Primary closure of the carotid artery is associated with poorer outcomes during carotid endarterectomy

Introduction: Arterial endarterectomy and reconstruction during carotid endarterectomy (CEA) can be performed in a variety of ways, including standard endarterectomy with primary closure, standard endarterectomy with patch angioplasty, and eversion endarterectomy. The optimal method of arterial reconstruction remains a matter of controversy. The objective of this study was to determine the effect of the method of arterial reconstruction during CEA on perioperative outcome. Methods: A retrospective cohort study of consecutive CEAs performed by 81 surgeons during 1997 and 1998 in six regional hospitals was performed. Detailed clinical data regarding each case and all deaths and nonfatal strokes within 30 days of surgery were ascertained by an independent review of the inpatient chart, outpatient surgeon record, and the hospitals' administrative databases. Two physician investigators - one neurologist and one internist - confirmed each adverse event by independently reviewing patients' medical records. Results: A total of 1972 CEAs were performed. The mean age of the patients was 72.3 years, and 57.2% were male. Preoperative neurologic symptoms occurred in 28.7% of cases (n = 566), and the remaining 71.3% were asymptomatic before surgery (n = 1406). The method of arterial reconstruction was chosen by the surgeon. Primary closure was performed in 11.8% (n = 233), patch angioplasty in 69.8% (n = 1377), and eversion endarterectomy in 18.4% (n = 362). There was no significant difference in the preoperative symptom status of patients who underwent primary closure compared with the other methods of reconstruction (72.5% asymptomatic vs 71.1%, p = NS). Primary closure cases were significantly more likely to experience perioperative stroke compared with the other closure techniques (5.6% vs 2.2%, P = .006). Primary closure cases also had a higher incidence of perioperative stroke or death compared with the other closure techniques (6.0% vs 2.5%, P = .006). There were no significant differences with regard to either perioperative stroke, or perioperative stroke/death noted when comparing patch angioplasty with eversion endarterectomy: stroke, 2.2% vs 2.5% (P = NS) and stroke/death, 2.5% vs 2.5% (P = NS) respectively. Conclusion: It appears that primary closure is associated with significantly worse perioperative outcomes compared with endarterectomy with patch angioplasty and eversion endarterectomy, even when the preoperative symptom status of the patient cohorts is equivalent. Although some of its advocates have reported that they can properly select appropriate patients for primary closure based on the size of the artery and other factors, the data demonstrate that these patients have poorer outcomes nonetheless. Primary closure during carotid endarterectomy should predominantly be abandoned in favor of either standard endarterectomy with patch angioplasty or eversion endarterectomy.