ObjectiveElectroencephalography (EEG) is a standard investigation after a first unprovoked seizure but the diagnostic value in adults remains unclear. Our objective was to investigate the diagnostic value of EEG after a first unprovoked seizure in a population-based cohort in Gothenburg, Sweden. MethodsThis retrospective population-based study included adult patients referred by a neurologist for EEG after a first unprovoked seizure from August 2016 - December 2019 in the greater Gothenburg catchment area. In total 520 patients were included and followed in medical records for a median of 4.42 years (range 0.33–6.93). Primary outcome measures were epileptiform activity. Medical records were reviewed for the influence of EEG on clinical management. The frequency of epileptiform activity was calculated and compared in different subgroup analyses. ResultsEEGs were recorded after a median of 70 days from the seizure. In total, 7.7 % had epileptiform activity, and 27.9 % had pathological slowing on their EEG. Seizure recurrence occurred in 33.7 % (29.6–37.8). Epileptiform activity predicted seizure recurrence with a recurrence risk of 82.5 % (70.7–94.3). The sensitivity of epileptiform activity was only 19.2 % (13.6–25.9) but the specificity was 98.0 % (95.9–99.2) and the relative risk of recurrence in presence of such activity was 2.8 (2.3–3.5). The EEG findings resulted in an alteration of diagnosis and/or medication in 4.4 % of the patients. Factors associated with epileptiform activity on EEG were younger age, EEG including sleep recording, generalized onset seizure, shorter time from seizure to EEG and longer duration of EEG recording. SignificanceThe yield of EEG recorded with a longer latency after a first unprovoked seizure in adults was low. However, epileptiform activity highly predicted seizure recurrence and our results suggest that epileptiform activity on EEG could be explored further as a biomarker for defining epilepsy already after a first seizure.
Read full abstract