Diagnosis of GH deficiency during childhood and adolescence is frequently challenging. Children whose height are below the 3rd percentile or -2 SD and have decreased growth velocity require clinical evaluation. Evaluation should begin with a detailed past medical history, family history, diet history, detailed review of prior growth data (including the initial post-natal period) and a thorough physical examination(1). Together, these should help the clinician identify the pattern and cause of growth failure, such as fetal growth restriction (e.g. SGA and IUGR), chronic illness, malnutrition/malabsorption, hypothyroidism, skeletal abnormalities or other identifiable syndromes, such as Turner syndrome. Once growth hormone deficiency is suspected, further testing of the hypothalamic-pituitary axes (including but not limited to the GH-IGF axis) along with radiological evaluation, should be performed (Table 1). It is important to note that the tests cannot be performed simultaneously, or in random order. Certain conditions (e.g. Hypothyroidism and Celiac disease) may mask the presence of others (e.g. GH deficiency), therefore requiring to a step-wise approach with screening tests preceding specific examinations. Since growth failure generally occurs outside of GHD, only those children with signs or symptoms undergo expensive, invasive and nonphysiologic GH provocative testing.
Table 1. Guidelines for initial clinical evaluation of a child with growth failure
|
Evaluation |
Key elements |
|---|---|
|
Birth history |
Gestational age, birth weight and length, delivery type, birth trauma, hypoglycemia, prolonged jaundice. |
|
Past medical and surgical history |
Head trauma, surgery, cranial radiation, CNS infection. |
|
Review of systems |
Appetite, eating habits, bowel movements. |
|
Chronic illness |
Anemia, Inflammatory Bowel Disease, cardiovascular disease, renal insufficiency, etc. |
|
Family history |
Consanguinity, parents and siblings' heights, family history of short stature, delayed puberty. |
|
Physical examination |
Body proportions (upper/lower segment ratios, arm span), head circumference, microphallus, dysmorphism, and midline craniofacial abnormalities. |
|
Growth pattern |
Crossing of percentiles, failure to catch-up. |
|
Screening Tests |
CBC, BCP, ESR, TFT's, UA, IGF1, IGFBP3, Bone age (and a Karyotype for females) |
The growth pattern is a key element of growth assessment and is best studied by plotting growth data on an appropriate growth chart. US growth charts were developed from cross-sectional data provided by the National Center for Health Statistics and updated in 2000(2), with body mass index included in this newest set. The supine length should be plotted for children from birth through age 3 years and standing height plotted when the child is old enough to stand, generally after 2 years of age. Ideally, growth data is determined by evaluating subjects at regular (optimally at 3 month) intervals, with the same stadiometer, and with the same individual obtaining the measurements, whenever possible. Three months is the minimal time interval needed between measurements to calculate a reliable growth velocity, and a six to twelve month interval is optimal. Age and pubertal staging must be considered when evaluating the growth velocity, with the understanding that there is great individual variation in the onset and rate of puberty (87).
Deviations across height percentiles should be noted and evaluated further when confirmed, with the understanding that during the first two years of life, the crossing of length and/or weight percentiles may reflect catch-up or catch-down growth. Crossing percentiles during this period is not always physiological, and must be examined in the context of family, prenatal, birth and medical histories. Additionally, between two and three years of age, statural growth measurement changes from supine to erect, and may also introduce variation. Growth below the normal range (e.g. >-2SD) even without further deviation is consistent with (but not pathognomonic of) GH deficiency. Short stature with a low BMI suggests an abnormality of nutrition/GI tract (e.g. malnutrition, Celiac Disease, etc.), while short stature with an elevated BMI suggests hypothyroidism, Cushing's syndrome or a central eating disorder, such as Prader-Willi syndrome, etc.
Figures 1-3 represent growth charts of children studied by the authors who have genetic defects leading to isolated growth hormone deficiency.
Growth failure can manifest as either gradual deceleration ("falling off the curve") (Figure 2), or alternatively, maintaining a growth pattern parallel to the 3rd percentile, but without catch-up growth (Figure 3).
Most children with GH deficiency have normal birth weight and length. However, in most cases, postnatal growth becomes severely compromised. This can be seen even in the first months of life. Although such children may show a normal growth pattern during the first 6 months, growth failure will eventually occur, as GH takes on a more physiologically dominant role and a child's growth falls below the normal range.
The most commonly used system to assess skeletal maturity is to determine the 'bone age' of the left hand and wrist, using the method of Greulich and Pyle (3). Children younger than 2 years of age should have their bone age estimated from x-rays of the knee. Tanner and Whitehouse and their colleagues developed a scoring system for each of the hand bones as an alternative method to the method of Greulich and Pyle (4).
As growth hormone is secreted in a pulsatile manner (usually 6 pulses in 24 hours and mainly during the night) with little serum GH at any given time, several methods have been recommended to assess the adequacy of GH secretion:
Stimulation testing: GH provocation utilizing arginine, clonidine, glucagon, L-Dopa, insulin, etc. This practice generally measures pituitary reserve-or GH secretory ability-rather than endogenous secretory status. Trained individuals should perform the GH stimulation test according to a standardized protocol, with special care taken with younger children/infants.
GH-dependent biochemical markers: IGF1 and IGFBP3: Values below a cut-off less than -2 SD for IGF1 and/or IGFBP3 strongly suggest an abnormality in the GH axis if other causes of low IGF have been excluded. Age and gender appropriate reference ranges for IGF1 and IGFBP3 are mandatory.
24-hour or Overnight GH sampling: Blood sampling at frequent intervals designed to quantify physiologic bursts of GH secretion .
IGF generation test: This test is used to assess GH action and for the confirmation of suspected GH insensitivity. GH is given for several days (3-5 days) with serum IGF-1 and IGFBP-3 levels measured at the start and end of the test. A sufficient rise in IGF-1 and IGFBP-3 levels would exclude severe forms of GH insensitivity (87,88).
Failure to raise the serum GH level to the threshold level in response to provocation suggests the diagnosis of GH deficiency, while a low IGF1 and/or IGFBP3 level is supportive evidence. Two significant problems with these ancillary variables are that IGF1 levels are very sensitive to the nutritional status (IGFBP3 less so), and also that the normative range for IGF1 and IGFBP3 values are extremely wide, often with poor discrimination between normal and pathological. Age/pubertal stage and gender-specific threshold values must be utilized for both IGF1 and IGFBP3, i.e. GH threshold of 10 ng/ml in children, 5 ng/ml in older adults, etc.
Children with severe GH deficiency can usually be diagnosed easily on clinical grounds, and fail GH stimulation tests. Studies have shown that despite clinical evidence of GH deficiency, some children may pass GH stimulation tests (87). In the case of unexplained short stature, if the child meets most of the following criteria, a trial of GH treatment should be initiated (80):
height >2.25 SD below the mean for age or >2 SD below the midparental height percentile;
growth velocity <25th percentile for bone age:
bone age >2 SD below the mean for age;
low serum insulin-like growth factor 1 (IGF-1) and/or insulin-like growth factor binding protein 3 (IGFBP3);
other clinical features suggestive of GH deficiency.
Height more than 2 SD below the mean.
Neonatal hypoglycemia, microphallus, prolonged jaundice, or traumatic delivery.
Although not required, a peak GH concentration after provocative GH testing of less than 10 ng/ml.
Consanguinity and/or a family member with GH deficiency.
Midline CNS defects, pituitary hypo- or aplasia, pituitary stalk agenesis, empty sella, ectopic posterior pituitary (‘bright spot’) on MRI.
Deficiency of other pituitary hormones: TSH, PRL, LH/FSH and/or ACTH deficiency.
Many practitioners consider GH stimulation tests to be optional in the case of clinical evidence of GH deficiency, in patients with a history of surgery or irradiation of the hypothalamus/pituitary region and growth failure accompanied by additional pituitary hormone deficiencies. Similarly children born SGA, with Turner syndrome, PWS and chronic renal insufficiency do not require GH stimulation testing before initiating GH treatment (80).