Selective versus routine intraoperative shunting during carotid endarterectomy: a multivariate outcome analysis
Graeme F Woodworth1, Matthew J McGirt, Khoi D Than
1Department of Neurosurgery, Johns Hopkins Hospital, Baltimore, Maryland, USA. gw@jhmi.edu
Objective:
The use of intraluminal shunting during carotid endarterectomy (CEA) remains controversial. Over the years, different shunting strategies have been used. More recently, the use of intraoperative electroencephalography and somatosensory evoked potential monitoring with selective intraluminal shunting has been explored. No studies have assessed the independent association of selective versus routine intraluminal shunting to outcomes after CEA.
Methods:
The clinical and radiological records of all patients undergoing CEA from 1994 to 2006 at an academic institution were reviewed retrospectively to assess outcomes at 72 hours. The independent association of selective intraluminal carotid artery shunting during CEA and perioperative stroke within 72 hours was assessed through multivariate logistic regression analysis.
Results:
In 1411 patients with both symptomatic and asymptomatic extracranial carotid artery disease, there were a total of 49 (3.5%) perioperative strokes after CEA. There were two (1%) cases of perioperative strokes among 194 patients in the selective shunting group compared with 47 out of 1217 (4%) in the routine shunting group (P = 0.04). Symptomatic carotid artery disease was associated with a twofold increase in the odds of experiencing perioperative stroke (odds ratio, 1.95; 95% confidence interval, 1.08-3.52; P = 0.03). Patients undergoing electrophysiological monitoring with selective intraluminal carotid artery shunting were more than seven times less likely to experience a perioperative stroke (odds ratio, 0.05; 95% confidence interval, 0.01-0.40; P < 0.01). Increasing cumulative surgical volume, particularly more than 200 total cases, was associated with more than a twofold decrease in perioperative stroke (odds ratio, 0.38; 95% confidence interval, 0.20-0.74; P < 0.01).
Conclusion:
Regardless of symptomatic carotid artery disease or cumulative surgical volume, patients undergoing CEA with intraoperative electroencephalography and somatosensory evoked potential monitoring with selective intraluminal carotid artery shunting had a stroke rate lower than that of the routine shunting group. Selective shunting based on electroencephalography and somatosensory evoked potential monitoring may be superior to the nonselective strategy.