Our data show that long-term GH treatment in GH deficient children causes hyperglycaemia and increased insulin secretion. These effects may in some patients induce glucose intolerance, which is reversible with appropriate dietary measures and does not require discontinuation of treatment.

The growth hormone (GH) response to iv administration of GH-releasing hormone (GHRH, 0.3 microgram/Kg) was evaluated in 21 short children (13 boys and 8 girls, age 6.7-13.8 yr). Fourteen had familial short stature and/or constitutional growth delay, one had coeliac disease, and 6 were ultimately diagnosed as non-classical GH deficiency or neurosecretory dysfunction on the basis of subnormal integrated spontaneous GH concentrations (ICGH). The response was compared with 12-h spontaneous GH secretion measured every 30 min from 20:00 to 08:00. Mean ICGH ranged from 2.0-17.7 micrograms/l, with a maximum nocturnal GH peak ranging from 5.4-74 micrograms/l. The maximum GH peak after GHRH ranged from 9.4-50 micrograms/l, and the area under the curve (AUC) from 406-3012 micrograms.min/l. No correlation was found between the maximum GH peak and the AUC after GHRH and the maximum overnight GH peak, the ICGH and the overnight AUC. This study confirms that the GH response to GHRH is highly variable among short children with a wide range of spontaneous GH secretion, and that this response is not correlated with the spontaneous ability to secrete GH.

We evaluated growth hormone binding protein (GHBP) activity in a group of obese children (12 boys and 12 girls, age 3.1-14.7 years, BMI 21.1-33.3, 11 prepubertal and 13 early pubertal) and in 26 age-matched normal weight children (14 boys and 12 girls, age 2.1-16.0 years, BMI 14.2-21.4, 18 prepubertal and 8 early pubertal). All children were of normal stature. GHBP activity was significantly higher in the obese (39.1 +/- 1.1%) than in the control children (28.3 +/- 1.0%, p < 0.0001). Mean serum GHBP was not different between boys and girls or between prepubertal and pubertal subjects. A positive correlation was found between BMI and GHBP levels only in the normal weight children (r = 0.425, p < 0.05). Baseline insulin concentrations in the obese children were 97.6 +/- 7.9 pmol/l (normal values, 45.0 +/- 18.6 pmol/l), and the mean insulin AUC following OGTT in the obese was 811.3 +/- 160.7 pmol/l (normal values, 373.1 +/- 150.1 pmol/l). Serum GHBP activity in the obese was not correlated with baseline serum insulin concentrations or with the insulin AUC following OGTT. In conclusion, we found that obese children have elevated GHBP activity, and speculate that this phenomenon may serve to compensate for their reduced GH secretion and accelerated GH clearance.