Contemporary results of carotid endarterectomy in "normal-risk" patients from the Society for Vascular Surgery Vascular Registry.

ABSTRACT

OBJECTIVE: Acceptable complication rates after carotid endarterectomy (CEA) are drawn from decades-old data. The recent Carotid Revascularization Endarterectomy versus Stenting Trial (CREST) demonstrated improved stroke and mortality outcomes after CEA compared with carotid artery stenting, with 30-day periprocedural CEA stroke rates of 3.2% and 1.4% for symptomatic (SX) and asymptomatic (ASX) patients, respectively. It is unclear whether these target rates can be attained in "normal-risk" (NR) patients experienced outside of the trial. This study was done to determine the contemporary results of CEA from a broader selection of NR patients. METHODS: The Society for Vascular Surgery (SVS) Vascular Registry was examined to determine in-hospital and 30-day event rates for NR, SX, and ASX patients undergoing CEA. NR was defined as patients without anatomic or physiologic risk factors as defined by SVS Carotid Practice Guidelines. Raw data and risk-adjusted rates of death, stroke, and myocardial infarction (MI) were compared between the ASX and SX cohorts. RESULTS: There were 3977 patients (1456 SX, 2521 ASX) available for comparison. The SX group consisted of more men (61.7% vs 57.0%; P = .0045) but reflected a lower proportion of white patients (91.3% vs 94.4%; P = .0002), with lower prevalence of coronary artery disease (P < .0001), prior MI (P < .0001), peripheral vascular disease (P = .0017), and hypertension (P = .029), although New York Heart Association grade >3 congestive heart failure was equally present in both groups (P = .30). Baseline stenosis >80% on duplex imaging was less prevalent among SX patients (54.2% vs 67.8%; P < .0001). Perioperative stroke rates were higher for SX patients in the hospital (2.8% vs 0.8%; P < .0001) and at 30 days (3.4% vs 1.0%; P < .0001), which contributed to the higher composite death, stroke, and MI rates in the hospital (3.6% vs 1.8; P = .0003) and at 30 days (4.5% vs 2.2%; P < .0001) observed in SX patients. After risk adjustment, the rate of stroke/death was greater among SX patients in the hospital (odds ratio, 2.05; 95% confidence interval, 1.18-3.58) although not at 30 days (odds ratio, 1.36; 95% confidence interval, 0.85-2.17). No in-hospital or 30-day differences were observed for death or MI by symptom status. CONCLUSIONS: The SVS Vascular Registry results for CEA in NR patients are similar by symptom status to those reported for CREST and may serve as a benchmark for comparing results of alternative therapies for treatment of carotid stenosis in NR patients outside of monitored clinical trials. The contemporary perioperative risk of stroke after CEA in NR patients continues to be higher for SX than for ASX patients.