Prepubertal organochlorine pesticide concentrations and age of pubertal onset among Russian boys

Abstract

Background: In animal studies, organochlorine pesticide (OCP) exposure alters pubertal development; however, epidemiological data are limited and inconsistent.

Objective: To evaluate the associations of serum OCP concentrations [hexachlorobenzene (HCB), β-hexachlorocyclohexane (β-HCH), and p,p'-dichlorodiphenyldichloroethylene (p,p'-DDE)] with male pubertal onset.

Methods: In Chapaevsk, Russia, a town environmentally contaminated with OCPs, 350 8-9 year old boys with measured OCPs were enrolled during 2003-2005 and were followed annually for eight years. We evaluated three measures of pubertal onset: testicular volume (TV)>3 mL in either testis, or stage 2 or greater for genitalia (G2+), or pubic hair (P2+). We used multivariable interval-censored models to evaluate associations of OCPs (quartiles) with physician-assessed pubertal onset.

Results: In adjusted models, boys with higher HCB concentrations had later mean ages of TV>3 mL and P2+ (but not G2+). Mean age at attaining TV>3 mL was delayed 3.6 (95% CI: -2.6, 9.7), 7.9 (95% CI: 1.7, 14.0), and 4.7 months (95% CI: -1.4, 10.9) for HCB Q2, Q3, and Q4, respectively, compared to Q1 (trend p: 0.06). Boys with higher HCB concentrations reached P2+ 0.1 months earlier (95% CI: -5.8, 5.6) for Q2, 4.7 months later (95% CI: -1.0, 10.3) for Q3 and 4.6 months later (95% CI: -1.1, 10.3) for Q4 compared to Q1 (trend p: 0.04). There were no associations of serum β-HCH and p,p'-DDE concentrations with age of pubertal onset.

Conclusion: Higher prepubertal serum HCB concentrations were associated with later age of gonadarche and pubarche.

Keywords: HCB; Male puberty; Organochlorine pesticides; p,p′-DDE; β-HCH.

Figure 1. Adjusted Mean Shifts in Age at Pubertal Onset (Months, 95% CIs) by Quartiles of Wet-Weight Serum OCP Concentrations Among 350 Russian Boys^a,b,c

*p ≤ 0.05 **p ≤ 0.10 Shift in months is relative to Q1 (reference) ^aG2+ model adjusted for baseline covariates: boys’ total serum lipids, birth weight, macronutrients (total caloric intake, percent calories from dietary carbohydrates, fat, and protein), blood lead levels; missing birth weight (n=1), macronutrients (n=3) ^bTV > 3 mL model adjusted for baseline covariates: boys’ total serum lipids, birth weight, macronutrients (total caloric intake, percent calories from dietary carbohydrates, fat, and protein), blood lead levels; missing birth weight (n=1), macronutrients (n=3) ^cP2+ model adjusted for baseline covariates: boys’ total serum lipids, macronutrients (total caloric intake, percent calories from dietary carbohydrates, fat, and protein), blood lead levels, maternal age at birth, household income; missing macronutrients (n=3), household income (n=1), maternal age at birth (n=2) ^dHCB wet-weight quartiles (Q1–Q4, pg/g serum): Q1: 169–516; Q2: 517–751; Q3: 752–1,156; Q4: 1,157–15,482 ^eβ-HCH wet-weight quartiles (Q1–Q4, pg/g serum): Q1: 209–567; Q2: 568–814; Q3:815–1,294 Q4: 1,295–13,732 ^fp,p′-DDE wet-weight quartiles (Q1–Q4, pg/g serum): Q1: 261–907; Q2: 908–1,406; Q3: 1,407–2,327; Q4: 2,328–41,301