Bispectral index variations in patients with neurological deficits during awake carotid endarterectomy

Abstract

The bispectral index (BIS) is derived from the EEG and therefore may be useful to diagnose intraoperative cerebral ischaemia. This study was undertaken to investigate BIS changes in awake patients with and without neurological deficits during carotid endarterectomy under regional anaesthesia. Seventy consecutive carotid endarterectomies under regional anaesthesia were divided into two surgical groups: patients with and patients without neurological deficits. Patients' neurological status was evaluated and neurological deficits were compared with BIS values. Measurements were made at different surgical stages: baseline, after sedation, at the beginning of surgery, at preclamping, at the 3 min clamping test, during shunt insertion, at declamping, 15 min after declamping and at the end of surgery. We performed intergroup and intragroup comparisons of BIS values. A decrease in BIS of at least 10 associated with neurological deficits was taken as the cut-off point for the classification of patients with logistic regression models (crude and adjusted for potential confounders). Thirteen patients (18.6% of the total) developed clinical cerebral ischaemia, though BIS values decreased in 10 of these patients (76.9%). The mean BIS values were 92.5+/-5.6 and 84.7+/-12.3 for patients without and with neurological deficits, respectively (P value<0.05). The odds ratios of a BIS decrease associated with neurological deficits were 8.5 (95% confidence interval 2.1-35.1) and 5.4 (95% confidence interval 1.2-24.3) adjusted for contralateral stenosis. Our results describe a relationship between BIS reductions and neurological deficits during carotid surgery in awake patients.