Background and Purpose – Plaque composition, but not stenosis degree, may play a key role in the development of recurrent ischemic events in patients with symptomatic, mild (<50%) carotid stenosis. This multi-center prospective cohort study was aimed to determine their clinical and radiological features and to evaluate the benefits of carotid endarterectomy (CEA) for them. Methods ? This study included 124 patients with cerebrovascular or retinal ischemic events ipsilateral to mild carotid stenosis. Best medical therapy (BMT) was administered to all participants. CEA or carotid artery stenting (CAS) was implemented at each institution's discretion. Baseline and 6-, 12-, and 24-month follow-up data were collected. Primary endpoint was ipsilateral ischemic stroke. Secondary endpoints included any stroke, ipsilateral TIA, ipsilateral ocular symptoms, any death, and plaque progression requiring CEA/CAS. Multivariate Cox proportional hazard model was used to evaluate the predictors for each endpoint. Results – Of 124 patients, 70 patients (56.5%) had the history of ipsilateral ischemic stroke and 51 (43.5%) had been treated with antiplatelet agents. Mean stenosis degree was 22.4±13.7%. Plaque composition was categorized into fibrous plaque in 22 patients, lipid-rich/necrotic core (LR/NC) in 25, and intraplaque hemorrhage (IPH) in 69. BMT was indicated for 59 patients, while CEA was performed in 63. The incidence of primary endpoint was significantly higher in BMT group than in CEA group (15.1% vs. 1.7%; HR, 0.18; 95%CI, 0.05–0.84, P=0.03). The predictors for ipsilateral ischemic stroke were CEA (HR, 0.18; 95%CI, 0.05–0.84, P=0.03) and IPH (HR, 1.92; 95%CI, 1.26–4.28, P=0.04). The results on secondary endpoints were very similar. Conclusion – IPH may highly predict subsequent cerebrovascular events, whereas CEA may reduce these risk during a 2-year follow-up in patients with symptomatic, mild carotid stenosis. Randomized clinical trials is warranted to validate these results.