The aim of this study was to assess the association between CHA2DS2-VASc score and carotid plaques in patients with non-valvular atrial fibrillation (NVAF). We conducted a retrospective study including 3,435 NVAF patients who underwent carotid ultrasound examinations from January 2015 to December 2017.We collected the clinical data on the medical records system. Chi-square trend test was used to analyze trends between the prevalence of carotid plaques with an increasing CHA2DS2-VASc score. Univariate and multivariate logistic regression was also used to assess the association between carotid plaques and CHA2DS2-VASc scores. The area under the receiver operating characteristic (ROC) curve (AUC) was used to determine the optimal cutoff points of different CHA2DS2-VASc scores in NVAF patients. NVAF patients with carotid plaques had higher CHA2DS2-VASc scores compared with patients who did not have carotid plaques (3.01±1.36 vs. 2.55±1.28, P < 0.05). In all participants, male participants and female participants, the prevalence of carotid plaques increased significantly as the CHA2DS2-VASc score increased (P for trend < 0.001). Multivariate logistic regression analysis demonstrated that for each 1-point increase in the CHA2DS2-VASc score, there was an associated 37% increase in the prevalence of carotid plaques. ROC curve analysis revealed that a CHA2DS2-VASc score ≥ 2 in male patients (sensitivity, 44.67%; specificity, 75.64%; AUC, 0.639) or ≥ 3 in female patients (sensitivity, 47.24%; specificity, 72.40%; AUC, 0.634) were associated with carotid plaques. The prevalence of carotid plaques in patients with NVAF was associated with the CHA2DS2-VASc score.