Lowered electroencephalographic spectral edge frequency predicts the presence of cerebral white matter injury in premature infants.

Abstract

Current methods for early identification of cerebral white matter injury in the premature infant at the bedside are inadequate. This study investigated the utility of advanced spectral analysis of the neonatal electroencephalogram (EEG) in the early diagnosis of white matter injury in the premature infant. The critical measurement used, suggested largely by previous studies in animal models, was the spectral edge frequency (SEF), calculated here as the frequency below which 90% of the power in the EEG exists. Fifty-nine very low birth weight infants (87% of eligible infants) had electrodes placed over the central and parietal regions (C3, P3, C4, and P4 sites according to the 10-20 international system) for the collection of EEG amplitude, intensity, and SEF. All averaged signals were analyzed off-line using software (Chart Analyzer; BrainZ Instruments, Auckland, NZ). All infants had a magnetic resonance imaging scan at term to identify the presence and severity of white matter injury. There was no significant difference between conventional EEG amplitude and intensity for infants with or without evidence of white matter injury. However, premature infants with increasingly severe white matter injury had progressively lower SEFs compared with infants who did not exhibit white matter injury. These data suggest that SEF-based measures are useful for defining the presence and severity of white matter injury at the bedside.