Abstracts

Rationale: Many electroencephalograms (EEGs) are ordered over weekends and holidays in the inpatient hospital setting. The value of a study identifying an abnormality of significance which alters treatment is high, however, this must be balanced against the cost effectiveness of performing these studies. Our aim was to study the rationale and yield of these studies.Methods: We retrospectively searched the Henry Ford EEG laboratory database to identify patients who underwent EEGs over weekends and holidays over a 2 year interval. After hour weeknight EEGs were excluded, as time of study was not listed on reports. Demographic data, rationale for study, and requesting physician data were collected on each patient. Each report was reviewed to determine if a significant finding was present, defined as a recorded seizure or periodic epileptiform discharges (PLEDs). The associations of gender, ordering physician, and reason for EEG with significant finding were tested using chi-square tests. Two sample t-tests were done to assess the association of age with significant finding. Where possible, analysis was adjusted for multiple EEGs in the same patient. All testing was done at the 0.05 level. Results: From 7/2007-7/2009, 274 emergent EEGs on 259 individual patients were performed. Fifty two percent (134) were performed on males and the mean age was 58.7 18.6 (SD) years. Only 11% of studies demonstrated a significant finding (15 seizures, 16 PLEDS). There was no association based on gender or ordering physician. Those with significant findings were significantly older than those without (mean 57.6 vs 67.7 years, p=0.04). Significant differences in yields were also seen based on study rationale (p<0.001), with a 0% yield for studies performed for syncope or after cardio-pulmonary arrest. The highest yield was seen when the study was a follow up on recent recorded status epilepticus or recent long term EEG (64%). Conclusions: The results of our study are limited as few studies were completed for some clinical rationales and the study was performed at a single tertiary care center. The results of our study demonstrate a low yield to performing emergent EEGs, especially for syncope and post cardio-pulmonary arrest. Future studies involving more than one center may help identify the indications for performing emergent EEG studies in the era of cost containment.