Discontinuation of antiepileptic drugs after successful epilepsy surgery. A Canadian survey

Introduction: To identify the perceived practice among Canadian epileptologists regarding discontinuation of antiepileptic drugs (AEDs) following successful resective surgery for temporal and extratemporal surgery. Methods: We performed a survey of pediatric and adult epileptologists in Canada, using a 77-item questionnaire to explore attitudes, timing, rate of withdrawal, and factors contributing to the decision to withdraw AEDs after successful epilepsy surgery. Surveys were mailed with a postage-paid return envelope. Two subsequent surveys were mailed to non-respondents at 15 days intervals. All procedures received institutional review board approval. Results: Surveys were sent to 82 epileptologists in all the Canadian provinces. Sixty-six physicians answered the survey (80.5%), representing all epilepsy centers across Canada. The minimum seizure free period required after epilepsy surgery before withdrawing AEDs, varied substantially among responders: <6 months in 10%, 6-11 months in 21%, >1 year in 50%, >2 years in 12%, >2 years in 3% after. The most important factors influencing the decision to withdraw AEDs a negative EEG before discontinuation (71%), patients' preferences (78%) and the presence of unilateral mesial temporal sclerosis (70%). The most important factors against reduction were the following: patients' wishes to resume driving (67%), focal (65%) or generalized (78%) epileptiform activity on EEG after surgery, persistent isolated auras (78%), any seizures after hospital discharge (81%), and presurgical multifocal/bilateral/diffuse findings (78%). Discussion: Canadian epileptologists indicated that AED levels, EEG and MRI are typically done before discontinuing AEDs. Generally, a good candidate for stopping AEDs has focal pathology, is completely seizure free, had an anterior temporal lobe resection, complete resection of seizure focus, and has no epileptiform discharges on postoperative EEG. The data pertaining to self-reported practice styles, and actual practice may differ.