Growth hormone (GH) provocative retesting of 108 young adults with childhood-onset GH deficiency and the diagnostic value of insulin-like growth factor I (IGF-I) and IGF-binding protein-3

Serum levels of total insulin-like growth factor I (IGF-I) and IGF-binding protein-3 (IGFBP-3) reflect the endogenous GH secretion in healthy children and exhibit little diurnal variation, which makes them good diagnostic markers for screening of GH deficiency (GHD) in short children, although some controversy still exists. In adults, the diagnostic value of IGF-I and IGFBP-3 suspected of GHD has been reported in only a few studies. We performed a GH provocative test, using oral clonidine, in 108 patients who had previously been treated with GH during childhood (73 men and 35 women). Basal IGF-I and IGFBP-3 levels were compared to those in 1237 healthy controls (312 controls >18 yr) as well as to peak GH levels. Seventy-nine patients had peak GH values below a cut-off value of 7.5 mu g/L (34 with isolated GHD), whereas 29 patients had a normal GH response (28 with previous isolated GHD), i.e. 45% of patients treated with GH during childhood because of isolated GHD had a normal GH response when retested in adulthood. Multiple regression analysis revealed that peak GH levels were dependent on the degree of hypopituitarism, body mass index, and duration of disease. IGF-I levels were below -2 so in 60 of 79 GHD patients and above -2 so in 21 of 29 patients with a normal GH response. IGFBP-3 levels were below -2 SD in 54 of 79 GHD patients and above -2 sn in 23 of 29 patients with a normal GH response. Multiple linear regression analysis demonstrated that IGF-I and IGFBP-3 were significantly dependent on peak GH levels and the number of other pituitary axes affected. In this analysis, duration of disease was significantly associated with both IGF-I and IGFBP-3, whereas body mass index was significantly associated with IGFBP-3, but not with IGF-I. We conclude that IGF-I and IGFBP-3 determinations predict the outcome of a GH provocative test in adults suspected of GHD and believe that IGF-I as well as IGFBP-3 serum concentrations are valuable diagnostic parameters in the evaluation of GHD in adults with childhood-onset disease. We suggest that children who have been treated with GH should undergo reassessment of their GH secretory status as young adults by provocative testing as well as by IGF-related peptides before continued adult GH replacement therapy is considered. However, our data suggest that it is not necessary to reconfirm GH deficiency by GH provocative testing in young adults who have two or more pituitary hormone deficiencies in addition to GHD.