Electroencephalography (EEG) has 25–56% sensitivity for the detection of epileptiform activity (EA) in children with seizures.1 Sleep-deprived EEG (SD-EEG) increases the activation rate by 20–50%.2 SD-EEG can be difficult for both families and children to tolerate.3 The groups of children who are likely to have abnormal SD-EEG were unknown.
To determine clinical and neurophysiological factors that would predict an abnormal SD-EEG, a retrospective, single centre review, over a 4-year period of children (0–16 years) who had SD-EEG within 6 months of non-epileptiform routine EEG (R-EEG) was conducted. R-EEG was defined as normal or with non-epileptiform abnormalities (NEA), and SD-EEG was normal or with EA. Age, sex, results of the routine EEG, achievement of sleep during SD-EEG, usage of anti-epileptic drugs, previous history of epilepsy and the presence of other neurological diseases were documented.
107 children met the inclusion criteria. SD-EEG showed EA in 39 (36%) children. Multiple logistic regression analysis demonstrated that male sex and the presence of NEA in R-EEG were significantly associated with epileptiform SD-EEG (adjusted odds ratio 2.95 and 2.45, p value 0.03 and 0.03, respectively). We found no association between increased yield of SD-EEG and other factors studied, including whether the child slept or not during the SD-EEG.
The sensitivity of 36% of the SD-EEG is comparable with other studies. There is a much higher incidence of EA in patients who have NEA on a routine EEG. SD-EEG should always be considered in children with a high suspicion of epilepsy, in particular those with NEA on R-EEG.