Prenatal and childhood exposure to chlordecone and adiposity of seven-year-old children in the Timoun mother-child cohort study in Guadeloupe (French West Indies).

backgroundExposure to persistent environmental organic pollutants may contribute to the development of obesity among children. Chlordecone is a persistent organochlorine insecticide with estrogenic properties that was used in the French West Indies (1973-1993) and is still present in the soil and the water and food consumed by the local population. We studied the association between prenatal and childhood exposure to chlordecone and the adiposity of prepubertal children.

methodsWithin the Timoun Mother-Child Cohort Study in Guadeloupe (French West Indies), 575 children had a medical examination at seven years of age, including adiposity measurements. A Structural Equation Modeling approach was used to create a global adiposity score from four adiposity indicators: the BMI z-score, percentage of fat mass, sum of the tricipital and subscapular skinfold thickness, and waist-to-height ratio. Chlordecone concentrations were measured in cord blood at birth and in the children's blood at seven years of age. Models were adjusted for prenatal and postnatal covariates. Sensitivity analyses accounted for co-exposure to PCB-153 and pp'-DDE. Mediation analyses, including intermediate birth outcomes, were conducted.

resultsPrenatal chlordecone exposure tended to be associated with increased adiposity at seven years of age, particularly in boys. However, statistical significance was only reached in the third quartile of exposure and neither linear nor non-linear trends could be formally identified. Consideration of preterm birth or birth weight in mediation analyses did not modify the results, as adjustment for PCB-153 and pp'-DDE co-exposures.

conclusionGlobally, we found little evidence of an association between chlordecone exposure during the critical in utero or childhood periods of development and altered body-weight homeostasis in childhood. Nevertheless, some associations we observed at seven years of age, although non-significant, were consistent with those observed at earlier ages and would be worth investing during further follow-ups of children of the Timoun Mother-Child Cohort Study when they reach puberty.