Abstracts

Rationale: Localization of motor and language function in brain tumor surgeries is important to allow for optimal resection while preserving eloquent cortex. To assist with localization, direct cortical stimulation (DCS) is used intraoperatively, but this may induce seizures requiring use of anesthetics that may alter the physiological responses of cortical neurons and alter the results of the mapping. The aim of this study is to determine whether there are patient- or electrocorticogram (ECOG)- related factors that can predict the risk of clinical seizures induced during DCS.Methods: The charts and ECOG recordings of patients who underwent DCS during brain tumor surgeries at Duke University Medical Center between 7/1/2014 and 5/1/2015 were reviewed. Previous history of seizures, use of antiepileptic drugs (AED), tumor location, tumor grade, and anesthetic regimen used during DCS were investigated as possible risk factors. Clinical seizures requiring cold saline irrigation or use of anesthetics or AED were noted. ECOGs were recorded using subdural 4, 6 or 8-contact-strips. Stimulation was done using an Ojemann stimulator with a square wave biphasic pulse of 1 ms at 60 Hz. The stimulus intensity, stimulus duration and interstimulus interval were recorded. In addition, duration, frequency, and morphology of afterdischarges (AD) prior to seizure onset were noted.Results: A total of 11 patients had DCS with ECOG recordings in the interval time studied. Four patients (36%) had intraoperative clinical seizures during stimulation that interrupted the functional mapping. In 3 of these patients (75%) with stimulation-induced seizures, a short interstimulus interval prior to seizure onset was the only factor that was consistently associated with higher risk of seizures (less than 3 seconds). In addition, in these patients, heralding afterdischarges were seen at the time of the stimulation preceding seizure onset. In two of these patients with a stimulation-induced seizure, a second seizure was induced despite decreasing stimulus intensity with repeated stimulation. In the other patient with seizures (25%), the ECOG showed evidence of ictal discharges about 30 seconds after onset of the clinical seizure with no clear preceding after-discharges. The history of seizures, use of AEDs, tumor location, tumor grade, anesthetics used, stimulus intensity and stimulus duration did not seem to affect the risk for stimulation-induced seizures.Conclusions: The best predictor for risk of clinical seizures induced by DCS during brain tumor surgeries was a decreased interstimulus interval of less than 3 seconds. This may be related to limited time available to evaluate for evolving afterdischarges as well as possible potentiation of the effect of the preceding stimulus on the cortical neurons so that threshold for seizure onset is lowered. As a result, using an interstimulus interval of more than 3 seconds may be an effective strategy to help prevent DCS-induced seizures during these types of surgeries.