Background: Cholesterol efflux capacity (CEC), a crucial atheroprotective function of high-density lipoprotein (HDL), has proven to be a reliable predictor of cardiovascular risk. Inflammation can damage CEC, but few studies have focused on the relationship between the systemic inflammation marker high-sensitivity C-reactive protein (hsCRP) and CEC in patients with coronary artery disease (CAD). Methods: Thirty-six CAD patients and sixty-one non-CAD controls were enrolled in this observational, cross-sectional study. CEC was measured using a [3H] cholesterol loading Raw 264.7 cell model with apolipoprotein B-depleted plasma (a surrogate for HDL). Proton nuclear magnetic resonance (NMR) spectroscopy was used to assess HDL components and subclass distribution. hsCRP was measured with a latex particle, enhanced immunoturbidimetric assay.Results: CEC was impaired in CAD patients compared to controls (11.9±2.3% vs. 13.0±2.2%, p=0.022). In the control group, CEC was positively correlated with enzymatically measured HDL cholesterol (HDL-C) levels (r=0.358, p=0.006) or NMR-determined HDL-C levels (r=0.416, p=0.001). However, in the CAD group, the significance of correlation disappeared (enzymatic method: r=0.216, p=0.206; NMR spectroscopy: r=0.065, p=0.708). Instead, we found that the level of hsCRP was negatively correlated with CEC (r=-0.351, p=0.036), and this relationship was not modified by CAD risk factors, HDL-C, and HDL subclasses. NMR showed that HDL particles shifted to larger ones in patients with high hsCRP levels, and this phenomenon was accompanied by decreased CEC. Conclusions: In patients with CAD, the level of HDL-C cannot reflect HDL function, but hsCRP is independently associated with HDL dysfunction. The impaired correlation between HDL-C and CEC is possibly due to an inflammation-induced HDL subclass remodeling. Trial registration: Chinese Clinical Trial Registry, ChiCTR1900020873. Registered on 21 January 2019 - Retrospectivelyregistered.