Objective: Both IGF-I and IGFBP-3 reflect spontaneous GH secretion in healthy individuals. We have evaluated the clinical usefulness of urinary IGF-I and IGFBP-3 measurements in the diagnosis of children with disordered growth.
Design: Serum IGF-I and IGFBP-3 radioimmunoassays (RIA) were developed, and modified for quantitation in urine. The relationship between serum and urine levels, and the performance of these tests in the diagnosis of GH deficiency (GHD) were examined.
Patients: Sixty-nine children (age 9.5 +/- 3.6 years; 37 boys, 32 girls) provided a timed overnight urine collection and a serum sample collected on the same morning. Subjects were defined as GHD (n = 22) or short normal (SN; n = 47) on the basis of medical history, clinical examination, auxology and peak response to a GH stimulation test (< 20 mU/L in GHD patients).
Measurements: IGF-I and IGFBP-3 in serum and urine were measured by RIA, urinary GH (uGH) by immunoradiometric assay (IRMA) after dialysis and urinary creatinine by the alkaline picrate method. Urine results were expressed as total amount excreted (tulGFBP-3 (microgram), tulGF-1 (ng), tuGH (ng), tuCrt (mmol).
Results: Urine IGF-I and IGFBP-3 excretion correlated significantly to serum levels of IGF-I and IGFBP-3 and also to tuGH excretion. There was a strong positive relationship between both urinary peptides and tuCrt, which suggested that renal filtration was the source of these peptides in urine. In addition, there were significant correlations with age, bone age and height SD score, of similar magnitude to those for tuGH. In prepubertal children, serum IGF-I and IGFBP-3 were significantly lower in GHD compared with SN children, while in puberty only serum IGFBP-3 was significantly lower in GHD. There was no difference however, in tulGF-I or tulGFBP-3 between GHD and SN children either prepubertally or in puberty with near complete overlap of the values between groups.
Conclusions: Measurements of tulGF-I and tulGFBP-3 have no place in the diagnosis of childhood GHD. Nonetheless, the significant correlations between serum and urinary IGF-I and IGFBP-3 levels and their correlation to uGH indicate that these peptides could be used as non-invasive physiological markers of the GH-IGF axis.