Relationship of serum bisphenol A with diabetes in the Thai population, National Health Examination Survey IV, 2009.

BACKGROUND

Epidemiological studies of the association between bisphenol A (BPA) exposure and diabetes have been inconsistent. The present study determined serum BPA concentrations in the Thai population and their association with hyperglycemia and diabetes.

METHODS

In all, 2581 serum samples from the Thai National Health Examination Survey (2009) were used to determine BPA levels. Impaired fasting glucose (IFG) was defined as fasting plasma glucose ≥100 and <126 mg/dL. Diabetes was defined as a history of a physician's diagnosis or fasting plasma glucose ≥126 mg/dL. Multinomial logistic regression was used to examine the association of serum BPA with IFG and diabetes.

RESULTS

Of 2581 samples tested, BPA was detected in 2135 samples (weighted percentage 78.1%), with a geometric mean concentration of 0.34 ng/mL BPA. Serum BPA levels were significantly higher among those with diabetes or IFG compared with normoglycemic individuals (0.52 and 0.38 vs 0.33 ng/mL, respectively; P < 0.001). After adjusting for potential confounders, compared with the first quartile (≤0.071 ng/mL), the overall adjusted odds ratios (OR) of serum BPA concentration in the third and fourth quartiles (0.319-0.745 and ≥0.746 ng/mL, respectively) for IFG were 1.72 (95% confidence interval [CI] 1.19, 2.49) and 1.23 (95% CI 0.80, 1.89), respectively; for diabetes, the adjusted OR were 1.88 (95% CI 1.18, 2.99) and 1.83 (95% CI 1.12, 2.95), respectively, with a slightly stronger association among men than in women.

CONCLUSIONS

Serum BPA concentrations were not associated with IFG, but were positively associated with diabetes in the Thai population. Further prospective studies are needed to confirm the relationship.