Abstract
BACKGROUND: Chronic kidney disease (CKD) is diagnosed in more than 26 million U.S. people, which increases the risk of many adverse events. α-Klotho was reported to have potential effects on kidney function. The purpose of this study was to investigated whether CKD prevalence is associated with α-Klotho levels in the U.S. people aged 40-79 years. METHODS: Thirteen thousand five hundred eighty-nine participates in the National Health and Nutrition Examination Survey 2007-2016 aged 40-79 with information of Klotho and kidney function were included. The association between CKD and Klotho was calculated using multivariate linear or logistic regression models with adjustment of several possibly confounding variables. Subgroup analyses stratified by age, BMI, and diabetes mellitus were conducted. The non-linear relationship between Klotho and dependent variables with a non-normality of residues was assessed using smooth curve fitting and the segmented regression (also known as piece-wise regression) models. RESULTS: Among 13,589 participants, the median of Klotho levels was 803.10 pg/mL, mean eGFR of all participants was 86.96 (SD = 19.88) mL/min/1.73 m(2), and CKD was diagnosed in 20.11% of them (N = 2733). In the fully adjusted model, eGFR was positively associated with Klotho (β = 5.14, 95%CI 4.13-6.15, p < 0.001), while CKD was negatively associated with Klotho (stage ≧ 1, OR = 0.62, 95% CI 0.50-0.76, p < 0.001; stage ≧ 3, OR = 0.31, 95% CI 0.24-0.41, p < 0.001). The non-linear relationship showed that occurrence of CKD stage> 1 and albuminuria were negatively associated with Klotho when Klotho smaller than turning point (for whether CKD stage> 1, turning point K = 6.85, Klotho < K, OR = 0.44, p < 0.001; for albuminuria, turning point K = 6.84, Klotho < K, OR = 0.59, p < 0.001). CONCLUSION: Serum soluble Klotho levels were positively associated with eGFR and negatively associated with the prevalence of CKD, especially in elderly, obese, and diabetic patients.