Cortical spreading depression occurs during elective neurosurgical procedures

Abstract

OBJECTIVE: Cortical spreading depression (CSD) has been observed with relatively high frequency in the period following human brain injury, including traumatic brain injury and ischemic/hemorrhagic stroke. These events are characterized by loss of ionic gradients through massive cellular depolarization, neuronal dysfunction (depression of electrocorticographic [ECoG] activity) and slow spread (2-5 mm/min) across the cortical surface. Previous data obtained in animals have suggested that even in the absence of underlying injury, neurosurgical manipulation can induce CSD and could potentially be a modifiable factor in neurosurgical injury. The authors report their initial experience with direct intraoperative ECoG monitoring for CSD. METHODS: The authors prospectively enrolled patients undergoing elective craniotomy for supratentorial lesions in cases in which the surgical procedure was expected to last > 2 hours. These patients were monitored for CSD from the time of dural opening through the time of dural closure, using a standard 1 x 6 platinum electrode coupled with an AC or full-spectrum DC amplifier. The data were processed using standard techniques to evaluate for slow potential changes coupled with suppression of high-frequency ECoG propagating across the electrodes. Data were compared with CSD validated in previous intensive care unit (ICU) studies, to evaluate recording conditions most likely to permit CSD detection, and identify likely events during the course of neurosurgical procedures using standard criteria. RESULTS: Eleven patients underwent ECoG monitoring during elective neurosurgical procedures. During the periods of monitoring, 2 definite CSDs were observed to occur in 1 patient and 8 suspicious events were detected in 4 patients. In other patients, either no events were observed or artifact limited interpretation of the data. The DC-coupled amplifier system represented an improvement in stability of data compared with AC-coupled systems. Compared with more widely used postoperative ICU monitoring, there were additional challenges with artifact from saturation during bipolar cautery as well as additional noise peaks detected. CONCLUSIONS: CSD can occur during elective neurosurgical procedures even in brain regions distant from the immediate operative site. ECoG monitoring with a DC-coupled full-spectrum amplifier seemed to provide the most stable signal despite significant challenges to the operating room environment. CSD may be responsible for some cases of secondary surgical injury. Though further studies on outcome related to the occurrence of these events is needed, efforts to decrease the occurrence of CSD by modification of anesthetic regimen may represent a novel target for study to increase the safety of neurosurgical procedures.