Developmental comparison of sleep EEG power spectral patterns in infants at low and high risk for sudden deathComparison au cours du développement des spectres de puissance de l'EEG du sommeil chez des nourrisons à haut et bas risque de mort subite
Abstract
Power spectral density measured from central cortical EEG recordings during sleep was compared in two groups of 20 infants each at intervals between 1 week and 6 months of age. One group consisted of siblings of infants who had previously died of the sudden infant death syndrome, as determined by history and autopsy report (sibling group). The second group was made up of age, sex and socioeconomically matched infants with no familial history of sudden infant death (control group).
All-night, 12 h polygraphic recordings were obtained and classified into sleep and waking states according to established scoring criteria. Left and right bipolar central cortical EEG tracings for continuous 10 min periods were sampled during the first, middle and last quiet sleep (QS) and active sleep (AS) epochs of the night. Sequential 16 sec segments of sampled EEG data were digitized, subjected to FFT, corrected, transformed and sorted in terms of power into adjacent 4 c/sec bands between 0 and 19 c/sec. Statistical evaluation was focused on sample, age and group differences in both QS and AS.
Sample effects were noted in both groups and in relation to both sleep states. During QS a consistent increase in power focused at 4–11 c/sec was observed in the first epoch of the night after 8 weeks of age. This emergent ‘time-of-night’ difference was more generalized in left hemisphere and more pronounced at 8 weeks in the sibling group. Evidence cited suggested that this effect could be mediated by a maturing circadian modulation of other physiological systems. Sample effects during AS were present at 1 week and disappeared after 16 weeks of age. This effect, which consisted of a generalized increase in EEG power during the final AS epoch of the night, was observed more extensively over the right hemisphere and was comparable in both groups.
Characteristic changes in power occurred within specific frequency bands as a function of age in both groups. During QS primary increments in power were noted in the 0–3 and 12–15 c/sec,bands, the latter preceded by a postnatal decrease. The largest developmental increase in power observed was at 12–15 c/sec, and appeared between 4–8 weeks in siblings and 8–12 weeks in controls. Since this frequency band is associated with the EEG spindle pattern during QS, observed changes were interpreted in terms of the maturation of established thalamocortical substrates for this pattern. Power spectral density changes in AS were minimal, except for the 4–7 c/sec band, which showed changes indicative of a distinctive maturation in certain elements of the EEG frequency spectrum studied here. In quiet sleep differential functions were noted for bands corresponding to the so-called delta (0–3 c/sec) and theta (4–7 c/sec) frequencies labeled initially by Davis et al. (1938). Changes in higher frequency bands were less independent and appeared to be primarily focused in the 12–15 c/sec (sigma) range. In fact, the greatest change registered with age in any frequency band during QS was at 12–15 c/sec, corresponding functionally to the appearance of spindle activity in the sleep EEG. Spectral density changes during active sleep were generally less remarkable, with one exception. Activity at 4–7 c/sec showed a dynamic developmental sequence, similar in many respects to that seen in the 12–15 c/sec band during QS.
The similarity between developmental patterns at 12–15 c/sec during QS and 4–7 c/sec in AS is worthy of note. Numerous findings have attributed rhythmic EEG activity in sensorimotor cortex to intrinsic thalamocortical circuits in the somatosensory system. When rhythmic activity is present a ‘gated’ or oscillatory discharge is thought to result from the absence of movement and the corresponding stability of somatosensory afferent discharge (for review see Creutzfeldt 1974). The parallel development of these two rhythmic frequencies in the two different sleep states suggests that both are dependent upon the maturation of similar or perhaps identical thalamocortical connections. Moreover, their presence in either sleep stage may signal a common condition of somatosensory ‘stability’. The different frequencies manifested under this condition could be a result of established differences in thalamic and cortical excitability during the two sleep states.
Virtually every EEG power spectral measure obtained in the present study indicated that changes occurred earlier in the sibling group. This outcome was confirmed by both orthogonal and rate-of-change statistical analyses. In relation to the primary developmental characteristics outlined above, the difference can best be described as a 1 month advance in normal developmental sequencing. Several frequency components which peaked or stabilized at 12 weeks of age in the control group did so at 8 weeks in the siblings, particularly in left hemisphere data. Additionally, the key rhythmic EEG frequencies in sleep, namely 12–15 c/sec during QS and 4–7 c/sec in AS, both showed advanced patterns of maturation in the sibling group. The sample effects described above also emerged earlier in the sibling group, at least during QS. This finding is consistent with evidence indicating an earlier circadian effect on autonomic patterns during QS in these infants (Hoppenbrouwers et al. 1980a; Hoppenbrouwers 1981) and supports the conclusion that the EEG sample effect in QS was related to circadian influences. Finally, it should be noted that a new set of changes was documented in siblings between 16 and 24 weeks of age, after a period of relative stability starting at 8 weeks. Control infants showed stability only after 12 weeks of age and generally failed to demonstrate this later shift.
Collectively, these findings suggest that central nervous system maturational sequences are accelerated in siblings of SIDS infants. There are several possible explanations for such a conclusion. First, numerous studies have suggested that risk for SIDS is associated with a mild chronic hypoxia beginning early in life and leading to a number of adaptive physiological responses (for review see McGinty and Sterman 1980). The high-risk sibling group studied here showed elevated respiratory and cardiac rates in all states, a greatly reduced incidence of apnea and a variety of other characteristics consistent with this interpretation (Hoppenbrouwers et al. 1980b; Harper et al. 1978, 1981). It is possible that factors directly or indirectly associated with this adaptation result in an acceleration of CNS development. Animal studies have shown clearly that CNS maturation can be altered by both behavioral and similar to those seen for 12–15 c/sec activity in QS. This similarity suggested maturation of a common or parallel substrate. Characteristic changes during AS also appeared earlier in siblings than controls.
The observed advance in the appearance of many key developmental features of the sleep EEG among siblings was attributed to accelerated CNS maturation in this group. It was suggested that this acceleration could result either as an adaptive response to a suspected congenial, mild hypoxia in siblings or as a consequence of altered parenting in families who had experienced a sudden infant death.
Résumé
La densité de puissance spectrale mesurée à partir des enregistrements EEG corticaux rolandiques au cours du sommeil a été comparée dans deux groupes de 20 nourrissons chacun, vus à des intervalles de 1 semaine à 6 mois. L'un des groupes est constitué par les frères et soeurs de nourrissons précédemment morts du syndrome de mort subite du nourrison, établli par l'histoire et le rapport d'autopsie (groupe ‘fratrie’). Le deuxième groupe est constitué de bébés appariés en ce qui cocerne l'âge, le sexe et le niveau socioéconomique, sans aucune histoire familiale de mort subite du nourrisson (groupe ‘contrôle’).
Des enregistrements polygraphiques de toute la nuit, durant 12 h sont obtenus et classés en stades de sommeil et de veille suivant les critères de codage établis. Des périodes continues de 10 min des tracés EEG rolandiques bipolaires droit et gauche ont étééchantillonnées au cours de l'époque initiale, médiane et terminale du sommeil calme (SC) et du sommeil actif (SA). Des segments successifs de 16 sec de données EEG échantillonnées ont été numérisées, soumises à transformée rapide de Fourrier, corrigées, transformées et sorties sous forme de spectres de puissance de bandes contiguës de 4 c/sec, de 0 à 19 c/sec. L'évaluation statistique a été centrée sur les différences suivant les échantillons, l'âge et les groupes, portant sur le SC et le SA.
Des différences suivant les échantillons ont été notées dans les deux groupes, et pour les deux stades de sommeil. Durant le SC, une augmentation constante de la puissance centrée autour de 4 à 11 c/sec s'observe dans la première époque de la nuit après l'âge de 8 semaines. Cette première différence, liée au moment de la nuit, est plus généralisée pour l'hémisphère gauche et plus prononcée à 8 semaines dans le groupe ‘fratrie’. Cette donnée suggère que cet effet pourrait être médiatisé par une modulation circadienne de la maturation d'autres systèmes physiologiques. Des différences suivant les échantillons au cours du SA s'observent à 1 semaine et disparaissent après 16 semaines. Ces effets, qui consistent en augmentation généralisée de la puissance EEG lors de la dernière époque de SA de la nuit s'observent de façon plus marquée sur l'hémisphere droit et sont comparables dans les deux groupes.
Des modifications caractéristiques de puissance surviennent à l'intérieur de bandes spécifiques de fréquence en fonction de l'âge dans les deux groupes. Au cours du SC les augmentations initiales de puissance s'observent dans les bandes 0 à 3 et 12 à 15 c/sec, les dernières étant précédées par une diminution à la période postnatale. Les plus grandes augmentations de puissance observées au cours du développement se produisent à 12–15 c/sec, et apparaissent entre 4 à 8 semaines dans le groupe ‘fratrie’ et entre l'âge de 8 à 12 semaines chez les ‘contrôles’. Puisque cette bande de fréquence est liée aux fuseaux EEG au cours du sommeil tranquille, les modifications observées sont interprétées en termes de maturation des substrats thalamo-corticaux dont on sait qu'ils sont à l'origine de ce pattern. Les modifications de densité de puissance spectrale au cours du SA sont minimes, à l'exception de la bande de 4 à 7 c/sec qui montre des modifications similaires à celles que l'on observe pour l'activité de 12 à 15 c/sec au cours du SC. Cette similarité suggère la maturation d'un substrat commun ou parallèle. Les modifications caractéristiques au cours du SA apparaissent également plus précocément dans le groupe ‘fratrie’ que chez les ‘contrôles’.
L'avance que l'on observe dans l'apparition de plusieurs données-clés du développement de l'EEG du sommeil parmi le groupe ‘fratrie’ est attribuée à une accélération de la maturation du système nerveux central dans ce groupe. Les auteurs suggèrent que cette accélération pourrait résulter soit d'une résponse adaptive à une légère hypoxie congénitale suspectée chez la fratrie ou comme une conséquence d'une perturbation de l'attitude familiale dans les familles qui ont déjà eu l'expérience d'une mort subite du nourrisson.
References (33)
- M.W. Church et al.
Changes in frequency and amplitude of delta activity during sleep
Electroenceph. clin. Neurophysiol.
(1975) - I. Feinberg et al.
Period and amplitude analysis of 0.5–3 c/sec activity in NREM sleep of young adults
Electroenceph. clin. Neurophysiol.
(1978) - M.B. Sterman et al.
Quantitative analysis of infant EEG development during quiet sleep
Electroenceph. clin. Neurophysiol.
(1977) - R. Bickford et al.
The compressed spectral array (CSA): a pictorial EEG
- A.A. Borbély
Sleep: circadian rhythm versus recovery process
- O.D. Creutzfeldt
The neuronal generation of the EEG
- H. Davis et al.
Human brain potentials during the onset of sleep
Neurophysiology
(1938) - I. Feinberg et al.
Flurazepam effects on sleep EEG
Arch. gen. Psychiat.
(1979) - P. Froggatt
Epidemiologic aspects of the Northern Ireland study
- H. Gastaut et al.
On the significance of ‘wicket rhythms’ (rythmes en arceau) in psychosomatic medicine
Electroenceph. clin. Neurophysiol.
(1954)
Computer analysis procedures
Polygraphic studies of normal infants and infants at risk for the sudden infant death syndrome: heart rate and variability as a function of state
Pediat. Res.
Periodicity of sleep states is altered in infants at risk for the sudden infant death syndrome
Science
Clinical characteristics of subsequent siblings of SIDS
Long term, continuous monitoring of multiple physiological parameters in newborn and young infants
Acta paediat. scand.
Ontogenesis of sleep and waking
Cited by (26)
Spontaneous smiling and facial mimics in the neonate: A contribution to the functional role of paradoxical sleep during development
2010, Medecine du SommeilLe visage du nouveau-né endormi révèle l’existence de mimiques qui donnent l’impression de joie ou d’émotions plus négatives. Parmi celles-ci, les sourires et les froncements de sourcils sont les plus facilement reconnus.
Essayer de comprendre la signification neurophysiologique des mimiques faciales observées au cours du sommeil. Amener des arguments en faveur des hypothèses concernant les fonctions du sommeil paradoxal au cours de l’ontogenèse.
Douze nouveau-nés normaux nés à terme, trois nourrissons porteurs d’une malformation cérébrale majeure : deux hydranencéphales, un nourrisson ayant subi une hémispherectomie gauche, ont été enregistrés pendant une période de trois heures par enregistrement vidéo et polygraphique de sommeil simultanés. Il a été calculé la fréquence, par 100 minutes de SA, des sourires et froncements de sourcils, leur modalité d’occurrence et leur distribution dans chaque période de sommeil agité (SA). Il a été déterminé pour les sourires leur durée, leur latéralité, leur relation avec les modalités d’occurrence des mouvements oculaires rapides. Les corrélations entre sourires, froncements de sourcils et mouvements corporels globaux.
Chez les nouveau-nés, 377 sourires et 1039 froncements de sourcils ont été analysés au cours du SA. Le sourire néonatal est identique au futur sourire social. Les sourires unilatéraux gauches sont significativement plus fréquents que les sourires unilatéraux droits (P < 0,0001). La durée des sourires est corrélée avec la densité des mouvements oculaires rapides survenant en salves (P < 0,01). Aucune corrélation n’a été trouvée entre les sourires et les mouvements corporels globaux ou les stimulations. Chez les nourrissons porteurs d’une malformation cérébrale, il existe une absence totale de mimiques faciales lorsque l’hydranencéphalie est totale, une absence de froncements de sourcils lorsqu’elle est partielle, pas d’asymétrie pour les sourires chez le nourrisson hémisphérectomisé.
Les données obtenues chez les nouveau-nés normaux et les altérations observées chez les nourrissons présentant une malformation cérébrale appuient les hypothèses d’un rôle du SA dans la programmation génétique des mimiques faciales. Cette programmation serait générée à partir du tronc cérébral avec une possible modulation par les structures corticales.
Sleeping newborns display facial mimics which give the impression of negative or positive affects. Smiling and frowning are the simplest and easily recognized ones.
This study is an attempt to understand the neurophysiological meaning of these mimics and to define their connection with the functional role of REM sleep during ontogenesis.
Twelve healthy, full term neonates; three infants with brain malformations: two hydranencephalics and one with a left hemispherectomy were recorded, for a period of 3 hours during sleep, with the simultaneous use of a polygraphic and a video recording. It was determined during active sleep the frequency of smiling and frowning, their modality of occurrence and their distribution within each active sleep. It was calculated for smiling their duration and laterality, their relationship with rapid eye movements and gross body movements.
In the healthy neonates, 377 smiles and 1039 smiles were analyzed during active sleep. Spontaneous smiling is found to be similar to adult social smiling. Smiles were significantly more intensely expressed on the left side of the face (P < 0.0001). Their duration was correlated with the density of REMs occurring in burst (P < 0.1). No correlation was found between smiling or frowning and gross body movements or stimulations. In the infants with brain malformation there was no facial mimics in the case of total hydranencephaly, no frowning in that of partial hydranencephaly and no smiling asymmetry in the infant with left hemispherectomy.
The data obtained in the healthy neonates and the alterations observed in the infants with brain malformation support the hypotheses of a role of active sleep in the genetic programming of facial mimics. Such programming would be generated in the brainstem with a possible modulation at cortical level.
Temperature differences during sleep between fullterm and preterm neonates at matched post-conceptional ages
2003, Clinical NeurophysiologyObjective: Altered physiologic behaviors during sleep have been described for healthy preterm neonates at post-conceptional fullterm ages. These differences may reflect brain dysmaturity as a result of conditions of prematurity. The present study examines if differences in state-specific temperature changes exist in a healthy preterm cohort as another expression of brain dysmaturity.
Methods: Rectal and skin temperatures during sleep state transitions are reported in 59 asymptomatic post-conceptional age term infants, comparing 25 full term and 34 preterm infants. Three-hour 24-channel electroencephalogram (EEG)-sleep studies were recorded for each child. One of 4 sleep states were assigned for each of 7339 min, based on both cerebral and non-cerebral measures. For each study, average rectal and skin temperatures for each sleep state were calculated. Repeated measures MANOVA were performed using 4 explanatory variables, average skin and rectal temperatures and variance of rectal and skin temperatures, comparing preterm/fullterm status and 4 sleep states.
Results: Rectal temperature differences between neonatal cohorts during specific sleep states were noted: F=8.58, P<0.0001. Significant differences were noted for both average and variance of rectal temperatures during all 4 sleep states with higher temperatures in the preterm group. For all neonates, both skin and rectal temperature differences were also noted among sleep states (F=4.22, P<0.0004). Differences were specifically noted between two specific EEG segments, mixed frequency active sleep and tracé alternant quiet sleep (P<0.0004).
Conclusions: In summary, significant differences in temperatures were noted across sleep state transitions for two neonatal cohorts, with higher average rectal temperatures in the preterm cohort. These findings highlight an altered measure of brain function during sleep in preterm infants affecting temperature regulation. This altered physiologic behavior reflects adaptation of the infant's brain function to conditions of prematurity which may contribute to vulnerabilities at older ages.
The meaning of character combination for the separation of EEG-data
2001, Theory in BiosciencesStandard EEG risk evaluation works on scoring systems that use different types of questionnaires. Here, an alternative for SIDS (Sudden Infant Death Syndrome) risk detection is presented that is based exclusively on EEG data which possibly could substitute the procedure of questioning the parents and allow a direct qualification of the physiological disposition of the individual neonate: Using EEG-characters an approved SIDS-case could be discriminated as well against the group of “healthy” infants as against the “high-risk-group”. The results of this study may confirm the evidence that the EEG analysis can be a promising approach to predict an increased SIDS risk.
Aberrant temporal patterning of slow-wave sleep in siblings of SIDS victims
1995, Electroencephalography and Clinical NeurophysiologyWe assessed the patterning of slow-wave EEG activity during sleep in siblings of sudden infant death syndrome (SIDS) victims over the first 6 months of life. Twelve hour overnight physiologic recordings were obtained from 25 apparently healthy subsequent siblings of SIDS victims and 25 control infants at 1 week, and 1, 2, 3, 4 and 6 months of age. The EEG activity was electronically bandpass filtered, leaving primarily activity ranging from 0.5 to 2.5 Hz (the delta frequency), and the filtered traces were full-wave rectified and integrated over 1 min periods. The recordings were divided into four 3 h segments beginning at sleep onset, and the mean integrated delta activity during quiet sleep was determined for each segment of the night. At 3 and 4 months postnatal age, SIDS siblings displayed increased integrated delta amplitude in the early morning hours relative to control infants. Most SIDS deaths occur in the early morning hours during the 2–4 month age range. We thus speculate that increased delta activity may be indicative of increased arousal thresholds in the early morning, which may contribute to SIDS deaths.
Automatic analysis of sleep - wake states during the first year of life
1992, Neurophysiologie Clinique / Clinical NeurophysiologyL'analyse automatique des tracés de sommeil du nourrisson, considérée d'abord comme une simple aide à la réalisation rapide d'un hypnogramme a longtemps paru plus difficile que chez l'adulte en raison des modifications importantes de l'électrogenèse au cours de la première année de la vie, et donc de la difficulté de s'adapter aux critères de Rechtschaffen et Kales (1968) même modifiés ou de Anders et al (1971). En fait, ces travaux ont fait apparaître des aspects nouveaux : l'analyse automatique de l'EEG dans le sommeil permet d'observer plus facilement et de façon plus objective que par l'analyse visuelle, les modifications de l'électrogenèse. De plus, les analyses automatiques des différents paramètres, fluctuant avec les stades de vigilance et de sommeil, mettent parfaitement en évidence la continuité de ces fluctuations, et donc l'aspect un peu artificiel des classiques hypnogrammes. Ces informations traitées mathématiquement facilitent la visualisation des résultats et permettront à l'avenir l'interprétation de la maturation des organisations circadiennes et ultradiennes qu'il était difficile de formaliser aussi nettement par analyse visuelle uniquement. Dans le domaine de la maturation des premiers mois de la vie, le traitement automatique des données conduira à s'adapter à une nouvelle forme de raisonnement.
Automatic analysis on infant sleep tracing was first considered as an aid to the rapid construction of hypnograms. It has long been thought more complicated than in adults, because of the significant changes in electrogenesis during the first year of life, resulting in difficulty to adapt to the criteria of Rechtschaffen and Kales (1968), even if modified, or to those of Anders et al (1971). In fact, these studies have shown that automatic analysis of sleep EEG tracings permit an easier analysis of the changes in electrogenesis according to subject age, sleep stage and time of night. Moreover, automatic analysis of parameters which change with the sleep-wake stages shows the continuity of these changes, and therefore the somewhat arbitrary nature of conventional hypnogram classification. The mathematical treatment of this information facilitates their visualisation, and permits a better analysis of circadian and ultradian variations that could scarcely be formalized by classical hypnogram analysis. In the study of maturation occurring during the first months of life, automatic signal processing will require adaptation that will lead to new forms of reasoning.
Functional role of REM sleep during ontogenesis
1992, Neurophysiologie Clinique / Clinical NeurophysiologyLes théories pouvant être impliquées dans le développement neurocognitif de l'enfant: le rôle du SP dans les processus de mémorisation et d'apprentissage et les hypothèses de Roffwarg et de Jouvet sont analysées et étayées sur l'étude: des mécanismes neurophysiologiques du SP, des expériences de privation de sommeil paradoxal (SP) chez l'animal immature, du développement du SP et du rêve chez l'enfant et sur une étude personnelle de la signification neurophysiologique des mimiques faciales observées au cours du sommeil chez le nouveau-né. La conclusion de cette analyse est qu'il semble que le sommeil agité (SA) puisse assez tôt, dans l'ontogenèse, être assimilé au SP, qu'il joue vraisemblablement pendant la période fœtale et postnatale immédiate un rôle important dans la mise en place de réseaux neuronaux probablement génétiquement programmés et peu spécifiques, qu'il pourrait intervenir dans l'engrammation des stimulations épigénétiques.
The hypotheses directly linked to cognitive and neurologic ontogenic processes ie consolidation of memory and learning, the maturation hypothesis of Roffwarg and the hypothesis of endogenous genetic programming of Jouvet, are analysed. The discussion of these theories are based on the analysis of: the neurophysiologic mechanism of REM sleep and its ontogenesis in human, the results of REM sleep deprivation in young animals and by a personal study of facial mimics during sleep in neonates. Active sleep could be assimilated, very early during ontogenesis, to REM sleep, it probably plays an important role in brain maturation during early development but the stimulation is probably, at this time, not very specific, later it could be a link between genetic programming and epigenetic processes.