Association between mixed dioxin exposure and hyperuricemia in U.S. adults: A comparison of three statistical models

Previous studies on the relationship between dioxin exposures and hyperuricemia have usually been based on multi-chemical linear models. However, the complex nonlinear relationship and interaction between mixed dioxin exposures and hyperuricemia have seldom been studied. In this study, we applied three different statistical models to assess the joint effect of 12 dioxins on hyperuricemia.

A total of 7 dioxin-like polychlorinated biphenyls (DL-PCBs), 3 polychlorinated dibenzo-p-dioxins (PCDDs), and 2 polychlorinated dibenzofurans (PCDFs) were measured in the serum of adults by the National Health and Nutrition Examination Survey (NHANES) from 2003 to 2004. We fitted multivariable logistic regression, weighted quantile sum (WQS) regression, and Bayesian kernel machine regression (BKMR) models to estimate the association of individual and mixed dioxin exposures with hyperuricemia.

Among the 1008 individuals included in our analysis, 20.04% had hyperuricemia. In the multivariable logistic regression established for each single dioxin, PCB28, PCB74, PCB105, PCB118, and 1,2,3,4,6,7,8-HPCDD were positively associated with hyperuricemia. With including all dioxins in the multivariable logistic regression model simultaneously, only PCB28 and 1,2,3,4,6,7,8-HPCDD were positively associated with hyperuricemia. In the WQS regression model, the WQS index was significantly associated (OR (95% CI): 2.32 (1.26, 4.28)) with hyperuricemia, and 1,2,3,4,6,7,8-HPCDD (weighted 0.22) had the largest contribution. In BKMR analysis, a significant positive association was found between mixed dioxin exposure and hyperuricemia when all dioxins were at their 60th percentile or above, compared to their 50th percentile. The univariate exposure-response function showed that PCB105 and PCB118 were positively associated with hyperuricemia.

By comparing the three statistical models, we concluded that the whole-body burden of 12 dioxins was significantly positively associated with hyperuricemia. PCB105, PCB118, and 1,2,3,4,6,7,8-HPCDD played the most important roles in hyperuricemia.