Comparison of Electroencephalography and Cerebral Oximetry to Determine the Need for In-Line Arterial Shunting in Patients Undergoing Carotid Endarterectomy

Objective: To compare cerebral near-infrared regional spectroscopy (NIRS) with the 12-lead electroencephalogram for the detection of ischemia during carotid artery clamping for carotid endarterectomy (CEA). Design: Prospective, observational. Setting: Single, tertiary care center. Participants: Ninety patients older than 18 undergoing elective, unilateral CEA. Interventions: In addition to EEG monitoring, all patients underwent continuous blinded NIRS monitoring with sensors placed bilaterally above the supraorbital ridge. Measurements and Main Results: Seventeen patients were excluded, leaving 73 patients available for evaluation. Four patients (5.5%) required shunting based on EEG findings. Changes in cerebral oxygen saturation (rSO(2)) were assessed on the operative side using the average value for the 1 minute prior to cross-clamp and the lowest rSO(2) value the first 5 minutes postclamp. Each 1% absolute decrease and each 1% relative decrease from baseline conferred a 50% increase in the need for shunt placement (OR 1.5; 95% CI (1.03-2.26); p = 0.03 and OR 1.4; 95% CI (1.02-1.81); p = 0.04 respectively). Sensitivity, specificity, and positive and negative predictive values were determined using significant cutoffs of >= 5% absolute change or >= 10% relative change. Positive predictive value was low ( <25%) for both absolute and relative changes. Conclusions: A decrease in rSO(2) during carotid crossclamping for CEA is associated with EEG-determined need for shunting, but the positive predictive value is low. Using the above cutoffs in the current series would have resulted in an increase in the shunt rate by approximately 20% when it was not indicated by EEG. (C) 2013 Elsevier Inc. All rights reserved.