Electroencephalographic study of sleep architecture in infants at four months
Loading...
Files
Full Text E-thesis
Date
2024
Authors
Ventura, Soraia
Journal Title
Journal ISSN
Volume Title
Publisher
University College Cork
Published Version
Abstract
Early infancy is a period of high neuroplasticity. Infants spend a large portion of their time asleep, and sleep reflects and contributes to neurodevelopment changes. The study of the EEG features of infant sleep provides a unique window on brain maturation and offers a number of opportunities to benefit infant development, including measures of normal and of divergent development due to factors such as premature birth, enabling early intervention; and also, as outcome biomarkers for interventions to optimise development, such as routine massage.
Aims and Objectives:
This thesis aims to deepen our knowledge of infant sleep EEG characteristics at 4 months of age and their utility in assessing neurodevelopment. It is divided into four main objectives. 1) To explore the normative values of sleep macrostructure and sleep spindles of term-born infants. 2) Examine whether routine massage is associated with physiological changes in the sleep EEG. 3) To establish the ability of sleep EEG markers at 4 months to predict neurodevelopmental scores at 18 months. 4) To compare sleep biomarkers of moderate to late preterm-born infants at corrected age with their term-born peers.
Methodology:
The cohorts analysed include ex-term infants randomised into intervention (4-month daily parent-led massage routine) and non-intervention (care-as-normal) groups (BabySMART cohort) and a healthy moderate-to-late preterm group (SLEEPi cohort) recruited at Cork University Maternity Hospital study.
The 4-month multichannel video-EEG polysomnography consisted of 21 EEG electrodes following the 10-20 electrode placement system, and extra polygraphic channels. Griffiths Scales of Child Development, third edition (Griffiths III) assessments were performed at 4 and 18 months and sociodemographic questionnaires for BabySMART, at 2 weeks, 4 and 18 months.
Initial analysis of the sleep EEG included manual annotation of fronto-central sleep spindles (a feature of sleep microstructure) and sleep stages (sleep macrostructure). Sleep macrostructure, spindle number, duration and density were calculated manually. Spindle power, frequency and synchrony were calculated with MATLAB R2020a. Brain Symmetry Index, Coherence and Spectral power of the background quantitative EEG features were calculated using the bespoke NEURAL software. Descriptive statistics and statistical analysis were calculated with IBM SPSS statistics 26.
Results:
1) I describe the normative values for sleep macrostructure and sleep spindles at 4 months. I found sex differences in spindle spectral power and BSI, and that spindle spectral power decreased from the 1st to the 2nd sleep cycle.
2) A 4-month routine of infant massage was associated with physiological changes reflected in sleep EEG features: higher EEG power, lower coherence and higher sleep spindle spectral power.
3) Sleep latencies, duration of non-rapid eye movement stage 3 of sleep (N3), spindle duration and synchrony, delta power, and coherence at 4 months were associated with neurodevelopment at 18 months.
4) The moderate-to-late preterm-born group showed changes in sleep parameters at 4-months corrected age compared to term peers and had a shorter N2 duration, higher delta power and higher frontal connectivity and lower connectivity between bilateral parietooccipital areas.
Conclusions:
We characterised normative data on sleep EEG biomarkers using the largest sample to date at four-months of age, providing a baseline for future studies of neurodevelopmental trajectory. Sleep spindle differences were observed between sexes and across sleep cycles, confirming and extending knowledge from older paediatric cohorts. I show for the first time that parent-led massage is associated with a more mature developmental pattern of the EEG of healthy term-born infants. Using a wide range of sleep features, I identified those relevant for the prediction of neurodevelopmental outcomes at 18 months. Lastly, I showed differences of sleep spindles in moderate-to-late preterm infants at 4-months corrected age relative to term-born peers, denoting altered brain organisation.
Description
Keywords
EEG , Polysomnography , Sleep spindles , Sleep macrostructure , Brain connectivity , Spectral power , Infancy , Prematurity , NREM , REM , Brain maturation , Sleep , Infant massage , Environmental enrichment
Citation
Ventura, S. 2024. Electroencephalographic study of sleep architecture in infants at four months. PhD Thesis, University College Cork.