Spot urine sodium and associated estimating equations provide a suitable alternative assessment of 24-hour sodium excretion in many large-scale studies, but not in chronic kidney disease (CKD) patients with decreased renal function. Herein, we aimed to develop a novel predictive equation. We retrospectively enrolled all CKD patients at Stage 1-4 who received spot and 24-hour urinary analysis in our single center from January 1, 2014 to December 31, 2017. Multiple linear regression analysis generated a predictive equation for estimating 24-hour sodium excretion from spot urine samples in the derivation cohort admitted from 2014 to 2015, and then we assessed this predictive equation in a validation cohort admitted from 2016 to 2017. All 5,235 patients were finally analyzed and divided into derivation (n = 2,460) and validation (n = 2,775) cohort according to the admission date. We generated a predictive equation and defined it as "CKDSALT" equation because it was used for the estimation of salt intake in CKD patients. When we measured sodium excretion as the gold standard, we compared this novel validation with other 3 equations: Kawasaki, INTERSALT, and Tanaka. The Bland-Altman plots indicated that the CKDSALT equation showed the lowest bias with limits of agreement (bias = -1.25 mmol, 95% confidence interval -121.3 to 123.8), and the best performance in any subgroup analysis: male and female, old and young, different levels of body mass index, various levels of estimated glomerular filtration rate, and 24-hour urine volume. The CKDSALT equation also had the highest Pearson (0.745) and intraclass correlation coefficient (0.853, 95% confidence interval 0.841-0.863) in all validation cohort and the above subgroups. Spot urine method by CKDSALT equation may be promising for estimating 24-hour urinary sodium excretion in CKD patients with normal renal function and patients with decreased estimated glomerular filtration rate.