Dear Sirs,
We read the recent paper by Inder et al. (1) with great interest. The
authors claim that the development of white matter injury in premature
infants can be detected by a reduction of the value of the parameter
Spectral Edge Frequency (SEF). SEF is derived from a spectral analysis of
frequencies of the EEG signal and is defined as the frequency below which
a predefined amount of the activity resides (in this paper 90 %).
We have also endeavoured to find better ways of monitoring central
nervous (CNS) function (2). We have, in particular, taken notice of the
fact that the EEG more often than not is discontinuous in preterm
neonates. Discontinuity can be both physiological and pathological.
Discontinuity is more pronounced the more immature the neonate but also
escalates with increased degree of pathology and increased pharmacological
load (3).
In the paper by Inder et al., SEF has been calculated from EEGs
recorded in premature neonates in which later no- or different degrees of
white matter injury was diagnosed. In these, the EEG can be assumed to be
discontinuous. No information is given as to how the EEG as such was
evaluated and how the degree of discontinuity was taken into account in
the spectral analysis and the calculation of the SEF. If these factors
were not considered, the main finding of the study i.e. a reduction of SEF
in relation to the degree of later diagnosed white matter injury, may be a
result of increasing amounts of discontinuous EEG in the neonates in
proportion to the degree of injury. An indication in this direction may be
the observation of an extremely low EEG intensity in the most
disadvantaged group. Since pharmacological effects may be at play, the
discrepancy concerning the number of neonates receiving morphine infusion
given in the Methods and Result sections (five and all respectively) makes
an evaluation even more difficult.
We believe that improved methods for monitoring of the CNS in
neonates are needed. However, both during development of such methods and
when new methods have come into practice, it is important to be able to
control and evaluate the original signal.
Magnus Thordstein
MD, PhD
Dept. Clinical Neuroscience
Göteborg University
Sweden
Ingemar Kjellmer
Professor
Dept. Pediatrics
Göteborg University
Sweden
Nils Löfgren
MSc, Electrical engineering
Dept. Signals and Systems, Chalmers University of Technology, Göteborg
Sweden
References
1. Inder TE, Buckland L, Williams CE, Spencer C, Gunning MI, Darlow
BA, Volpe JJ Gluckman PD. Lowered electroencephalographic spectral edge
frequency predicts the presence of cerebral white matter injury in
premature infants. Pediatrics 2003; 111: 27-33
2. Thordstein M, Bågenholm R, Andreasson S, Ouchterlony J, Löfgren N,
Göthe F, Nivall S, Hedström A, Lindecrantz K, Kjellmer I, Wallin B G.:
Long term EEG monitoring in neonatal and pediatric intensive care. In
Ambler Z, Nevsimalova S, Kadanka Z, Rossini PM. (eds): Clinical
Neurophysiology at the Beginning of the 21st Century. Clinical
Neurophysiology Suppl 53, p 76-83. Elsevier Science BV, 2000
3. Scher MS. Electroencephalography of the newborn: Normal and
abnormal features. In Niedermeyer E and Lopes da Silva F (eds):
Electroencephalography; Basic principles; Clinical applications and
related fields. p 896-946, Lippincott, Williams & Wilkins, 1999
eLetters: ![[Read eLetters]](/icons/misc/sectionDOWN.gif){kind=icon}
