The association of aldehydes exposure with diabetes mellitus in US population: NHANES 2013-2014.

Background The association of mixed aldehydes exposure with diabetes remains unclear. Objective We aimed to explore associations between serum aldehydes concentration and diabetes. Methods We analyzed associations between aldehydes and diabetes using data from 1795 participants in the National Health and Nutrition Examination Survey (NHANES) from 2013 to 2014 by multiple logistic regression models. Bayesian kernel machine regression (BKMR) was used to evaluate the combined association of serum aldehydes on prediabetes and diabetes. Results Isopentanaldehyde increased the risk of diabetes 2.09 fold (95%CI:1.05-4.16) in the highest tertile, compared to the lowest-tertile concentration after adjusting for covariates, with a p-value for trend (P-t) equal to 0.041, in females. The adjusted OR of prediabetes with a 95% CI for the highest tertile was 0.52(0.28, 0.97) for benzaldehyde in females (P-t = 0.034). We also found associations in the male group between butyraldehyde and diabetes for the second (OR:2.80, 95%CI:1.35-5.79) and third (OR:2.59, 95%CI:1.30-5.17) tertile levels (P-t = 0.010). The risk of diabetes increased 2.55 fold (95%CI: 1.26-5.16, P-t = 0.008), in subjects in the highest tertile of hexanaldehyde concentration. Other aldehydes did not show a statistically significant association with diabetes or prediabetes. The BKMR model showed a positive association of mixed aldehydes with diabetes in males, and butyraldehyde showed a significant positive trend with the highest posterior inclusion probability (PIP = 0.85). Mixed aldehydes increased female's risk from prediabetes to diabetes in which isopentanaldehyde had the highest posterior inclusion probability (PIP = 0.67). Conclusions The mixed aldehydes might increase the risk of suffering from diabetes in males and accelerate the progression of diabetes in females, in which butyraldehyde and isopentanaldehyde play the most important roles.