Carotid angioplasty with or without stent placement (CAS) has emerged as an alternative to carotid endarterectomy (CEA) for revascularization of severe carotid artery stenosis in patients with high risk for surgery. This meta-analysis compared the efficacy and safety of both treatments from data for previous randomized trials. We did a literature search using Medline, PubMed, Cochrane database, and relevant articles for randomized trials comparing CAS with CEA. A meta-analysis using both random-effects and fixed-effects models compared outcome events of death, stroke, myocardial infarction, and cranial nerve injury at 30 days, 6 months, or 1 year after procedure. A total of nine trials involving 3138 patients (1564, CEA; 1574, CAS) with symptomatic or asymptomatic stenosis were included for analysis. By random-effects model, there was no significant difference of event rates between treatments for any stroke (odds ratio for CAS [95% confidence interval], 1.46 [0.91-2.36]), death or any stroke (1.37 [0.90-2.10]), or death, any stroke, or myocardial infarction (1.02 [0.49-2.11]) at 30-day, and death and any stroke at 6 months (1.50 [0.69-3.23]) or 1 year (1.25 [0.59-2.63]). But, there were significantly higher 30-day event rates after CAS than CEA for death or any stroke (1.37 [1.04-1.81]) by fix-effects model, accompanied with significant heterogeneity (p=0.04). Risk of cranial nerve injury was much lower in CAS than in CEA (0.12 [0.05-0.29]). Except for lower risk of cranial nerve injury, CAS is neither safer nor associated with a better short-term outcome as compared to CEA in treating carotid artery stenosis.