Abstracts

Rationale: Electroencephalograms (EEGs) are commonly used in the investigation of patients presenting with seizures, paroxysmal events, loss of consciousness, confusion and new onset psychiatric symptoms in order to rule out the possible diagnosis of a seizure disorder. Objective: To determine the incidence of diagnostic epileptic activity on EEGs performed on adults and to determine whether the yield of EEG is different based on the reason for referral. Methods: Single-center retrospective study, of sequentially referred adult patients more than 17 yrs of age with clinical seizures, paroxysmal event of unknown etiology, altered mental status or new psychiatric symptoms. Patients having long-term video-EEG recording and patients who had EEG in an intensive care unit were excluded. Standard 30 minute EEGs were performed by an EEG technician using a commercially available cap and interpreted by one of three fellowship trained electroencephalographers. Between group difference in results was compared using Analysis of Variance (ANOVA). Results: Over an 18 month period, 569 patients underwent 600 routine EEGs. Twenty-seven had two EEGs, two had three EEGs and the remaining patients had a single EEG. One hundred and eighty-two patients were referred for EEG with a diagnosis of epilepsy or seizures, 276 with paroxysmal events of unknown etiology, 75 with altered mental status and 36 with a new onset of psychiatric symptoms. Overall, epileptic activity was found in 41 of 569 patients (7.2%). Epileptic activity was found in 20 of 182 (11%) patients with seizures or epilepsy, 18 of 276 (6.5%) patients with paroxysmal events, three of 75 (4%) of patients with altered mental status and zero of 36 (0%) patients with new onset psychiatric symptoms. Overall ANOVA demonstrated a significant difference in the incidence of epileptic activity for different reasons of referral (p=0.045) with no specific between group comparisons reaching significance on post-hoc analysis. Conclusions: In patients referred with paroxysmal events and altered level of consciousness, the incidence of a diagnostic finding on EEG was relatively low (6.5% and 4% respectively). Given the fact that epileptic EEG findings were only observed in 11% of patients with a known diagnosis epilepsy or seizures, a nondiagnostic result cannot be considered to rule out epilepsy. The observation that no patients with new onset psychiatric symptoms had a positive EEG could imply that epilepsy patients are being accurately diagnosed by primary care physicians and not mistakenly referred for psychiatric evaluation.