Abstract Background The relationship between non-fasting remnant cholesterol and cardiovascular outcome in the era of potent statin therapy remained to be elucidated. Methods A cohort study of three hundred and twenty eight diabetics diagnosed with new-onset stable coronary artery disease (CAD) by coronary angiography were enrolled. All cases were followed up for an average duration of twelve months. The association between baseline remnant cholesterol levels and major cardiovascular outcomes were evaluated using the receivers operating characteristic (ROC) curves and Cox proportional hazards regression analysis. Results During a period of 12-month’s follow-up, 14.3% patients (47/328) underwent pre-specified adverse outcomes. The remnant cholesterol associated with high sensitivity C-reactive protein, neutrophil count and fibrinogen ( R 2 = 0.20, 0.12 and 0.14; P = 0.000, 0.036 and 0.010 respectively). Area under the ROC curves (AUC) indicated discriminatory power of the remnant cholesterol to predict the adverse outcomes for this population (AUC = 0.64, P