Abstracts

Rationale: Although there are MEG systems tailored for infants1,2, MEG systems designed primarily to accommodate adult subjects have been routinely used for children. The practicality of the adult MEG device has not been systematically reviewed in infant epilepsy patients. The aim of this study is to quantify the probability of successful, clinically useful MEG recordings in infant epilepsy patients.Methods: We retrospectively reviewed the database of patients at our center who had undergone routine MEG examination for epilepsy between January 2008 and December 2014.Results: A total of 887 patients had simultaneous MEG/EEG recorded by a whole-head neuromagnetometer, with 204 planar gradiometers and 102 magnetometers (Vectorview, Elekta, Helsinki, Finland). Eight patients were less than 2 years old (infants), while 585 of them were 18 years or older (adults) at the time of MEG examination. The youngest patient recorded was 9 months (the figure shows an infant patient lying comfortably on the MEG bed accompanied by a parent). In 4 of 585 adult patients, the analysis was technically difficult due to a failure of the head position indicator measurement apparatus or artifacts from metallic implants. All the infant MEG records were processed with a Signal Space Separation (SSS) or a temporally extended SSS (tSSS) algorithm to remove environmental artifacts. Six of 8 were also processed with movement compensation (MC) processing using continuous head position monitoring. In all the infant patients, the brain activity during both awake and sleep state was satisfactorily recorded without sedative medication and without complications. MEG captured epileptic abnormalities in all infants: interictal in 8 (100%) and ictal in 4 (50%) of the patients. The yield in adults was lower: interictal in 399 (69%) and ictal in 37 (6.4%). Despite the relatively low number of tiny patients, Fisher's exact probability test showed that infants had significantly more frequent ictal abnormalities than adult patients. Half of the infant patients (4) underwent epilepsy surgery at our center after MEG. In 3 of them (1 frontal lobectomy, 1 lesionectomy, and 1 hemispherectomy) the MEG findings provided important help for planning of surgery, and their seizures were reduced postoperatively. In the other patient, MEG showed generalized epilepsy, a frontal lobectomy was performed, but the patient's severe seizures continued.Conclusions: The customary MEG device successfully records epileptic activity in infants, facilitated by monitoring the patients' head position and postprocessing with SSS or tSSS, with or without MC. Interictal and ictal MEG abnormalities were more frequent in infants than in adult patients. MEG is safe and useful to guide the therapeutic management of infant epilepsy patients using the conventional machine without special adaptations for small heads. 1. Okada et al. Review of scientific instruments 2006;77(2):024301 2. Roberts et al. Front Hum Neurosci 2014;8(99):1-10 Sources of funding none