Association between polychlorinated biphenyls exposure and incident type 2 diabetes mellitus: A nested case-control study

Previous epidemiological studies indicated that the association between polychlorinated biphenyls (PCB) and type 2 diabetes mellitus (T2DM) was inconclusive. We investigated the association between PCBs exposure and incident T2DM in a nested case-control study, and further explored the relationship between PCBs and 5-year fasting blood glucose (FBG) changes. Baseline concentrations of seven indicator-PCB (PCB-28, 52, 101, 118, 138, 153, 180) were measured in 1006 pairs of incident T2DM cases and matched controls nested within the Dongfeng-Tongji cohort. Conditional logistic regression models and pre-adjusted residuals method were used to assess the associations between PCBs and incident T2DM. We further computed beta coefficients (βs) of 5-year FBG changes using multivariable generalized linear regression. Non-dioxin-like PCBs (NDL-PCBs) were significantly associated with higher T2DM incidence after adjustment for all covariates. Significant differences were observed for extreme quartiles comparisons (Q4 vs. Q1) of PCBs except PCB-138, and the incidence of T2DM were 1- to 3-fold higher among those in the highest versus lowest PCBs quartiles. Serum NDL-PCBs were positively associated with changes in FBG (P for overall association ≤0.01). Additionally, triglycerides mediated the associations between PCBs and T2DM incidence. Our findings showed positive associations of NDL-PCBs with incident T2DM and 5-year FBG changes. PCBs increased incident T2DM via lipid metabolic pathways.