Comparison between insulin-like growth factor-I (IGF-I) and IGF binding protein-3 (IGFBP-3) measurement in the diagnosis of growth hormone deficiency.

To analyze the utility of insulin-like growth factor-I (IGF-I) and insulin-like growth factor binding protein-3 (IGFBP-3) radioimmunoassay in the diagnosis of growth hormone deficiency (GHD) we measured IGF-I and IGFBP-3 in sera from normal children (n=309), short children (n=99) and patients with GHD (n=73). In 80% and 93% of classical GHD (cGHD), IGF-I and IGFBP-3 levels, respectively, were below the age-related cutoff levels (lower limit). In 81% and 88% of normal short children (NS), IGF-I and IGFBP-3 levels, respectively, were above the age-related cutoff levels. Thus, both IGF-I and IGFBP-3 were good parameters for screening GHD. In contrast, in more than half of partial GHD (pGHD), either IGF-I or IGFBP-3 was above the age-related cutoff levels. The poor discrimination between patients with pGHD and NS by using these two parameters may be the result of their relatively similar GH levels, as compared to cGHD, or due to the limitations of GH stimulation tests. In about 80-90% of NS, IGF-I and IGFBP-3 were above the age-related cutoff levels at all ages. A hundred percent of cGHD under 10 years old had IGFBP-3 below the age-related cutoff levels, whereas 79% of cGHD under 10 years old had IGF-I below the age-related cutoff levels. Thus in the younger age groups, IGFBP-3 may be more sensitive than IGF-I. It may be because IGFBP-3 levels are relatively higher than those of IGF-I in younger subjects. IGFBP-3 may be less sensitive for diagnosing GHD in older children than in younger children because IGFBP-3 levels may also increase during puberty by mechanisms independent of the GH-IGF-I axis. There was a significant correlation between IGF-I and IGFBP-3 in all the subjects. However, IGF-I and IGFBP-3 classified subjects differently in 25% of patients with GHD and 19% of those with NS. This may reflect differences in daily coefficient of variation in IGF-I and IGFBP-3, in assay sensitivity and in non-GH dependent pubertal effect. The other explanation for the difference between these two parameters in terms of above and below the cutoff levels is that it may be due to the limitation of GH stimulation tests in the diagnosis of GHD.