Prognostic value of neonatal discontinuous EEG

The burst suppression pattern on the neonatal electroencephalogram (EEG) is associated with a poor outcome. However, this serious abnormality constitutes only a small proportion of discontinuous neonatal EEGs. We sought to establish whether any easily measurable parameters among the broad range of excessively discontinuous neonatal EEGs are predictive of outcome. We retrospectively reviewed the EEGs and medical records of 43 term infants with excessively discontinuous EEGs. We quantitated 10 parameters in the bursts and interburst intervals, among them the predominant interburst interval duration (defined as the duration of more than 50% of all interburst intervals of an EEG). Univariate and multivariate analyses were performed on the 10 EEG variables in relation to neurologic outcome and subsequent epilepsy. Based on multivariate analysis, a single easily measurable EEG parameter related significantly to outcome. A predominant interburst interval duration of more than 30 seconds correlated with the occurrence of both unfavorable neurologic outcome and subsequent epilepsy (P = 0.040 and P = 0.033, respectively). In conclusion, a infant whose EEG contains a predominant interburst interval duration of more than 30 seconds has a 100% probability of experiencing severe neurologic disabilities or death and an 86% chance of developing subsequent epilepsy. This easily quantitated EEG parameter could be valuable for the early estimation of neurologic prognosis. (C) 2002 by Elsevier Science Inc. All rights reserved.