OBJECTIVETo explore the clinical and neuroelectrophysiological characteristics of myoclonus of different origins in children.

METHODThirty-two children with myoclonic seizure were analyzed by video electroencephalogram-electromyogram (VEEG-EMG) polygraphic recordings, jerk-locked back averaging (JLA) and short latency somatosensory evoked potential (SSEP). They were classified into cortical myoclonus (CM), subcortical myoclonus (SCM), and unidentified group according to their generating locations, and also were classified into epileptic and non-epileptic myoclonus based on their different properties.

RESULTThe 32 patients included 14 with CM, 14 with SCM and 4 with unidentified origin. (1) CM group: the myoclonic patients presented as focal and/or multifocal seizures in 11 cases and as generalized in another 3 patients besides focal myoclonus. Arrhythmic jerks were shown completely in 11 cases and rhythmic seizures were concomitant in another 3 patients. Myoclonus sensitivity to sensory stimulus was observed in 10 patients. The durations of EMG burst were 10-52 ms. Background EEGs were presented normal in 4 patients and slowing in 10 patients. The epileptiform discharges in interictal EEG were variable. The ictal EEG showed epileptic discharges with each clinical jerk in 9 cases but only with some jerks in 4 patients. Another one had no any EEG abnormality in each jerk. The myoclonus-related spikes were disclosed in 13 cases by JLA. Of the 10 cases who underwent SSEP, giant SSEPs were seen in 3 cases including the one with normal EEG and JLA analyses. (2) SCM group: myoclonus was presented as generalized in 8 cases and as focal in 6 cases. All the patients showed arrhythmic jerks and 14 cases were not sensitive to stimulus. The durations of EMG burst were from 60 ms to 400 ms. Normal background EEGs were presented in 6 patients and slowing in 8 patients. The interictal EEG showed no consistent abnormality. Epileptic discharges associated with myoclonus seizures were not found in any of 9 patients but were observed with some seizure changes in 5 cases. There was no myoclonus-related spike by JLA in this group. SSEPs were normal in all patients. (3) The group with unidentified origin: the durations of EMG were from 60 ms to 400 ms, and their EEG and SSEP recordings were normal. In addition, 32 patients could be classified as epileptic myoclonus in 14 cases and nonepileptic myoclonus in 18 cases by the polyneurophysiological tests.

CONCLUSION(1) It is not reliable to identify myoclonus seizures and their clinical properties depending on their interictal and ictal EEGs only. (2) Polyneuroelectrophysiological tests, including EEG-EMG, JLA, and SSEP, seem to be valuable and useful to identify the generating locations and properties for different myoclonus in children.