Comparison of Epileptiform Activity in Routine and Sleep-Deprived Electroencephalogram (EEG) Studies: A Five-Year...

Epilepsy is a neurological syndrome with a significant associated morbidity and mortality. An electroencephalogram (EEG) is an important tool to assist in the diagnosis of epilepsy. Sleep deprivation is a well-known risk factor for seizures and forms a basis for why sleep-deprived EEGs may assist in diagnosing epilepsy. There is mixed evidence regarding the utility of sleep deprivation in identifying signs of seizure activity, known as epileptiform activity. This study aimed to assess whether sleep-deprived EEGs increase the yield of epileptiform activity, compared with routine EEGs.

This was a retrospective observational study (January, 2018 to January, 2023) in patients of all ages who underwent routine and sleep-deprived EEGs at a major tertiary referral centre. Descriptive statistical analysis was undertaken in addition to the Fisher exact test and McNemar test to facilitate comparison of matched pairs.

There were 116 patients included in the study. The median age at time of first EEG was 32.5 years (interquartile range (IQR) 18-53). Fifty four percent of patients were male and 80% of patients were aged ≥18 years. The diagnostic yield of epileptiform activity was 21% when the results of both the routine and sleep-deprived EEG studies were included. There was no significant difference in the yield of epileptiform activity in routine versus sleep-deprived EEGs (12% versus 13% respectively, p = 1.0). Nine patients had epileptiform activity present on routine but not sleep-deprived EEG, and the reverse was true for 10 patients. There were several study limitations, including the varied time interval between EEG studies and limited referral documentation.

The study findings suggest that, in the correct clinical context, undertaking both a routine and sleep-deprived EEG, regardless of the order, may have benefit in identifying epileptiform activity. These findings also reinforce important data points that should be included with EEG referrals, to help optimise future research in this area.