Abstracts

Rationale: Currently, it is not possible to reliably predict the occurrence of epileptic seizures noninvasively. If possible, such prediction might allow patients and/or caregivers to implement interventions to prevent seizures and associated complications. Based on previous fMRI and SPECT studies, it has been suggested that seizures may be preceded by increased cerebral blood flow. Recently, we demonstrated that transcutaneous regional cerebral oxygen saturation (rSO2) sensors are feasible for use in patients with convulsive seizures undergoing video EEG monitoring. It has been suggested by others that such noninvasive monitoring can also be used to detect pre-ictal increases in cerebral blood flow, thereby predicting subsequent seizure occurrence. Accordingly, we reanalyzed our data to determine if convulsive seizures were consistently marked by increased cerebral oxygenation. Methods: We retrospectively reanalyzed data gathered for a feasibility project involving noninvasive cerebral oximetry. Patients with histories of generalized tonic clonic seizures (GTCS) were recruited into that IRB-approved study between December 2011 and May 2012. All subjects were evaluated with continuous 30-channel scalp EEG and 2 rSO2 sensors placed on each side of the forehead. Data from the rSO2 sensors were recorded every 4 seconds by a Nonin Equanox Model 7600 Regional Oximeter (Nonin, Plymouth, MN, U.S.A.). We calculated the mean rS02 value from the most reliable sensor for the 1 hour epochs in the non-ictal (2 hours prior to seizure onset) and pre-ictal (1 hour prior to seizure onset) periods. The occurrence and timing of rSO2 values in the pre-ictal period >=3 standard deviations (SDs) from the mean rSO2 value calculated in the non-ictal period were recorded. Results: Five patients underwent prolonged video-EEG and rSO2 monitoring, during which 7 primary or secondarily GTCS with usable data were captured. The mean rSO2 value in the non-ictal period was 75.6+/-5.7%. This significantly increased to 76.0+/-6% in the pre-ictal period (Paired Samples T Test 2 tailed p=0.032). Four of the 7 GTCs (57.1%) were marked by at least 3 sequential rSO2 values in the pre-ictal period that were >=3 SDs greater than the mean rSO2 value recorded during the non-ictal period. On average, such values were noted 18 minutes 30 seconds (18:30) prior to electrographic seizure onset (range 7:40-33:54). Three GTCs (42.9%) were marked by sustained cerebral hyperoxia for 15 or more consecutive readings. Such sustained readings were noted a mean of 15:45 prior to seizure onset (range 6:52-33:06). All recorded elevations in cerebral oxygenation resolved prior to seizure onset. Conclusions: Increased cerebral oxygenation as measured by noninvasive rSO2 sensors occurred in a majority of our cohort's recorded GTCS. Our data suggests that increased cerebral oxygenation could potentially be noninvasively monitored and used to predict seizure occurrence. Larger studies are now needed to determine the value of noninvasively measured cerebral hyperemia in seizure prediction.