Combined Exposure to Multiple Metals and Kidney Function in a Midlife and Elderly Population in China: A Prospective Cohort Study

Abstract

[Background] Metal exposure is suspected to be correlated to kidney function. However, the combined effects of co-exposing to multiple metals, especially both toxic and protective metals, have not been completely evaluated. [Method] A prospective cohort study was conducted with the "135" cohort for the evaluation of how plasma metal levels are correlated to kidney function in a midlife and elderly community in southern China. An amount of 1368 subjects without kidney disease at baseline were enrolled in the final analysis. By using linear regression and logistic regression models, the correlation of individual metal values with renal function parameters was assessed. Measuring of the multiple metal exposure level was performed by principal component analysis (PCA). [Results] Diminished renal function, as evaluated based on fast kidney function decline, or estimated glomerular filtration rate (eGFR) < 60 mL/min/1.73 m², was positively associated with the plasma concentrations of chromium and potassium, but it was negatively associated with selenium and iron (p < 0.05). In multiple-metal analyses, linear and logistic regression models showed that the iron and chromium exposure pattern had a protective effect on renal function, whereas the sodium and potassium exposure pattern and the cadmium and lead exposure pattern increased the risk for fast kidney function decline, and eGFR < 60 mL/min/1.73 m². [Conclusions] Certain metals, including chromium, potassium, selenium, and iron, were correlated with kidney function in a midlife and elderly community in China. In addition, the potential combined influences of co-exposing to multiple metals were observed.

Keywords: chromium; environmental epidemiology; plasma metal; potassium; renal function; selenium.

Figure 1

Difference (95%CI) in a yearly decline of eGFR (mL/min/1.73 m²) correlated to baseline plasma metal levels in the linear regression model. CI: confidence interval. Model 1:crude model; Model 2: adjusted for age, gender, baseline eGFR, BMI, smoking status, and drinking status. The inclusion of plasma metal levels in models as natural log-transformed continuous variables. p values were Bonferroni corrected. Red line: Bonferroni-corrected positive significance Blue line: Bonferroni-corrected negative significance.