Abstracts

Rationale: Positron emission tomography (PET) is a metabolic-based imaging technique that can help localize the epileptogenic focus when other modalities such as magnetic resonance imaging (MRI) are unrevealing. However, seizures or other neurophysiological abnormalities around the time of a PET scan could alter the metabolic activity of the brain and thus the scan’s results. Simultaneous scalp electroencephalography (EEG) can reveal whether seizures or other abnormalities occurred during a PET scan. The purpose of this study is to determine the utility and concordance of EEG during PET scans. Methods: This is a retrospective chart review study for all patients with refractory epilepsy at Stanford Hospital and Clinics who underwent simultaneous EEG and PET scan between April 2010 and December 2015. We excluded patients less than 18 years old. A total of 71 charts were identified; 33 males and 38 females. Ages ranged 18-64 years (mean 39, median 37). Results: The duration of EEG recordings ranged from 1 hour and 42 minutes to 17 hours and 7 minutes (mean 3 hours and 11 minutes, median 2 hours and 58 minutes). 50 patients (70%) had abnormal EEGs, and 37 (52%) had abnormal PETs. Seizures occurred in 6 patients (8%). Of these, 2 (33%) were reported by the patient and not captured by EEG; one occurred after the PET scan ended, the other during EEG lead placement before the recording started. In the remaining 4 (67%) with seizures recorded on EEG, only 1 (25%) had an abnormal PET scan. In contrast, for the 67 patients without seizures on EEG, a larger proportion 54% (36 patients) had abnormal PET scans. Of the 21 (31%) patients with normal EEG, 12 (57%) had a PET revealing an abnormality. However, for the 46 patients (69%) with abnormal EEG, a slightly smaller proportion (52%) had abnormal PETs. EEG and PET had only 29% laterality concordance for right-sided abnormalities and 26% concordance for left-sided ones. Conclusions: This study demonstrates that a small but nontrivial proportion of patients have a seizure during PET scans, suggesting some utility for simultaneous EEG recording. In addition, a smaller proportion of patients who have seizures or other neurophysiological abnormalities have abnormal PETs, suggesting an effect of these factors on the scan. Investigation in a larger population will provide further clarification. Funding: No funding was sought for this project.