PURPOSE: EEG recording during sleep is widely used in the assessment of epilepsy, particularly in candidates for surgery, yet the diagnostic value of this procedure is not well established. We evaluated the predictive reliability of interictal epileptiform discharges (IEDs) for localization in presurgical patients with temporal lobe epilepsy (TLE) during non-REM sleep. METHODS: Preoperative scalp EEG recordings with waking and sleep states were assessed in 83 patients with TLE in whom localization of the epileptogenic zone was subsequently confirmed by successful surgical treatment (patient seizure-free >1 year). RESULTS: The accuracy of EEG recordings for prediction of lateralization significantly changed from 51.8% during waking to 78.3% during sleep. After exclusion of patients who showed no discharges, the predictive value changed from 74.1 to 86.7%. However, in patients in whom the waking scalp EEG lateralized incorrectly, no improvement in reliability was achieved by sleep recording. CONCLUSIONS: Our results suggest that IEDs occurring in non-REM sleep provide more accurate information for lateralization of epileptogenesis than do those occurring during waking. This gain of diagnostic information was obtained in patients who showed either bilateral or no discharges in waking records, because unilateral discharges arising de novo in sleep were always correctly lateralizing. On the other hand, in patients who showed unilateral discharges in the awake state, whether ipsilateral or contralateral to the epileptogenic zone, the findings were generally unchanged during sleep.
PURPOSE: EEG recording during sleep is widely used in the assessment of epilepsy, particularly in candidates for surgery, yet the diagnostic value of this procedure is not well established. We evaluated the predictive reliability of interictal epileptiform discharges (IEDs) for localization in presurgical patients with temporal lobe epilepsy (TLE) during non-REM sleep. METHODS: Preoperative scalp EEG recordings with waking and sleep states were assessed in 83 patients with TLE in whom localization of the epileptogenic zone was subsequently confirmed by successful surgical treatment (patientseizure-free >1 year). RESULTS: The accuracy of EEG recordings for prediction of lateralization significantly changed from 51.8% during waking to 78.3% during sleep. After exclusion of patients who showed no discharges, the predictive value changed from 74.1 to 86.7%. However, in patients in whom the waking scalp EEG lateralized incorrectly, no improvement in reliability was achieved by sleep recording. CONCLUSIONS: Our results suggest that IEDs occurring in non-REM sleep provide more accurate information for lateralization of epileptogenesis than do those occurring during waking. This gain of diagnostic information was obtained in patients who showed either bilateral or no discharges in waking records, because unilateral discharges arising de novo in sleep were always correctly lateralizing. On the other hand, in patients who showed unilateral discharges in the awake state, whether ipsilateral or contralateral to the epileptogenic zone, the findings were generally unchanged during sleep.
Authors: Jennifer L Dewolfe; Beth Malow; John Huguenard; Robert Stickgold; Blaise Bourgeois; Gregory L Holmes Journal: Epilepsy Curr Date: 2013-01 Impact factor: 7.500