LAUSR

ing CEA, with high sensitivity and negative predictive value. A 74-year-old gentleman having hypertension and symptomatic coronary artery disease had an episode of transient ischemic attack. Investigations revealed bilateral carotid artery (CA) stenosis (90% and 70% in left and right CA; respectively). Brain computed tomography (CT) was normal (Fig. 1A). A combined surgery of left CEA and CABG was planned. Standard American Society of Anesthesiologists monitors with transesophageal echocardiography and CO (INVOS) with bifrontal probes were used. Balanced anesthesia was used with isoflurane, fentanyl, and cisatracurium. Transesophageal echocardiographyguided fluids, inotrope and vasopressors were used to maintain cardiac output and mean arterial pressure (MAP). Optimal hemoglobin levels and ventilation strategies were used to maintain periprocedural significant drop (>20% of baseline) of cerebral oxygen saturation. Patient was adequately heparinized before the carotid clamping and subsequently for cardiopulmonary bypass during CABG. During left CA clamping cerebral oxygen saturation was well maintained bifrontally (obviating the need of intraluminal shunt) and thereafter, throughout the procedure with transient, occasional troughs which responded promptly to elevation of MAP. Anticoagulation was adequately reversed at end of CABG. Wide fluctuations in MAP and post-CEA hypertension, were distinctly avoided. Patient was extubated 6 hours later and found to have right-sided paresis. CT revealed a left temporoparietal intracranial hemorrhage (ICH). Monitoring with CO has shown to reduce the incidence of stroke in CEA and in major cardiac surgery. In CEA, intraoperative CO has been shown to be a better predictor of need for intraluminal shunt compared with transcranial Doppler5 and postoperatively, can detect cerebral hyperperfusion syndrome.2,6 CO is useful in traumatic brain injury, ICH and detecting delayed cerebral ischemia in subarachnoid hemorrhage.7 In this particular patient, CO monitoring may not have detected the stroke because of the limited interrogation depth of the CO (2 to 2.5 cm). The area of ICH in the CT image (Fig. 1B) was deeper than the interrogation depth of the device. Unfortunately, CO monitoring was not continued in the immediate postoperative period. Postoperative hypertension was not encountered in this patient ruling out the possibility of that being the cause of ICH. Hence the insult could have occurred intraoperatively while CO was being monitored; demonstrating the possible limitation of this monitoring modality. Studies show incidence of cerebral hyperperfusion syndrome is minimal (0.6%, mostly late onset; day 5).6 This report highlights that despite CO being a safe, noninvasive, easy to use tool for monitoring cerebral oxygenation, at times it fails to detect stroke promptly. We would further suggest extension of its use in postoperative period after CEA, in combination with another monitor (transcranial Doppler) or posterior/ lateral application of probes to widen the interrogation area for added accuracy.