GH is a peptide that is secreted by the acidophil cells of the pituitary. GH release depends on the interplay between two opposing hormones secreted by the hypothalamus; growth hormone-releasing hormone (GHRH), which stimulates GH release, and somatostatin, which inhibits GH release. GH exerts its effect through stimulation of various tissues to produce a group of peptide growth factors called “somatomedins.” The most important somatomedin is somatomedin-C, which is produced by the liver and which acts on cartilage. High levels of somatomedin participate in a GH feedback mechanism by inhibiting pituitary release of GH and also by stimulating hypothalamic secretion of somatostatin. In addition, there apparently are numerous other factors that influence GH production to varying degree, some of which act through the hypothalamus and others that apparently do not. Normal GH secretion takes place predominantly (but not entirely) during sleep in the form of short irregularly distributed secretory pulses or bursts. GH deficiency most often is suspected or questioned as part of the differential diagnosis of retarded growth or short stature. A considerable number of hormones and factors are necessary for adequate growth. Some of the most important are listed in the box on this page. Etiologies of short stature (whose extreme result is dwarfism) are listed in the box. Diagnosis of many of these conditions is discussed elsewhere. Deficiency of GH will be discussed in the context of pediatric or childhood growth retardation. As noted previously, GH deficiency may be part of multihormone pituitary dysfunction but may exist as an isolated defect in an otherwise apparently normal pituitary.

GH deficiency tests

Growth hormone assay. The most widely used screening test for GH deficiency is serum GH assay. Values are elevated by sleep, exercise of any type, and various foods. Therefore, if a single basal level is obtained, the specimen should be secured in the morning after an overnight fast, and the patient should be awake but not yet out of bed. Because basal normal range for single specimens includes some normal persons with values low enough to overlap the range of values found in patients with GH deficiency, only a result high enough to rule out GH deficiency is significant. Low results either may be due to GH deficiency or may simply be low-normal. Other conditions that increase GH levels besides those previously mentioned include severe malnutrition and chronic liver or renal disease. Patients with psychiatric depression or poorly controlled diabetes mellitus secrete more GH than normal persons during waking hours but not during sleep.

Body Factor Snecessary for Normal Growth
Sex hormones
Adequate nutritional factors
Partial List of Short Stature Etiologies
Familial short stature
Constitutional growth delay
Turner’s syndrome
Trisomy syndromes (e.g., Down’s syndrome)
GH deficiency (isolated or hypopituitarism)
Severe malnutrition
Uncontrolled type I diabetes mellitus
Severe chronic illness with organ failure (chronic renal disease, congenital heart disease, severe chronic hemolytic anemia)
Excess androgens or estrogens
Excess cortisol (Cushing’s syndrome or iatrogenic)
Psychosocial retardation (abused or severely emotionally deprived children)

Two relatively simple screening tests for GH deficiency have been advocated. It has been found that a major part of daily GH secretion takes place in a short period beginning approximately 1 hour after onset of deep sleep (electro-encephalogram [EEG] stage 3 or 4). Therefore, one method is to observe the patient until deep sleep occurs, wake the patient 60-90 minutes later, and quickly draw a blood sample for GH assay. About 70% of normal children or adults have GH values sufficiently elevated (і7 ng/ml; 7 µg/L) to exclude GH deficiency. Strenuous exercise is a strong stimulus to GH secretion. An exercise test in which the patient exercises vigorously for 20-30 minutes and a blood sample is drawn for GH assay 20-30 minutes after the end of the exercise period has been used. About 75%-92% of normal children have sufficiently elevated levels of GH after exercise to exclude GH deficiency.

A third method involves integrated multihour time period blood specimens; this method is being used in research and in some medical centers. Because of the irregular pulsatile nature of GH secretion, it was found that a more reliable estimate of GH secretion levels could be obtained by drawing blood specimens for GH assay every 20-30 minutes from an indwelling catheter over a 24-hour period. Some investigators report that the same information can be obtained using an 8 P.M. -8 A.M. 12-hour period while the patient is sleeping. The integration (averaging) of the GH specimen values correlate much better with clinical findings, especially in patients with partial GH deficiency (also known as “neurosecretory GH deficiency”), than does a single-specimen value. This test is being used when standard tests are equivocal or normal in spite of clinical suspicion of GH deficiency.

Somatomedin-C assay. Somatomedin-C assay has been proposed as a screening test for GH deficiency, pituitary insufficiency, and acromegaly. As noted previously, somatomedin serum levels depend on serum GH levels, and it has been postulated that the growth-promoting actions of GH are actually carried out by the somatomedins. Somatomedin-C is produced in the liver and circulates in serum predominantly bound to certain long-lived high molecular weight serum proteins. Normal values for females are about 10%-20% higher than those for males in both children and adults. Normal values also depend on age and developmental stage. There is a considerable increase during the pubertal adolescent growth spurt. Radioreceptor and immunoassay measurement techniques are available in large reference laboratories for somatomedin-C. Serum levels are low in most patients with isolated GH deficiency or pituitary insufficiency. The assay has some advantages in that the result does not depend on relation to food intake or time of day. Disadvantages are that only large medical centers or reference laboratories offer the assay and that various conditions besides pituitary dysfunction may produce a low result. These nonpituitary conditions include decreased caloric intake and malnutrition, malabsorption, severe chronic illness of various types, severe liver disease, hypothyroidism, and Laron dwarfism. Therefore, a clearly normal result is more diagnostic than a low result, since the low result must be confirmed by other tests. Somatomedin-C levels are elevated in acromegaly (discussed later).

Growth hormone stimulation tests. Since only a clearly normal single-specimen GH result excludes pituitary GH secretion deficiency, in patients with lower single-specimen GH values a stimulation test is needed for detection of deficiency or confirmation of normality. The classic procedures are insulin-induced hypoglycemia and arginine infusion. Pituitary insufficiency cannot be documented on the basis of pituitary nonresponse to either test agent alone, since about 20% of normal persons may fail to respond satisfactorily to insulin alone or to arginine alone. Estrogen administration tends to increase GH secretion, and some investigators have obtained better results from arginine infusion after administration of estrogens.

Stimulation tests with other substances, including glucagon, tolbutamide, levodopa, and clonidine, have been proposed. Of these, levodopa stimulation seems to give best results; some believe that it has an accuracy equal to or better than that achieved with insulin or arginine stimulation and, in addition, is easier and safer to perform. One investigator found a greater number of unreliable results in older persons and depressed patients. Blood samples are drawn 60 and 90 minutes after oral levodopa administration. Investigators have explored various means to improve the accuracy of these tests. Sympathetic nervous system alpha-adrenergic stimulants enhance GH release while beta stimulation is inhibitory. The beta blocker propranolol, given with levodopa, enhances levodopa stimulation of GH secretion and improves identification of normal persons. For example, one investigator reported only about 5% false low GH results with the combination of levodopa and propranolol but 20% false low results with levodopa alone. Clonidine stimulation has attracted interest because it has fewer side effects and is more convenient to perform than the other tests. However, there is disagreement regarding test accuracy, partially because some investigators used lower doses that did not provide maximal effect. Overall effectiveness with higher doses is said to be similar to that of the other stimulation tests. The major problems with the stimulation tests are the need for more than one test to confirm a failure to stimulate GH levels and the existence of partial (neurosecretory) GH secretion deficiency, in which GH secretion is not sufficient for normal bone growth but stimulation test results are normal. At present, the diagnosis of partial GH deficiency is made through 12-hour or 24-hour integrated GH measurements or by therapeutic trial with GH. GHRH is now available for investigational use. Preliminary reports using GHRH as a stimulation test in GH deficiency have yielded results rather similar to those of standard stimulatory tests. GHRH testing may be useful in differentiating hypothalamic dysfunction from pituitary GH dysfunction. However, some patients with hypothalamic dysfunction may show apparent failure to respond to initial stimulation of the pituitary by GHRH, thus simulating pituitary GH secretory dysfunction, although they respond normally after repeated stimulation.

Bone X-ray studies. Hand-wrist x-ray studies for bone age (bone maturation) are strongly recommended in patients with suspected growth disorders. The actual genetic sex (which in some instances may be different from the apparent sex) must be furnished to the radiologist so that the correct reference range may be used. Bone maturation should be compared with linear growth (height) and chronologic age. The box on this page lists some conditions associated with retarded or accelerated bone age.