The impact of idiopathic childhood-onset growth hormone deficiency (GHD) on bone mass in subjects without adult GHD

Objective Despite seemingly adequate growth hormone (GH) treatment during childhood, children with GH deficiency (GHD) have reduced bone mineral density (BMD) at final height. The aim was to evaluate BMD and bone mineral content (BMC) in adults treated for idiopathic childhood-onset (CO) GHD, 18 years after stopping GH treatment. Subjects and methods Twenty-six (11 females) patients with idiopathic CO GHD participated. All patients but two had been treated for isolated GHD in childhood. The childhood diagnosis was established by an insulin tolerance test (ITT) and reassessed in adulthood by an ITT (N = 21) or arginine test (n = 5), revealing that 10 patients had GHD according to adult criteria. Accordingly, the patient group was divided into (1) patients who did not have persistent GHD in adulthood and (2) patients who did have persistent adult GHD. Twenty-six healthy subjects acted as age-, gender- and body mass index (BMI)-matched controls. Results The patients who did not have persistent GHD had significantly lower IGF-I values and whole-body, femoral neck and lumbar spine BMD compared to controls [0.994 +/- 0.10 vs. 1.114 +/- 0.11 g/cm(2) (P = 0.003), 0.842 +/- 0.12 vs. 0.962 +/- 0.11 g/cm(2) (P = 0.006) and 1.026 +/- 0.14 vs. 1.127 +/- 0.13 g/cm(2) (P = 0.004), respectively]. Femoral neck BMD was significantly reduced in the patients who had persistent GHD, compared to controls (0.842 +/- 0.09 vs. 0.938 +/- 0.11, P = 0.04). Significant correlations were observed between all bone variables and IGF-I in all subjects, whereas no correlations were observed between bone variables and GH peak levels in the 26 patients. Conclusion In conclusion, we found that (1) patients with idiopathic CO GHD, who at retest in adulthood did not have GHD according to adult criteria, had reduced serum IGF-I and BMD/BMC compared to controls. (2) This observation was also made in the patients who did have persistent GHD in adulthood. The findings may reflect the fact that the present diagnostic criteria for adult GHD (i.e. response to the ITT) do not reflect the clinical consequences of disordered GH-IGF axis in CO GHD young adults who were treated with GH in childhood. Alternatively, despite seemingly adequate GH treatment in childhood an optimal peak bone mass in adolescence may never have been reached in either of the groups. (3) IGF-I levels correlated with clinical signs of the adult GHD syndrome. We believe that further studies on the indications and diagnostic procedures for GH treatment after cessation of linear growth are necessary.