RESUMO
Background: Patients undergoing carotid endarterectomy (CEA) may suffer from cerebral hypoperfusion during the carotid cross-clamping. Near-infrared spectroscopy cerebral oximetry (NIRS) is a non-invasive method of regional cerebral oxygen saturation measurement reflecting changes in cerebral blood flow during CEA. The main goal of the study was to evaluate the accuracy of the NIRS in detecting cerebral hypoperfusion during CEA under regional anesthesia (RA) and compare it with awake neurological testing. Patients and methods: A prospective observational study of 28 patients that underwent CEA in RA and manifested neurologic deficits, and 28 consecutive controls from a tertiary and referral center, was performed. All patients were monitored with NIRS cerebral oximetry and awake testing as the control technique. Subsequently, operating characteristic curve and Cohen's kappa coefficient were determined to evaluate the reliability of the monitoring test. Results: NIRS presented a sensitivity of 27.3% and a specificity of 89.3% in comparison to awake testing. Receiver operating characteristic (ROC) curve analysis demonstrated that a decrease of at least 20% in cerebral oxygen saturation is the best threshold to infer cerebral hypoperfusion. However, the respective area under the curve (AUROC) was 0.606 (95% CI: 0.456-0.756, P = 0.178) with a calculated Cohen's kappa of 0.179, P = 0.093. Regarding 30-days outcomes, only awake testing has shown significant associations with stroke and postoperative complications (P = 0.043 and P = 0.05), which were higher in patients with post-clamping neurologic deficits. Conclusions: NIRS demonstrated a reduced discriminative capacity for critical cerebral hypoperfusion, and does not seem to add substantial clinical benefits to the awake test.