Abstracts

Rationale: Long-term video-electroencephalography monitoring (LTM) in epilepsy monitoring units (EMUs) exposes patients to a variety of serious adverse events (SAEs) and safety issues. It has been reported that up to 25% of patients undergoing LTM might suffer from a SAE possibly or likely related to AEDs withdrawal. However, because of the retrospective design of these studies, there is still a lack of high-quality evidence on the risks and risk factors of adverse events in EMUs. Here, we report the preliminary data of a prospective multicenter assessment of patient’s safety in 22 French EMUs. Methods: The SAVE study is an ongoing multicenter stratified cluster-randomized controlled, parallel group, open-label trial, focusing on the management of AEDs withdrawal during LTM in patients with drug resistant seizures (NCT02679846). SAVE is assessing the impact of a standardized protocol of AEDs withdrawal during LTM on the frequency of seizure-related SAEs occurring during these monitorings. The setting of the study includes a prospective evaluation phase, prior to randomization, during which the organizational characteristics, baseline EMUs' activity, current management of AEDs withdrawal, and rate of SAEs of each participating center are evaluated. The present report corresponds to the analysis of the data collected during this baseline period. SAEs were defined as the occurrence of one or more of the following events: (i) Convulsive or non-convulsive status epilepticus; (ii) 4-hour seizure cluster (≥ 3 seizures / 4 hours) ; (iii) secondarily generalized seizure (SGTCS) unusual for the patient (≤ 1/year) ; (iv) vertebral compression ; (v) other fracture (including broken tooth) ; (vi) post-ictal psychosis ; (vii) post-ictal aspiration pneumonia ; (viii) cardio-respiratory arrest ; (ix) any other seizure-related serious injury or adverse events. Results: The baseline period of the SAVE study started in February 2016 and ended in January 2017. A total of 313 patients with refractory epilepsy were included over this period, including 288 with available data regarding AEDs withdrawal and occurrence of SAEs during the LTM. Among these 288 patients, 43 underwent invasive recordings using stereo-electroencephalography. A minority of patients (n=15, 5%) did not receive any AED at the time of the LTM. Among the 273 others, at least one of the ongoing AEDs was tapered during the LTM in 251 (87%), including 40 (14%) in whom all AEDs were withdrawn. Overall, SAEs occurred in 29 patients (10%). 13 patients (4.2%) demonstrated unusual SGTCS, eight (2.7%) a 4-hour seizure cluster, one a seizure-related fracture of a finger, one a depressive episode and one a deep venous thrombosis. In the four remaining patients, the SAEs were related to SEEG, consisting in breakage and/or pull-out of an electrode during a seizure in two, sub-cutaneous bleeding in one and severe headache in the third. SAE recovered in all patients, though three patients required prolonged hospitalization. Conclusions: In this prospective study, the overall rate of SAEs was lower than previously reported in retrospective studies. This difference might primarily be related to the lower incidence of psychiatric SAEs and of status epilepticus in our cohort. Analysis of the impact of the management of AEDs withdrawal on the frequency and/or type of seizure-related SAEs is ongoing Funding: French Ministry of Health (PHRC National 2013 : 13-0536)