Abstracts

Rationale: Childhood absence epilepsy (CAE) is relatively common and affected patients may suffer cognitive impairment both during seizures and in the interictal period. Recent studies have suggested that effective medications for controlling seizures might also have a disease-modifying effect on long-term prognosis. To further investigate this possibility we sought to determine historical trends in long-term prognosis before and after effective treatments were introduced for treatment of CAE. Methods: We performed a meta-analysis of 33 studies published between 1923 and 2013 to address whether CAE prognosis has improved over time, and, if so, if this improvement can be linked to the treatment of CAE with valproic acid (VPA) and ethosuximide (ESX). Studies were included on the basis of subjects meeting diagnostic criteria for CAE, specifically bilateral synchronous 3 Hz spike-wave discharges and onset of seizures in childhood, and tabulation of the remission rate in the study. For each study, seizure freedom, defined as the absence of seizures for the duration of follow-up, was calculated. In studies where subjects were trialed off medication, seizure freedom once off of medication was additionally calculated. Results: Within the 33 identified studies, 2982 subjects were identified, 936 of whom were trialed off medication. Rate of seizure freedom off medication was plotted against year of publication, and the inflection point for the curve of best fit was found to be 1975. Studies published prior to 1970 show significantly worse seizure control off medication than studies published after 1980 (33% vs. 77%, t(10) = 5.76, p < 0.001). The percentage of patients experiencing seizure freedom was significantly greater in patients on ESX or VPA versus all other medications. This effect was seen both while patients were on medication (78% vs. 52%, ?2 (1,N=1474)>10, p < 0.001), and once medication had been stopped (64% vs. 28%, ? 2(1,N=940)>10, p < 0.001). Age of onset of CAE ranged from 1 ?" 16 years, peaking at 6 ?" 11 years. Follow-up ranged from 8 months ?" 25 years. Dropout rate ranged from 0-52% and was not correlated to year of publication. Conclusions: This retrospective study suggests that CAE prognosis has improved significantly over time, and that improved prognosis may be linked to usage of effective medications such as ESX and VPA. In addition, this study suggests that early and effective treatment of CAE might provide long-term clinical benefit to patients. Further research defining the factors determining long-term seizure freedom in patients with CAE, as well as studies examining the mechanism by which effective treatments may improve prognosis, is warranted. Funding: None.