Carotid artery stenting (CAS) for restenosis (RS) after carotid endarterectomy (CEA) is presumed to have fewer complications than CAS for primary atherosclerotic (PA) lesions. It has been proposed that interventionalists may limit themselves to CAS for RS initially, while they gain additional experience during their learning curve. However, there are few studies objectively comparing the outcomes of the two groups of patients to substantiate this assumption. We analyzed prospectively collected data on CAS performed at our institution from 1996 to April 2006. Complication rates were compared between CAS performed for RS versus PA lesions. Specific end points studied included in-hospital and 30-day stroke and death rates. The incidence of transient ischemic attack (TIA) was also recorded. Patient demographic features (gender, age, hypertension, diabetes mellitus, coronary artery disease, smoking, hypercholesterolemia, and presence of preoperative neurological symptoms) were recorded. A neurologist examined all patients before and after CAS. Patients with previous CAS with in-stent RS and tandem common carotid artery-internal carotid artery or arch ostial stenoses were excluded from this analysis. CAS procedures (n = 217) performed on 210 patients fulfilled inclusion criteria for this study. Indications for CAS included RS (n = 118, 54%) and PA (n = 99, 46%). The two groups were well matched for all demographic features except hypercholesterolemia, which was more common in the PA group. Thirty-day stroke and stroke + death rates for the entire series were 2.8% and 4.1%, respectively. Within this cohort, 30-day stroke and stroke + death rates were not significantly different between the RS (2.5% and 5.1%) and PA (3.0% and 3.0%) groups. Within the RS group, these outcomes were also similar when patients treated for late recurrence (>24 months after CEA, n = 49) were compared to those treated for early recurrence (< or = 24 months after CEA, n = 67). Only when stroke and TIA were combined was a difference observed between the late recurrence (10.0%) and the early recurrence (1.5%) groups (p = 0.049). Contrary to general opinion, 30-day stroke and stroke + mortality rates from CAS for RS versus PA were not significantly different. Lower neurological event rates were only seen in CAS for early RS compared with late RS after endarterectomy when TIAs were included as an end point in the analysis. CAS for RS must therefore not be considered a low-risk procedure. Technical proficiency for CAS must be equivalent regardless of the etiology of the stenosis. These observations also underscore the need for appropriate patient selection and close follow-up of all patients undergoing CAS.