The effects of 6 months of treatment followed by a 5- to 6-week withdrawal period of ethosuximide (ESM) in 3 children, ages 5, 10, and 10 years, with new onset absence epilepsy are reported from the Departments of Neurology, State University of New York at Stony Brook, and the Division of Child Neurology, Schneider Children’s Hospital and Albert Einstein College of Medicine, New York. The patients had responded promptly to ESM at levels of 75-90 mcg/ml, and their 24-hour ambulatory EEGs were normal. After this early withdrawal of ESM, 2 patients remained seizure-free and their 24-hour ambulatory EEG was burst free, and one was seizure-free but showed 3/sec spike-wave bursts in the EEG. Optimizing medication until both clinical and electrographic seizure activity are continuously suppressed, as determined by 24-hour ambulatory EEG, allows early withdrawal of treatment without increased risk of relapse. 
COMMENT. In this study, ethosuximide was chosen as the drug of first choice in the treatment of childhood onset absence epilepsy. In my experience, while ethosuximide and valproate are equally effective in controlling clinical absence attacks, valproate appears, in uncontrolled studies, to be superior in the suppression of electrographic 3/sec spike-wave discharges. A comparative trial of these drugs using 24-hour ambulatory EEG monitoring would be of interest. The authors found that a burst-free routine EEG will not guarantee a complete remission of electrographic seizure activity, and a clean 24-hour ambulatory EEG is essential before attempting this recommended early drug withdrawal.
A 2 year seizure free interval is the generally recommended time before antiepileptic drug withdrawal.