Abstract： Objective: To analyze the serum levels of small, dense low-density lipoprotein cholesterol (sdLDL-C) in patients with coronary artery disease (CAD) and evaluate their predictive values for the risk of major cardiovascular adverse events (MACE). Methods: Serum sdLDL-C levels were measured in 93 patients with acute coronary syndrome (ACS), 48 patients with stable CAD (SCAD) and 123 healthy controls. The Gensini scores of CAD patients were calculated. The incidence of MACE in the CAD patients was followed up for 1 year. Spearman correlation and multiple linear regression analyses were performed to investigate the association of serum sdLDL-C with Gensini scores in CAD patients. The predictive values of serum sdLDL-C for CAD risk were explored by multivariate Logistic regression analyses and their predictive values for the occurrence of MACE in CAD patients was investigated by Cox regression analyses. Results: The serum sdLDL-C levels in ACS patients were significantly increased compared with SCAD patients (P=0.038) and controls (P＜0.001). The serum sdLDL-C levels were independently correlated with Gensini scores in CAD patients (β=0.315,P=0.017, adjusted R²=0.083). Multi-factor logistic regression analyses revealed that high sdLDL-C levels were closely associated with ACS risk (OR=7.895,95%CI：2.344-26.589, P=0.001) and they could discriminate between ACS and SCAD (OR=5.948,95%CI：1.158-30.558,P=0.033). During 1 year of follow-up, the incidence of MACE in the CAD patients was 22.70%. The serum sdLDL-C level of the CAD patients with MACE was significantly higher than that of CAD patients without MACE (P=0.001). Multi-factor Cox regression analyses indicated that high sdLDL-C levels were closely associated with MACE risk in CAD patients (HR=5.326,95%CI：1.623-17.483,P=0.006). Conclusion: Serum sdLDL-C levels were significantly elevated in ACS patients. Serum sdLDL-C may function as a predictor for assessing the risk of MACE in CAD patients.