The association between estimated glomerular filtration rate and cadmium exposure: An 8-year follow-up study

Background The associations between cadmium exposure and chronic kidney disease have rarely been reported in longitudinal studies. In this study, we investigated the associations between the estimated glomerular filtration rate and cadmium exposure in a cross-sectional study in a longitudinal cohort. Materials and methods In total, 790 subjects (≥35 years of age) living in southeastern China were included at 1998. Cadmium in blood (BCd) and urine (UCd) as well as renal dysfunction biomarkers, urinary N-acetyl-β d-glucosaminidase (UNAG) and albumin (UALB), were determined. 497 subjects were followed at 2006 and a total of 456 subjects were finally included after excluding subjects that did not have exposure or effects biomarkers. The BCd, UCd, UNAG and UALB were determined using baseline methods. At follow-up, the estimated glomerular filtration rate (eGFR) was computed using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation. Single nucleotide polymorphisms (SNPs) in metallothioneins 1A (MT1A) rs11076161 and MT2A rs28366003 in blood sample were detected using TaqMan allelic discrimination assays. Results The median of UCd and BCd at baseline was 5.38 μg/g cr and 4.69 μg/L, respectively, and the median UCd and BCd at follow-up was 4.88 μg/g cr and 2.20 μg/L at follow-up. The mean eGFR at follow-up was 88.0 mL/min/1.73 m2 and 15 subjects had new onset of eGFR <60 mL/min/1.73 m2. The eGFR at follow-up was associated with baseline age (β = −0.66, 95% confidence interval (CI): 0.80 to −0.52), BCd (β = −0.46, 95% CI: 0.68 to −0.25) and UALB (β = −0.29, 95% CI: 0.41 to −0.16) after adjusting for confounders. Subgroup analysis in subjects who had low baseline UALB or subjects with or without hypertension showed similar results. A logistic regression model further showed that baseline BCd and UALB were independent risk factors for follow-up CKD. The odds ratios (ORs) were 1.09 (95% CI:1.03–1.16) for UALB, 1.16 (95% CI:1.01–1.33) for BCd, and 6.74 (0.87–29.63) for current hypertension. Baseline BCd, UALB and current hypertension were used to construct the nomogram. Linear discriminant analysis (LCA) showed that 87.6% of CKD was accurately predicted based on the three factors. Conclusion Baseline age, BCd and UALB were associated with follow-up eGFR, and baseline BCd and UALB were predictive factors for incidence of CKD.