Lead exposure, childhood weight status and sexual maturation. Prenatal lead exposure has been associated with smaller size at birth and in infancy [22, 124,135-138]. Among children exposed to high lead level in utero, growth delays appear to persist only among those with high postnatal lead exposure [89, 90], pointing to the importance of repeat measures of exposure to understand genesis of growth delays associated with lead levels in children. Cross-sectional analyses of survey data and observational studies document a 1-2 cm decrement in stature for every 10 pg/dL difference in blood levels in nationally representative samples from NHANES and HHANES, but findings are less consistent for the association with weight [139-141]. Variance in childhood weight reflects both stature and the relationship of weight with stature [142]. Only two studies have investigated the relationship between lead and body habitus in children. Mid-pregnancy blood lead was marginally associated with BMI change from birth to 1 yr (|3=0.04, 95% CI?0.06 to 0.14) and 1-4 yr (P=0.04, 95%C!=0.00 to 0.08) among 100 and 58 mother-infant pairs, respectively, in Kosovo [143], Among Boston school children followed by Dr. Howard Hu, Co-lnvestigator for the proposed P20, a 10-fold increase in (logio) dentin lead level at 7 yr of age was cross-sectionally associated with a 1.023 increment in BMI among 236 children (P=1.023, SE=0.458, P=0.03). Among 58 ofthese subjects followed-up 10 years later, a 10-fold increase in dentin lead at 7 yr of age was associated with a 2.65 unit increase in BMI (P=2.650, SE= 1.156, P=0.03) [144]. Although dentin lead is a measure of chronic lead exposure, this study did not examine prenatal or early childhood lead exposure, limiting inferences about timing of exposure. In the Kosovo study, child blood lead levels were measured but not adjusted for analyses due to high correlation with maternal blood lead. A recent animal study examined the long term effects of gestational lead exposure (GLE) in a murine model of human equivalent GLE [1], finding inverse dose-response effects on late onset obesity of male but not female offspring. GLE did not significantly affect weight of female offspring and was significantly associated with weight of male mice at 1 yr but not at eariier ages. Compared with controls, male mice weighed significantly more: +26% low, +21% moderate, +13% high GLE. These findings point to the need to consider associations of perinatal lead exposure with analytic approaches that account for nonmonotonic responses in physical growth across spectrum of sensitive periods in childhood. In addition, use of outcome measure of body habitus (e.g. BMI) may reveal whether effects on weight of mice with high-dose GLE were in fact also stunted. Molecular mechanisms for lead effect on obesity are not well understood [1], but developmental studies suggest effects could be related to altered hypothalamic pituitary axis (HPA) and hypothalamic dopaminergic dysfunction [91], to genetic polymorphisms related to obesity and metabolic disorders, or diet or environmental exposures that disrupt endocrine signaling [92, 93]. Cross-sectional studies of national survey data (NHANES III) implicate low-level lead exposure in delayed age at self-reported menarche and physician-observed onset of puberty (Tanner stage 2, pubic hair development) in giris, after adjustment for race/ethnicity, BMI, age, family size, urban residence and poverty income ratio [145]. In analyses among giris 8-18 yr of age in NHANES III stratified by race/ethnicity, higher (3pg/dL) compared with lower (Ipg/dL) blood lead concentrations, were associated with significant delays in Tanner stages for both breast development (Adjusted OR 0.76;95% Confidence Interval (Cl) 0.63-0.91 and pubic hair development (OR 0.70 (0.54-0.91) in Mexican-American girls and in all pubertal measures (breast, pubic hair, age at menarche), among non-Hispanic black giris. In white giris, blood lead concentration was associated with delays in pubertal development but these findings were not statistically significant [87]. In a study of 489 Russian boys aged 8-9 yr, those with blood lead levels >5pg/dL had a 44% reduced odds of having entered Tanner genital staging G2 (95%CI= 0.34 to 0.95, P=0.03) and demonstrated marginally reduced testicular volume (OR=0.72, 95%CI=0.48 to 1.07, P=0.11) [88]. Cross-sectional designs nevertheless limit inferences from these studies about the timing of lead exposure and underscore the need to examine relationships of perinatal exposure to growth and maturation in longitudinal data. Studies among rodents suggest lead may have a dual site of action: at the level of hypothalamic pituitary axis (HPA), and directly at the level of gonadal steroid biosynthesis, although these mechanisms may not operate similariy among humans. Among animals, lead is believed to act on the Hypothalamic-Pituitary-Gonadal (HPG) axis by blocking the release of gonadotropin releasing hormone (GnRH) thus decreasing puberty-related hormones such as LH and estradiol. Prenatal exposure to lead among rats has been demonstrated to cause a decrease in several puberty-related hormones such as insulin-like growth factor-1 (IGF-1) [146], luteinizing hormone (LH), and estradiol (E2)J30, 32,147]. The effect of lead on puberty-related hormones has been demonstrated to be reversible by the administration of IGF-1 to prepubertal female rats [148]. At the gonadal steroid biosynthesis level (second site of action), lead has been shown to impair leydigcell and Sertoli cell functions [149, 150]. Lead has been associated with a decrease in testicular weight, seminiferous tubular diameter, and sperm count among mice subcutaneously exposed to lead or exposed to lead though their diet [32, 150-152]. Prenatal and dietary lead exposure has also been observed to delay the timing of puberty (as measured by age of vaginal opening and age of estrus for example) among female rats and mice [30, 32, 153-155]. In rat studies, lead has been related to a reduction in testosterone and estradiol levels during puberty [146, 150, 156].
Showing the most recent 10 out of 57 publications
