Associations between dietary exposure to dioxins and polychlorinated biphenyls (PCBs) and Longitudinal changes in weight and waist circumference- an EPIC study.

Abstract

BACKGROUND: Obesity is a growing global health concern. Some evidence suggests that exposure to polychlorinated biphenyls (PCBs) and dioxins, may play a role in weight gain, but human prospective data are limited and have shown inconsistent results. Therefore, this study investigate the association between dietary exposure to dioxins and PCBs and changes in weight and in waist circumference after 5 years of follow-up in a large prospective cohort. METHOD: We included 215,556 participants recruited between 1992 and 2000; of whom 99,046 provided data on waist circumference. Body weight or waist circumference were measured at recruitment and self-reported at follow-up. Intakes of dioxins and PCBs were estimated using country-specific dietary questionnaires collected at baseline, and food contamination concentrations based on a European Food Safety Authority database. Associations were estimated using multilevel mixed linear regression models. RESULTS: Higher intake of both dioxins and dioxin-like PCBs (dioxins + DL-PCBs) (Q4vsQ1 = 0.07kg/5-years (95%CI 0.01, 0.13)), and non-dioxin like PCBs (NDL-PCBs) (Q4vsQ1 = 0.27kg/5-years (95%CI 0.20, 0.35), p-trend<0.001)) were associated with weight gain. Inverse associations were observed between dietary intake of dioxins + DL-PCBs and NDL-PCBs and waist circumference change (Q4vsQ1 = -0.44cm/5-years (95%CI -0.56, -0.31), p-trend<0.001 and Q4vsQ1 = -0.21cm/5-years (95%CI -0.34, -0.07), p-trend<0.001, respectively). These inverse associations were primarily caused by a subset of participants from one country who provided most of the waist circumference data. Results were consistent across stratified and sensitivity analyses. CONCLUSION: Results obtained in this large prospective study show a positive association between dietary intake of both dioxins + DL-PCBs and NDL- PCBs and weight gain. Although the observed associations were small and there may be measurement errors, the consistency of these associations across multiple stratified analyses and sensitivity analyses strengthens the validity of the findings. The findings suggest that the effect of dioxins and PCBs are still present in the food chain despite regulatory bans. Efforts should be strengthened to reduce the exposure levels in the general population not only to lower the risk of obesity, but also to prevent various chronic conditions.