ABSTRACT Objective Early establishment of an ideal risk assessment model is the key point of the prevention and treatment of chronic kidney disease (CKD). There are advantages of ultrasound parameters which is rapid, repeatable and easy to obtain. This study aims to explore the value of ultrasound parameters in predicting the degree of renal damage in CKD patients, and to improve the screening of CKD high-risk population by conventional renal color Doppler ultrasound. Methods 252 CKD in-patients were selected and divided into mild CKD reduction group (G1-2 stage, 83 cases) and moderate/severe CKD reduction group (G3-5 stage, 169 cases). Clinical data and laboratory tests [including estimated glomerular filtration rate (eGFR) and serum creatinine (Cr)] were collected. Renal length, parenchymal thickness and parenchymal echo were measured by two-dimensional ultrasound, and intrarenal arterial resistance index (RI) was measured by Doppler ultrasound. The differences of above parameters between the two groups were compared. Logistic regression was used to analyze the predictive ability of ultrasound parameters for mild and severe CKD damage. The receiver operating characteristic (ROC) curve was drawn to analyze the diagnostic efficacy of ultrasound parameters for mild and severe impaired CKD. Results There were significant differences in age, hypertension and coronary heart disease between the two groups (P < 0.05). There were significant differences in eGFR, urea, creatinine and all ultrasound parameters between the two groups (P < 0.0001). ROC curve analysis of renal parenchyma echo intensity and ultrasound CKD score had a high AUC (0.838 vs 0.820), and the cut-off values for predicting mild CKD damage were 1.75 and 4.75. The cut-off values for predicting severe CKD damage were 2.25 and 5.25, respectively. Multivariate logistic regression analysis showed that the AUC of combined ultrasound parameters in predicting mild and severe CKD were 0.867 and 0.847, respectively (both P < 0.0001). Conclusion Ultrasound parameters can independently predict mild renal function impairment in CKD patients, among which ultrasound CKD score and renal parenchyma echo index have relatively higher values. Combined with renal artery RI parameters, the prediction efficiency can be improved.