As common contaminants, metals are non-negligible risk factors for diabetes and chronic kidney disease. However, whether there is an association between multiple metals exposure and incident chronic kidney disease (CKD) risk in patients with diabetes is unclear. We conducted a prospective study to evaluate these associations. In total, 3071 diabetics with baseline estimated glomerular filtration rate (eGFR) ≥ 60 mL/min/1.73 m2 from the Dongfeng-Tongji cohort were included. We measured baseline plasma concentrations of 23 metals and investigated the associations between plasma metal concentrations and CKD in diabetics using logistic regression, the least absolute shrinkage and selection operator (LASSO), and the Bayesian Kernel Machine Regression (BKMR) models. During average 4.6 years of follow-up, 457 diabetics developed CKD (14.9 %). The three models consistently found plasma levels of zinc, arsenic, and rubidium had a positive association with incident CKD risk in patients with diabetes, while titanium, cadmium, and lead had an inverse correlation. The results of BKMR showed a significant and positive overall effect of 23 metals on the risk of CKD, when all of the metals were above the 50th percentile as compared to the median value. In addition, potential interactions of zinc and arsenic, zinc and cadmium, zinc and lead, titanium and arsenic, and cadmium and lead on CKD risk were observed. In summary, we found significant associations of plasma titanium, zinc, arsenic, rubidium, cadmium, and lead with CKD in diabetes and interactions between these metals except for rubidium. Co-exposure to multiple metals was associated with increased CKD risk in diabetics.
Keywords: Bayesian kernel machine regression (BKMR); Chronic kidney disease; Diabetes; LASSO; Metal mixtures; Prospective study.
Copyright © 2022 The Authors. Published by Elsevier Inc. All rights reserved.