release of somatostatin.

Figure 7.3 Metabolic actions of growth hormone.

The unit of measurement of GH was changed in the UK in 2008 from mU/L to µg/L: values here are given in µg/L with the equivalent in mU/L in brackets. The concentration of GH in the blood varies widely through the day and may at times be undetectable (<0.3 µg/L (<1 mU/L)) with currently available assays. Physiological secretion occurs in sporadic bursts, lasting for 1–2 h, mainly during deep sleep. Peak concentrations may be as high as 13.3 µg/L (40 mU/L). The rate of secretion increases from birth to early childhood, and then remains stable until puberty, when a massive increase occurs, stimulated by testosterone in males and oestrogens in females; thereafter the rate of secretion declines to a steady level, before falling to low levels in old age. Secretion can be stimulated by stress, exercise, a fall in blood glucose concentration, fasting and ingestion of certain amino acids. Such stimuli can be used in provocative tests for diagnosing GH deficiency. GH secretion is inhibited by a rise in blood glucose, and this effect provides the rationale for the use of the oral glucose tolerance test in the diagnosis of excessive GH secretion. Excessive secretion (usually due to a pituitary tumour) causes gigantism in children and acromegaly in adults; deficiency causes growth retardation in children, and can cause fatigue, loss of muscle strength, impaired psychological well-being and an adverse cardiovascular risk profile (elevated plasma total and low density lipoprotein (LDL) cholesterol concentrations and hyperfibrinogenaemia) in adults.

Somatostatin, the 14 amino acid hypothalamic peptide that inhibits GH secretion, has many other actions, both within the hypothalamo-pituitary axis and elsewhere. For example, it inhibits the release of thyroid-stimulating hormone (TSH) in response to thyrotrophin-releasing hormone (TRH), and it is present in the gut and pancreatic islets, where it inhibits the secretion of many gastrointestinal hormones, including gastrin, insulin and glucagon. The physiological significance of these actions is poorly understood. Rare somatostatin-secreting tumours of the pancreas have been described, and somatostatin secretion can also occur from medullary carcinomas of the thyroid and small cell carcinomas of the lung. Somatostatin analogues are used therapeutically to stop bleeding from oesophageal varices (an unlicensed indication), to inhibit hormone secretion by tumours and to treat acromegaly. A third hormone, ghrelin, also affects GH secretion. The main site of its production is the stomach, and it is involved in the regulation of appetite (see Chapter 20), but it is also produced in the hypothalamus and stimulates GH secretion.

Gonadotrophins

Follicle-stimulating hormone (FSH) and luteinizing hormone (LH) are both glycoproteins having a molecular weight of approximately 30 kDa, and consist of two subunits: the β-subunits are unique to each hormone, but the amino acid sequence of the α-subunits is the same, as it also is in those of both TSH and hCG.

The synthesis and release of both hormones are stimulated by the hypothalamic decapeptide, gonadotrophin-releasing hormone (GnRH), these effects being modulated by circulating gonadal steroids. GnRH is secreted episodically, resulting in pulsatile secretion of gonadotrophins with peaks in plasma concentration occurring at approximately 90-min intervals. In males, LH stimulates testosterone secretion by Leydig cells in the testes: both testosterone and oestradiol, derived from the Leydig cells themselves and from the metabolism of testosterone, feed back to block the action of GnRH on LH secretion. FSH, in concert with high intratesticular testosterone concentrations, stimulates spermatogenesis; its secretion is inhibited by inhibin (Fig. 7.4), a hormone produced during spermatogenesis.

Figure 7.4 Control of testicular function by pituitary gonadotrophins.

In females, the relationships are more complex. Oestrogen (mainly oestradiol) secretion by the ovaries is stimulated primarily by FSH in the first part of the menstrual cycle; both hormones are necessary for the development of Graafian follicles. As oestrogen concentrations in the blood rise, FSH secretion declines until oestrogens trigger a positive feedback mechanism, causing an explosive release of LH and, to a lesser extent, FSH. The increase in LH stimulates ovulation and development of the corpus luteum, but rising concentrations of oestrogens and progesterone then inhibit FSH and LH secretion; inhibin from the ovaries also appears to inhibit FSH secretion. If conception does not occur, declining concentrations of oestrogens and progesterone from the regressing corpus luteum trigger menstruation and the release of LH and FSH, initiating the maturation of further follicles in a new cycle (Fig. 7.5). Before puberty, plasma concentrations of LH and FSH are very low and unresponsive to exogenous GnRH. With the approach of puberty, FSH secretion increases before that of LH.

Figure 7.5 Changes in the plasma concentration of pituitary gonadotrophins during the menstrual cycle. The resultant changes in oestrogens (17β-oestradiol) and progesterone concentration are also shown.

Increased concentrations of gonadotrophins are seen in ovarian failure in women, whether pathological or after the natural menopause. High concentrations of FSH are seen in azoospermic men, and LH is increased if testosterone secretion is decreased.

Gonadotrophin-secreting tumours (secreting either LH or FSH) of the pituitary are rare. Decreased gonadotrophin secretion, leading to secondary gonadal failure, is more common. It can either be an isolated phenomenon, due to hypothalamic dysfunction, or occur with generalized pituitary failure. A case of hypogonadotrophic hypogonadism is described in Case history 10.1.

Adrenocorticotrophic hormone

Adrenocorticotrophic hormone (ACTH) is a polypeptide (molecular weight 4500 Da), comprising a single chain of 39 amino acids. Its biological function, which is to stimulate adrenal glucocorticoid (but not mineralocorticoid) secretion, is dependent on the N-terminal 24 amino acids. ACTH is a fragment of a much larger precursor, pro-opiomelanocortin (POMC, molecular weight 31 kDa) (Fig. 7.6), which is the precursor not only of ACTH but also of β-lipotrophin, itself the precursor of endogenous opioid peptides (endorphins). The control of the release of β-lipotrophin and the endorphins has not been fully elucidated, but ACTH release is controlled by a hypothalamic peptide, corticotrophin-releasing hormone (CRH). ACTH secretion is pulsatile and also shows diurnal variation, the plasma concentration being highest at approximately 08:00 h and lowest at midnight. Secretion is greatly increased by stress and is inhibited by cortisol. Thus cortisol secretion by the adrenal cortex is controlled by negative feedback, but this and the circadian variation can be overcome by the effects of stress. The normal value for plasma ACTH concentration at 09:00 h is <50 ng/L.

Figure 7.6 ACTH is derived by proteolysis of a precursor, pro-opiomelanocortin. β-Lipotrophin is derived from the same precursor and is itself the precursor of endorphins and enkephalins (naturally occurring peptides with opioid-like activity). Melanocyte-stimulating hormones are secreted in some species, but not humans. (Those hormones that are secreted physiologically in humans are shown in heavier outlines.)

Increased secretion of ACTH by the pituitary is seen with pituitary tumours (Cushing’s disease) and in primary adrenal failure (Addison’s disease). The hormone may also be secreted ectopically by non-pituitary tumours. Excessive ACTH synthesis is associated with increased pigmentation, owing to the melanocyte-stimulating action of ACTH and other POMC-derived peptides. Decreased secretion of ACTH may be an isolated phenomenon but is more commonly associated with generalized pituitary failure.

Measurement of anterior pituitary hormones

Measurements of pituitary hormone concentrations are required in both suspected hypofunction and hyperfunction (the latter is usually the result of a pituitary tumour, and is often accompanied by partial hypofunction). The investigation of suspected pituitary hypofunction should begin with measurement of pituitary and target organ hormones in a blood sample taken at 09:00 h. TSH deficiency will be apparent from a low total or free thyroxine concentration without the elevation of TSH characteristic of primary hypothyroidism. Plasma TSH concentration may be normal or low in hypopituitarism: it is rarely undetectable.

In males, a normal plasma testosterone concentration indicates normal LH secretion. In hypopituitarism, plasma testosterone concentration is low and LH and FSH concentrations are normal or low. In premenopausal females, amenorrhoea with a low plasma oestradiol concentration and normal or low gonadotrophins suggests hypothalamic or pituitary dysfunction. A clomifene test (see p. 170) may help to distinguish between these. A normal ovulatory plasma progesterone concentration (see p. 176) indicates the integrity of the hypothalamo-pituitary–ovarian axis without the need for further testing; a history of regular, normal menstrual cycles also effectively excludes gonadotrophin deficiency. In normal postmenopausal women, plasma gonadotrophin concentrations are grossly elevated; in hypopituitarism, they are normal or low.

Tests involving the administration of TRH and GnRH followed by measurement of TSH and gonadotrophins have traditionally been used in the investigation of pituitary disease, often combined with the insulin hypoglycaemia test (IHT, see below). However, the use of these tests has been criticized on the grounds that the responses to these releasing hormones only reflect the readily releasable pituitary pools of the hormones concerned and do not assess the physiological integrity of the pituitary. Normal responses can occur in spite of other evidence of pituitary hypofunction. The response to the releasing hormones is often delayed in patients with hypothalamic, as opposed to pituitary, dysfunction, but such delayed responses can also occur in pituitary disease. In practice, the releasing hormone tests often add little to what can be deduced from clinical observation and the results of basal hormone measurements.

Because GH is secreted sporadically, it may be undetectable in the plasma of normal individuals. Thus while a concentration of >6.7 µg/L (>20 mU/L) in a single sample excludes significant deficiency, a low concentration is not necessarily indicative of deficiency. Growth hormone secretion can be assessed using the IHT: a peak plasma concentration <6.7 µg/L (<20 mU/L) after adequate hypoglycaemia (blood glucose concentration <2.2 mmol/L) is reliable evidence of GH deficiency.

Because the IHT is potentially hazardous, various other tests of GH secretion have been devised, involving the administration of, for example, GHRH, glucagon, arginine, yeast extract or l-DOPA, although the relevance of these pharmacological stimuli to the physiological secretion of GH is questionable. GH concentrations >6.7 µg/L (>20 mU/L) are usually regarded as excluding GH deficiency, but lesser responses are not conclusive evidence of deficiency. Vigorous exercise also stimulates GH secretion, but even with standardized protocols an apparently subnormal response may not indicate GH deficiency. More reliable information may be provided by the measurement of GH secretion during sleep, by means of frequent blood sampling through an indwelling cannula, but there are obvious practical drawbacks to this procedure.

Measurements of IGF-1 are increasingly being used together with GH stimulation tests in the investigation of suspected GH deficiency. A low plasma concentration of IGF-1, together with an impaired or absent GH response to stimulation, confirms GH deficiency. Some patients who appear clinically to have GH deficiency have normal or elevated plasma GH concentrations but, because of a receptor or intracellular signalling defect, are resistant to its action. This syndrome is known as Laron dwarfism: patients have low plasma IGF-1 concentrations. It should be noted that, while plasma IGF-1 concentrations are much more stable than those of GH, they do vary with age and nutritional status: measured values should always be assessed with reference to age-and sex-matched reference values. The IGFs are carried in the plasma bound to IGF-binding proteins (IGFBPs), and measurement of IGFBP-3 may be a better marker of growth hormone deficiency in children.

The integrity of the hypothalamo-pituitary–adrenal axis can also be tested using the IHT. A rise in plasma cortisol concentration to at least 550 nmol/L after adequate hypoglycaemia indicates a normal axis. It has been shown that if the basal (09:00 h) plasma cortisol concentration is <100 nmol/L, the cortisol response to hypoglycaemia is never normal, whereas it invariably is normal if the basal concentration is >400 nmol/L. A formal IHT may therefore not be necessary in patients whose basal plasma cortisol concentrations are outside the range 100–400 nmol/L. The protocol for the IHT is given in Figure 7.7. The short ACTH stimulation test (tetracosactide or Synacthen test, see p. 141), used primarily in the investigation of adrenal failure, has also been advocated as a test for ACTH deficiency. This may seem illogical, but the rationale is that ACTH deficiency causes adrenal atrophy and thus decreases adrenal responsiveness to ACTH. A good correlation between the results of the IHT and short ACTH stimulation tests has been demonstrated: a plasma cortisol concentration >550 nmol/L 30 min after the administration of synthetic ACTH (250 µg, i.v.) excludes ACTH deficiency. Experience with the low dose (1 µg) tetracosactide test in this context is presently limited, but it may be less sensitive in identifying partial failure of ACTH secretion.

Figure 7.7 Combined test (triple bolus test) of anterior pituitary function. In patients thought to be very likely to be hypopituitary, the insulin dose should be 0.1 U/kg body weight; in patients with Cushing’s disease or acromegaly, a dose of 0.30 U/kg may be used. When glucagon (1 mg i.m.) is used instead of insulin, blood samples for cortisol and GH should be taken at 30 min intervals from 90 to 240 min after the injection (GH and cortisol responses occur later than when insulin is used).