Urinary metabolites of dibutyl phthalate and benzophenone-3 are potential chemical risk factors of chronic kidney function markers among healthy women

Abstract

Chronic kidney disease (CKD) is a global health threat of growing concern. Recently, exposure to endocrine disrupting compounds (EDCs) such as phthalates and bisphenol A has been suggested as a risk factor for CKD. However, most epidemiological studies have focused on a limited number of urinary chemicals. This study aimed to identify chemical determinants of the urinary albumin-to-creatinine ratio (ACR), which is a kidney function marker, among multiple major EDCs including phthalate metabolites, bisphenols, and benzophenones in a Korean female population (20-45 years old, n = 441). First, the creatinine-adjusted urinary concentration of each urinary chemical was associated with ACR in a linear regression model (single-pollutant model). Then, compounds with a significant association with ACR in the single-pollutant model were added in a multi-pollutant model and evaluated for their association with ACR. Moreover, to prevent potential reverse causality due to impaired kidney function, quartile analyses were performed for the subjects with healthy renal function (ACR < 9.71 mg/g). In addition to creatinine adjustment, the statistical analysis was also conducted with specific gravity-adjusted concentrations of urinary chemicals, and the results were compared. Several compounds measured in the urine showed a significant association with ACR in the single-pollutant model. In the multi-pollutant model, however, only monobutyl phthalate and benzophenone-1, which are metabolites of dibutyl phthalate and benzophenone-3, respectively, showed significant positive associations. The association of these chemicals remained significant in a couple of the sensitivity analyses with a different adjustment of urine dilution and in a subpopulation with normal ACR. In conclusion, among dozens of urinary chemicals, monobutyl phthalate and benzophenone-1 consistently showed a strong association with urinary ACR. Confirmation of our observation in other human populations and experimental studies is warranted.