Abstracts

Activity-evoked optical changes in cortex are known as [apos]intrinsic optical signals[apos] (IOS). [italic]In vivo[/italic] IOS is dominated by hemodynamic components, and hence a brief neuronal discharge generates a signal that is on the order of seconds. Although IOS can localize an area of tissue involved in the generation and spread of an epileptic discharge, it is believed to lack the temporal resolution required to distinguish sites of onset from sites through which the activity has propagated. In order for IOS to be a practical tool for the intraoperative localization of epileptogenic tissue, it is desirable to improve the temporal resolution of the technique. The goal of this study was to determine if real-time optical imaging of interictal activity could be achieved through use of improved detectors and signal processing. IOS data of interictal activity was acquired either from an acute primate seizure model (n = 3) or intraoperatively from patients undergoing surgical treatment for medically intractable seizures (n = 5). Epileptiform activity in primates was generated by placement of a bicuculline-soaked pledget (0.5 x 0.5 mm; 100 [mu]M) on the cortex for ten minutes. Optical maps of interictal activity were correlated with surface EEG electrode recordings (humans) or single-unit recordings (primates). A cooled digital CCD camera that provided 16-bits of true information was used. Heartbeat, respiration, and motion artifacts were removed from the data through signal processing algorithms based on wavelet transforms and nonparametric spline estimation. Single-unit recordings acquired simultaneously with optical images demonstrated that changes in IOS following the time course of interictal discharges could be recorded in primate cortex. In four intraoperative studies on human subjects, IOS mapping identified cortical areas involved in the generation of interictal activity that were closely correlated to EEG activity. In a fifth patient, optical imaging identified interictal and ictal activity which was not reflected in the EEG recordings; in this case, clinical correlates (i.e. motor movements) confirmed that epileptic activity was occurring. These studies demonstrate that IOS can provide maps of epileptic activity in cortex with sufficiently high temporal resolution to identify sites generating interictal activity. In one case, IOS mapping identified ongoing epileptic activity that could not be detected with standard EEG recording techniques. This suggests that in some cases, IOS mapping can localize epileptic activity that can not be mapped with other currently available methods. In particular, this work is supportive of the notion that IOS mapping has the potential to be a practical method for the intraoperative mapping of epileptogenic tissue. (Supported by a NIH/NINDS R21NS042341 (DWH), and NIH/NINDS K08NS01828 and R21NS048114 (MMH).)