Review/Mise au pointElectroencephalography in premature and full-term infants. Developmental features and glossaryÉlectroencéphalographie du nouveau-né prématuré et à terme. Aspects maturatifs et glossaire
Section snippets
Historical background
By D. Samson-Dollfus, 1999 [41]
In 1933, Berger [9] claimed that no cerebral electrical activity was apparent before the age of 35 days. Five years later, however, Smith [83], [84], [85] showed that the tracings recorded in full-term neonates displayed 4 to 7 Hz frequencies, mixed with slow waves (2 or 3 Hz) and 12–14 Hz rhythms. In 1942, Lindsley [42] published the tracing recorded in a fetus of gestational age 7 months, recorded through the mother's abdominal wall. Between 1948 and 1953, Hugues
Definitions
Gestational age (GA): number of weeks from the first day of the woman's last menstrual cycle to the birth date.
Conceptional or corrected age (CA): GA at birth plus week number postpartum.
Full-term neonate: neonate born after 38 weeks of amenorrhea.
Premature infant: infant born before 37 weeks of amenorrhea.
The infants born at 37 and 38 weeks are usually considered as having the same clinical and neurological characteristics as the completely full-term newborns, but their EEG show some
Recording methods
Neonatal EEG recording should disturb the infant's behaviour and standard care as little as possible. The EEG equipment must be carefully prepared in advance to minimize the time required for electrode positioning and to avoid interference during recording (for example, opening the incubator). The EEG/polysomnography recorder should be positioned close to the infant, allowing the technician to report behavioural observations and/or to make video recordings.
Interpretation requires knowledge of
Recognition of sleep and wakefulness
Prechtl [74] reported that “behavioural states are constellations of physiological and behavioural variables, which are repeated and remain stable over time not only in a given child, but also in different infants”.
Therefore, the analysis of behavioural states should never be based on a single variable.
In full-term infants, wakefulness is characterized either by agitation (active wakefulness [AW]), or by bright, wide-open eyes, without agitation (quiet wakefulness [QW]).
In daily practice,
Developmental features
The evolution of EEG and of behavioural features according to gestational age is presented in the following pages.
As EEG aspects are almost identical for the same GA and CA, no distinction is made in clinical practice.
The description of some aspects during maturation depends of the thresholds chosen for definitions, in particular for amplitude. They vary from study to study; so the values given for the durations of the activity and inactivity periods may be different [89] ; the main recent
Organization of behavioural states
- •
Alternate periods of activity and rest.
- •
Variations of EEG activity are inconsistently associated with behaviour.
Background activity
- •
Very discontinuous tracing; lability is present which leads to a variable discontinuity.
- •
Sequences of amplitude higher than 50 μV lasting less than 60 s.
- •
Sequences of 15 to 50 μV lasting around 20–25 s.
EEG features
- •
Mono- or diphasic delta waves. They are smooth, with very sparse superimposed theta or alpha rhythms. They are mainly of high amplitude (> 300 μV) and very low frequency (0.3–1 Hz).
- •
Theta
Organization of behavioural states
- •
Alternate periods with eye movements (less discontinuous EEG and higher amplitude) and periods without eye movements (discontinuous EEG and lower amplitude).
- •
Variations of EEG activity are inconsistently associated with behaviour.
Background activity
- •
Very discontinuous tracing; lability is present which leads to a variable discontinuity. In studies published after 1988, the longest interburst intervals last 29 to 46 s.
- •
Brief periods of semi-continuous tracing. The bursts of amplitude higher than 50 μV can last around
Organization of behavioural states
- •
Active wakefulness (AW).
- •
Active sleep (AS) with rapid eye movements.
- •
Quiet sleep (QS).
These are now well discernible.
Differentiation between AS and QS on EEG can be made from the discontinuous (QS) or continuous (AS) aspect, but not yet from EEG features.
Background activity
- •
Discontinuous tracing. Lability is present. The duration of the “hypoactive periods” (interburst intervals, IBI) is very variable (up to 30 s).
- •
Periods of continuity are longer (up to 160 s) and more numerous than before 28 weeks GA. They are more
Organization of behavioural states
- •
Active wakefulness (AW).
- •
Active sleep (AS).
- •
Quiet sleep (QS).
A good concordance is established between EEG activity and behavioural criteria.
Background activity
- •
AW: mainly artefacts are registered.
- •
AS: continuous or semi-continuous tracing.
- •
QS: discontinuous tracing; bursts of activity ≥ 3 s, periods of hypoactivity (IBI) ≤ 20 s.
EEG features
- •
Delta waves (0.7–2 Hz) have an amplitude of 100 to 200 μV. Smooth delta waves are no longer observed. Theta rhythms superimposed on delta waves persist.
- •
Delta waves with superimposed alpha or beta
Organization of behavioural states
AW, AS and QS are easily differentiated, but periods of quiet wakefulness (QW) are still rare.
Background activity
- •
AW: continuous activity with frequent artefacts.
- •
QW: continuous tracing.
- •
AS: continuous tracing identical to that for QW; the two states may be differentiated in terms of polygraphic patterns, or by visual observation.
- •
QS: discontinuous tracing: the burst duration lengthens between 32 and 34 weeks GA, with a concomitant decrease in the interburst interval. Despite lower values of interburst intervals
Organization of behavioural states
- •
AW, AS and QS.
- •
QW is clearly differentiated.
- •
Differentiation of two states of AS as of 36 weeks GA: AS 1 precedes QS, AS 2 follows QS.
Background activity
- •
AW and QW: continuous activity, activité moyenne (“polyfrequency activity”) appears at 36 weeks GA.
- •
AS 1: high-amplitude continuous tracing.
- •
AS 2: continuous tracing with more rapid activity and amplitude lower than that during AS 1.
- •
QS: still discontinuous or semi-discontinuous. The duration of periods of hypoactivity is < 10 s.
EEG features
- •
Delta waves with superimposed fast
Organization of behavioural states
- •
AS 1 and AS 2.
- •
QS.
- •
AW and QW are clearly differentiated.
Background activity
- •
AW, QW, AS 2: activité moyenne (“polyfrequency activity”).
- •
AS 1: continuous activity, mixed frequencies. The amplitude is higher, the activity is less rapid than during AS 2.
- •
QS: semi-discontinuous or sometimes tracé alternant.
EEG features
- •
Delta waves with superimposed fast rhythms (delta brushes) in all behavioural states, with a higher amplitude during QS.
- •
Clearly differentiated frontal transients during AS 1 and at the onset of QS.
- •
Slow anterior dysrythmia
Organization of behavioural states
AS 1, QS, AS 2, AW and QW are clearly differentiated.
Background activity
- •
AW, QW, AS 2: activité moyenne (“polyfrequency activity”)
- •
AS 1: continuous tracing, with decreased amplitude compared to the activity at 37–38 weeks GA (mixed frequencies).
- •
QS: tracé alternant and/or slow continuous tracing.
EEG features
- •
Delta brushes persist, but they are rare and only observed during QS.
- •
Frontal transients: identical to those observed at 37–38 weeks GA.
- •
Slow anterior dysrythmia: identical to that observed at 37–38 weeks GA.
- •
Appearance of
Glossary
Definitions of the main normal and pathological neonatal EEG aspects.
Normal EEG tracings
Figures from the previous sections illustrate some of these definitions.
Pathological EEG tracings
The diagnostic and prognostic value of the EEG in neonates is the result of detailed morphological analysis performed over many years, which has formed the basis for the following descriptions.
Some tracings and patterns frequently observed in neonatal pathology are highly suggestive of certain etiologies. However, the aim of this paper is not to report the correlations between EEG patterns and diseases. Should you so wish, please refer to the references for such correlations.
It should also be
Conflict of interest statement
None.
Acknowledgements
The authors are very grateful to Mickaël Walls for the revision of the English language.
References (95)
- et al.
Encéphalopathie myoclonique néonatale
Rev Electroencephalogr Neurophysiol Clin
(1978) - et al.
The EEG of the early premature
Electroencephalogr Clin Neurophysiol
(1985) - et al.
Densité et dynamique des encoches pointues frontales dans le sommeil du nouveau-né et du nourrisson
Rev Electroencephalogr Neurophysiol Clin
(1977) - et al.
The early diagnosis of periventricular leukomalacia in premature infants with positive rolandic sharp waves on serial electroencephalography
J Pediatr
(1998) - et al.
Background EEG activity in preterm infants: correlation of outcome with selected maturational features
Electroencephalogr Clin Neurophysiol
(1994) - et al.
Positive rolandic sharp waves in neonatal EEG; types and significance
Electroencephalogr Clin Neurophysiol
(1982) - et al.
Asymétrie EEG transitoire à l’entrée dans le sommeil calme chez le nouveau-né : étude sur 75 observations
Rev Electroencephalogr Neurophysiol Clin
(1984) - et al.
Essai d’évaluation des thérapeutiques (diazépam, phénobarbital) sur l’EEG néonatal pendant les premières 24 H du traitement
Rev Electroencephalogr Neurophysiol Clin
(1979) - et al.
Apport de l’EEG au diagnostic des hémorragies intra-ventriculaires du prématuré
Rev Electroencephalogr Neurophysiol Clin
(1972) - et al.
Hémorragie intraventriculaire et ischémie parenchymateuse chez le nouveau-né à terme. À propos de cinq cas
Neurophysiol Clin
(1997)