Abstracts

Rationale: Amplitude-integrated Electroencephalography (aEEG) is widely used in neonatal intensive care unit (NICU). It is useful as monitoring of seizures since seizures of neonates are so difficult to distinguish from physiological movements and, therefore, likely to be underestimated or overestimated. These features are often the case with the seizures of Severe Motor and Intellectual Disability (SMID). This study aimed to verify if aEEG could be applied to SMID for good control of epilepsy. Methods: aEEG recordings of 24-hour period were investigated in 18 bedridden patients (mean age, 24 years; age range, 14-41 years; 10 males and 8 males) with refractory epilepsy (total 28 recordings). All recordings were accompanied by simultaneous video monitoring (recorded in the same hard disk drive as aEEG) in order to compare aEEG with clinical manifestations. The nursing records including description of seizures and cares such as aspiration of sputum were also analyzed. The analysis was based on three bipolar EEG channels (Fz-C3, C3-C4 and C4-Fz according to the 10-20 system). The analysis was focused on 1) the correspondence between baseline changes on aEEG and clinical manifestations, 2) the confirmation of seizure or NOT and 3) better conditions and parameters for SMID than those in NICU. Results: It was easy to detect ictal EEG patterns by using aEEG trend monitor (figure 1), but most of ictal patterns on aEEG were unaccompanied by seizure-like manifestations. In addition, there was a profound difference between aEEG change and nursing records concerning seizures. The adjustment of medication according to aEEG trend monitor (not nursing records) did result in good control of seizures (figure 2: same patient of figure 1 after adjustment). A patient with "gelastic seizure" was identified having no seizure on aEEG and antiepileptic drugs were reduced to improve level of consciousness. As a whole, aEEG recordings were helpful for diagnosis of seizure as well as for treatment based on evidence. In 2 patients with high-voltage baseline on aEEG, it was too difficult to distinguish ictal period from the baseline. We tried to detect ictal period by changing parameters of aEEG but failed. However, ictal period couldn't be detected in only 3 recordings and, conversely, in 89% recordings it could be detected. The artifacts on aEEG such as aspiration of sputum could be easily recognized by shorter period than ictal period (vertical lines on figure 1, 2). They were confirmed by checking simultaneous videos and nursing records, too. Conclusions: aEEG recordings in SMID are also useful to distinguish seizures from similar physiological or voluntary movement and to adjust medications for better control of seizures as in NICU. The simultaneous video monitoring is so helpful to analyze seizures in detail. There, however, still remain the problems to resolve, particularly finding best parameters such as frequency filtering. To the best of our knowledge, this is the first study of aEEG applied to SMID in the world. COI: The authors have no conflict of interest to disclose with respect to this presentation.