Blood lead levels and serum insulin-like growth factor 1 concentrations in peripubertal boys

Abstract

Background: Childhood lead exposure has been associated with growth delay. However, the association between blood lead levels (BLLs) and insulin-like growth factor 1 (IGF-1) has not been characterized in a large cohort with low-level lead exposure.

Methods: We recruited 394 boys 8-9 years of age from an industrial Russian town in 2003-2005 and followed them annually thereafter. We used linear regression models to estimate the association of baseline BLLs with serum IGF-1 concentration at two follow-up visits (ages 10-11 and 12-13 years), adjusting for demographic and socioeconomic covariates.

Results: At study entry, median BLL was 3 μg/dL (range, < 0.5-31 μg/dL), most boys (86%) were prepubertal, and mean ± SD height and BMI z-scores were 0.14 ± 1.0 and -0.2 ± 1.3, respectively. After adjustment for covariates, the mean follow-up IGF-1 concentration was 29.2 ng/mL lower (95% CI: -43.8, -14.5) for boys with high versus low BLL (≥ 5 μg/dL or < 5 μg/dL); this difference persisted after further adjustment for pubertal status. The association of BLL with IGF-1 was stronger for mid-pubertal than prepubertal boys (p = 0.04). Relative to boys with BLLs < 2 μg/dL, adjusted mean IGF-1 concentrations decreased by 12.8 ng/mL (95% CI: -29.9, 4.4) for boys with BLLs of 3-4 μg/dL; 34.5 ng/mL (95% CI: -53.1, -16.0) for BLLs 5-9 μg/dL; and 60.4 ng/mL (95% CI: -90.9, -29.9) for BLLs ≥ 10 μg/dL.

Conclusions: In peripubertal boys with low-level lead exposure, higher BLLs were associated with lower serum IGF-1. Inhibition of the hypothalamic-pituitary-growth axis may be one possible pathway by which lead exposure leads to growth delay.

Figure 1

Adjusted mean IGF-1 concentrations for high versus low BLL by pubertal status. The adjusted mean IGF-1 difference for high versus low BLL was 14 ng/mL in prepubertal boys, 18 ng/mL in early-pubertal boys, and 42 ng/mL in mid-pubertal boys. Compared with prepubertal boys, the lead-associated IGF-1 difference was larger in mid-pubertal boys (p = 0.04) and larger, but not significantly larger, in early-pubertal boys (p = 0.65).

Figure 2

Adjusted mean IGF-1 concentrations by BLL category. Compared with BLL ≤ 2 μg/dL, the higher BLL levels of 5–9 μg/dL and ≥ 10 μg/dL were associated with significantly lower mean IGF-1 concentrations (p < 0.001 for both comparisons).