Unraveling the pesticide-diabetes connection: A case-cohort study integrating Mendelian randomization analysis with a...

There is no evidence on the longitudinal and causal associations between multiple pesticides and the incidence of type 2 diabetes mellitus (T2DM) in the Chinese rural population, and whether physical activity (PA) modified these associations remains unclear. Here, we aimed to investigate the longitudinal and causal associations between pesticides mixture and T2DM, and determine whether PA modified these associations.

A total of 925 subjects with normal glucose and 925 subjects with impaired fasting glucose (IFG) were enrolled in this case-cohort study. A total of 51 targeted pesticides were quantified at baseline. Logistic regression, quantile g-computation, and Bayesian kernel machine regression (BKMR) were used to assess the individual and combined effects of pesticides on IFG and T2DM. Mendelian randomization (MR) analysis was employed to obtain the causal association between pesticides and T2DM.

After 3-year follow-up, one-unit increment in ln-isofenphos, ln-malathion, and ln-deltamethrin were associated with an increase conversion of IFG to T2DM (FDR-P<0.05). One quartile increment in organochlorine pesticides (OCPs), organophosphorus pesticides (OPs), herbicides and pyrethroids mixtures were related to a higher incidence of T2DM among IFG patients (P<0.05). The BKMR results showed a positive trend between exposure to pesticides mixture and T2DM. The MR analysis indicated a positive association between exposure to pesticides and T2DM risk (P<0.05). No any significant association was found between pesticides and IFG. In addition, compared to subjects with high levels of PA, those with low levels of PA were related to increased risk of T2DM with the increased levels of pesticides among IFG patients.

Individual and combined exposure to pesticides increased the incidence of T2DM among IFG patients. MR analysis further supported the causal association of pesticides exposure with T2DM risk. Our study furtherly indicated that high levels of PA attenuated the diabetogenic effect of pesticides exposure.