Spectral analysis of electroencephalography in premature newborn infants: normal ranges

Continuous EEG monitoring has not been used widely in neonatal intensive care, especially in the care of extremely premature infants, probably in part because of a lack of a reliable quantitative method. The purpose of this study was to quantify the EEG of the very premature infants just after birth by using spectral analysis and to describe the characteristics of the spectral signal when infants were clinically stable. Digital EEG recordings were performed on 53 infants who were < or =30 wk gestation for 75 min each day during the first 4 d after birth. Artefact was rejected manually after visual inspection of trace. The EEG was analyzed by manual measurement of interburst interval and automatically by spectral analysis using Fast Fourier Transformation. Spectral analysis generated the normal ranges of the relative power of the delta (0.5-3.5 Hz), theta (4-7.5 Hz), alpha (8-12.5 Hz), and beta (13-30 Hz) frequency bands, spectral edge frequency, and symmetry. The median (range) relative power of the delta band increased significantly from 68% (62-76%) on day 1 to 81% (72-89%) on day 4 (p=0.001). The interburst intervals became progressively shorter between days 1 [14s (10-25)] and 3 [8s (6-12)]; there were no significant differences between days 3 and 4. The relative power of the delta band seemed to be the most useful and repeatable spectral measurement for continuous long-term monitoring. However, automatic artefact rejection software needs to be developed before continuous quantitative EEG monitoring can be used in the neonatal intensive care environment.