Abstracts

Rationale: EEG is a cornerstone for epilepsy diagnosis. The Centers for Medicare and Medicaid Services have discussed changing outpatient EEG reimbursement to be flat regardless time recorded, leaving neurophysiology labs to question the benefit of performing studies longer than 20 to 30 minutes. Therefore, we investigated the difference in the yield of epileptiform abnormalities in a 30 minute versus extended (45 minutes or longer) outpatient EEG in patients without a previous diagnosis of epilepsy who were newly referred for EEG following a clinical event. Methods: Outpatients with EEGs performed between October 15, 2013 and May 15, 2014 were reviewed. Patients of all ages were included if they had never been diagnosed with epilepsy and never had an EEG performed at our institution. Hospital inpatients and patients with EEGs less than 45 minutes were excluded. EEGs were reviewed by staff or fellows at the time they were performed. Within each single EEG, the reviewer's impression was marked after the first 30 minutes, and again at completion of the study, and the differences in impression compared to identify late discharges, defined as new discharges or discharges in a new focus seen after the first 30 minutes. Data were collected on recorded events including whether it occurred during or after the first 30 minutes of EEG, and on the indication for EEG. Generalized seizure, focal seizure, new seizure, or infantile spasm indications were considered seizure-specific. Statistics are descriptive with means and standard deviations for demographic and EEG duration data, as well as percentages with confidence intervals for other EEG and event data. Differences of greater than 10% were considered clinically significant for nonequivalence. Results: A total of 1803 EEGs were reviewed, with 940 meeting inclusion criteria. The average EEG duration was 59.4 minutes (SD +/- 6.4 minutes). Eighty patients (8.5%, 95% CI 6.9-10.5) had epileptiform abnormalities on EEG at any time during the recording, with 20 (25%, 95% CI 16.8-35.4) demonstrating late discharges. A total of 87 patients (9.3%) had a seizure-specific indication for EEG. Of these, 27 (31%, 95% CI 22.2-41.4) had an epileptiform change at any time, with 7 (26%, 95% CI 13.2-44.7) demonstrating late discharges. Additionally, 130 patients (13.8%) had events captured. Thirty-six events (27.7%, 95% CI 20.7-35.9) occurred only after 30 minutes. Ten patients with any epileptiform abnormalities had events captured, of these 8 were seizures, with 4 occurring after 30 minutes, and 2 were myoclonus associated with Creutzfeldt-Jakob disease. Conclusions: A quarter of epileptiform discharges and recorded events in this prospective study occurred after 30 minutes. These data caution against using a shortened recording time for patients referred to evaluate seizures and related events.