Abstract Background and Aims Soluble αklotho (sklotho) in systemic circulation, which is mainly derived from renal tubule, has been reported to play important roles in mineral bone disorders, cardiovascular complications and renal progression in CKD. Although decreased renal parenchymal mass is thought to be a main cause of decreased serum level of sklotho in CKD, various factors causing renal tubular injury such as hypoxia, oxidative stress, proinflammatory cytokines and uremic toxins are also known to decrease the renal tubular expression of αklotho. Because smoking contributes to hypoxia and oxidative damage, we hypothesized that smoking might be related to decreased sklotho in patients with non-dialysis CKD. Method To test this hypothesis, we analysed data of 2,101 subjects with non-dialysis CKD from a currently on-going Korean multi-centers prospective cohort study of CKD (KNOW-CKD, NCT01630486 at www.clinicaltrials.gov) in whom serum klotho was measured using an enzyme-linked immunosorbent assay kit (Immuno-Biological Laboratories Co, Gunma, Japan). The intra-assay and inter-assay coefficients of variations were 0.57-1.78% and 3.01-6.12% respectively. Results Mean age was 53.6±12.2 years and 1282 (61.0%) were men. The number of current, former and never smokers were 338 (16.1%), 642 (30.6%), and 1118 (53.2%) respectively. The median pack-year was 20.0 (interquartile range: 10.0-35.0). Mean eGFR by CKD-EPICr equation was 53.0±30.7 ml/min/1.73m2 and mean UACR was 903.9±1423.8 mg/g. Median serum sklotho was 536.0 (IQR: 419.5 – 667.0) pg/ml. In multivariable linear regression with backward stepwise adjustment by all significant factors for sklotho in univariate linear regression, smoking pack-year was independent variable for sklotho. (β= -0.786, p=0.032 95% CI for β: -1.504 to -0.069). Age and serum uric acid were also negatively associated with sklotho, while eGFR, serum alkaline phosphatase, and serum hemoglobin were positively associated with sklotho in the same analysis. Median sklotho of never, former, and current smokers were 551.0 (IQR:430 – 689.3) pg/ml, 509.5 (IQR: 407.8 – 645.3) pg/ml, and 521. (IQR: 397.8 – 645.0) pg/ml respectively. Linear mixed model adjusted by sex, age, eGFR, hemoglobin, alkaline phosphatase, and uric acid demonstrated that sklotho in never smokers was significantly higher than former (p=0.004) and current smokers (p=0.001). But there was no difference in sklotho between former and current smokers. Conclusion Smoking was associated with lower serum sklotho in non-dialysis CKD. But no difference in sklotho between former and current smoker casted doubt on causal relationship between smoking and serum sklotho