Context: Supraphysiological serum insulin-like growth-factor-I (IGF-I) concentrations have been a matter of concern in children treated with growth hormone (GH) because high IGF-I levels were associated with risk of later disease in former epidemiological studies.

Objective: To determine whether a single IGF-I measurement reliably reflects lifetime IGF-I exposure we evaluated intraindividual longitudinal tracking of IGF-I and IGF-binding-protein-3 (IGFBP-3) levels and we estimated cumulative lifetime exposure to IGF-I in healthy and GH-treated individuals.

Methods: We included 6,459 healthy participants (cross-sectional = 5,326; longitudinal = 1,133) aged 0----76 years (9,963 serum samples) and nine patients born small-for-gestational-age (SGA) with 238 serum samples during GH treatment.Intraindividual tracking of IGF-I and IGFBP-3 (SDS) was determined by intraclass correlation coefficients (ICC). Cumulative lifetime IGF-I exposure was estimated by area under the curve of the predicted SDS-trajectory from 0-76 years.

Results: For IGF-I (SDS), ICCs were 0.50 (95% CI: 0.47-0.53) for male and 0.53 (0.50-0.56) for female participants. Lifetime IGF-I exposure was significantly higher in female (mean 12,723 ± 3,691 SD) than in male participants (12,563 ± 3,393); p = 0.02. In SGA children, treatment with GH increased the lifetime exposure to IGF-I from 9,512 ± 1,889 to 11,271 ± 1,689, corresponding to an increase in lifetime IGF-I trajectory from -0.89 SD ± 0.57 to -0.35 SD ± 0.49.

Conclusion: Since IGF-I and IGFBP-3 levels track throughout life, a single measurement reliably reflects lifetime exposure. GH therapy increased the lifetime exposure to IGF-I only slightly and it remained below the average lifetime exposure in the reference population.

Keywords: GH treatment; IGF-I; IGFBP-3; SGA; short stature