FV 1178. The More the Merrier? Diagnostic Yield of EEG after the First Unprovoked Seizure

Background: By adhering to the long-established definition, epilepsy could be diagnosed after two unprovoked seizures more than 24 hours apart. According to the new guidelines of the Deutsche Gesellschaft für Neurologie published in 2017, diagnosis of epilepsy can be ascertained after a single unprovoked seizure if the probability of recurrent seizures within the next 10 years comparable to the risk of recurrence associated with a second unprovoked seizure (more than 60%). The presence of epileptiform activity (EA) on the EEG after a first unprovoked seizure implies a 77% post-test probability of relapse in adults and 66% in children, respectively. Thus, EA on the EEG after the first unprovoked seizure entails a risk of seizure recurrence more than 60%.

Purpose: Defining the optimal and clinically practical EEG diagnostics to prove EA after the first unprovoked seizure or to exclude it with the maximum likelihood.

Questions: How many EEG recordings are necessary? How long should a routine EEG be recorded? Is a sleep EEG more meaningful than an awake EEG, or should we recommend a 24-hour EEG?

Methods: We report the conclusions of the relevant literature for adult and child neurologists.

Results: (1) Sensitivity for EA is highest, ∼50%, in the first routine EEG (awake, 20 minutes of artifact-free recording, 3 minutes hyperventilation, photic stimulation with frequencies of 1–30 Hz). Cumulative sensitivity increases up to 80 to 90% after four EEGs with much lower incremental yield subsequently. (2) Sensitivity for EA on routine EEG is highest in (a) the first EEG in cases with generalized discharges and (b) the first and following routine EEGs in children, respectively, adolescents (up to an age of 19 years). (3) The first routine EEG should be performed within the first 24 hours after a seizure. A recording within the first 12 hours after the seizure may be more meaningful, but artificial alterations could be misleading. (4) An EEG after sleep deprivation recorded 48 hours to 4 weeks after a seizure reveals a similar sensitivity as a routine EEG recorded within 24 hours after a seizure. (5) Sleep EEG following a normal routine EEG increases the sensitivity of EA by 11 to 18%. For diagnosis of sleep-dependent epilepsy syndromes (especially in infancy and childhood), a sleep EEG is mandatory. (6) Long-term EEG up to 16 to 24 hours (including sleep recording) shows a similar sensitivity to detect EA as an additional sleep EEG after normal routine EEG. A long-term EEG as first EEG recording after the first unprovoked seizure possesses a similar sensitivity to detect EA as a routine EEG within 24 hours after the seizure.

Conclusion: In children and adolescents, few routine EEGs are often sufficient to detect epileptic abnormalities. The diagnostic yield to identify epileptic abnormalities after the fourth normal routine EEG is very low. Sleep EEG further increases the sensitivity (by 11–18%). A long-term EEG after the first unprovoked seizure results in no benefit compared with an early routine EEG (< 24 hours after seizure).