Abstracts

Rationale: Polysomnography (PSG) is used to analyze and diagnose sleep and sleep related events. The typical PSGrecords electrocardiogram (EKG), 2 limb electromyography, chest and abdominal belts and electroencephalogram (EEG). A routine PSG uses a limited 8 channel EEG montage. In certain cases a seizure montage is indicated and an expanded EEG montage is used with 22 leads. The American Academy of Sleep Medicine have multiple reasons for using an expanded EEG montage during a PSG: dream enactment, sleep walking or sleep talking (parasomnias). There is scarce literature on the frequency of expanded EEG montages and their utility in diagnosis. In addition, there is limited data on the preferred interpreter; comparing the electrographic interpretations of epilepsy trained versus sleep medicine trained physicians. Methods: An internal review board approved retrospective review of all sleep studies performed at the University of Miami Sleep Center from 1/1/2011 to 4/30/2014 was performed. Studies that had an extended EEG montage were included. Patients less than 15 years of age were excluded. The demographic data, reason for study, specialty of referring physician, findings, results and interpreting physician specialty were collected. In addition two board certified epileptologists reviewed the EEG portion of each study. The epileptologist were blinded to any data other than the age. EEGs were reviewed using Rembrandt software using a low frequency filter setting of 1 Hz and a high frequency filter setting of 70 Hz. These interpretations were compared to the official sleep medicine physician report to determine if there was a significant difference. Results: There were very few studies conducted with the extended seizure montage. Of the sleep studies in the database only 0.62% were done with a seizure montage. The most common reasons for the extended montage were to evaluate for nocturnal seizures, parasomnias/dream enactment . There was no difference regarding initial specialty in ordering an extended montage. Thirty-two percent of the studies were reported by the sleep physician as having abnormalities in the EEG: rare spikes and sharps. When the interpretations were compared, the kappa value was 0.57, showing moderate agreement between the two specialties. The sleep physicians reported 8 studies with abnormalities, while the epileptologists reported 4 studies as abnormal. Conclusions: There are not many seizure montages being conducted and there was no difference in specialty training when determining whether or not to order a seizure montage versus a routine study. Typical abnormalities reported in the studies were: rare sharps and spike waves. On comparison with board certified epileptologists there was a moderate agreement between the two specialties. The sleep specialists reported two times more abnormalities than the epileptologists. These increased reported abnormalities may be secondary to sleep physicians not being comfortable with benign/normal variants observed in EEGs. Having the studies reviewed by a board certified epileptologist may help eliminate referrals or further studies which may not be necessary.