​The third trimester of pregnancy is a crucial phase of brain development, as this is the period during which development follows a highly orchestrated sequence of neurobiological processes, including neuronal differentiation, synaptogenesis, the growth of neural connections, and the establishment of brain networks. Studying early trajectories of brain activity in newborns delivered before the ninth month of pregnancy is therefore essential for understanding alterations in brain function associated with prematurity and identifying early markers of atypical neurodevelopmental trajectories.

IMPORTANCE OF EEG IN MONITORING BRAIN MATURATION

Clinical assessments of neonatal brain development often rely on conventional EEG, a non-invasive examination performed during sleep that records brain waves across different frequency bands (delta, theta, alpha, beta, and gamma). In particular, an increase in theta waves, which play a key role in the early development of neural networks, has been reported between the late preterm period (around 34 weeks of gestational age) and approximately one month after the equivalent age at term (EAT). However, traditional EEG analyses do not distinguish between oscillatory and non-oscillatory activity, limiting our understanding of the underlying mechanisms. Recent spectral parameterization techniques overcome this limitation by decomposing the power spectrum into two physiologically distinct components: an aperiodic component, which reflects non-oscillatory background neural activity, and periodic components, which appear as peaks corresponding to oscillations at specific frequencies.

ADDED VALUE OF SPECTRAL ANALYSIS OF THE EEG

In this study, the authors used spectral parameterization to track the evolution of aperiodic activity and theta oscillation power in 41 very preterm infants (born before 32 weeks of gestational age) and 13 full-term infants, as well as to assess inter-individual variability within the preterm group. High-density EEG was recorded during active sleep/REM sleep at the EAT, and again at 2 months corrected age (2mCA), in both groups of infants. The researchers were thus able to extract parameters characterizing aperiodic activity and theta power, at both the global level and across different brain regions.
Between EAT and 2mCA, aperiodic EEG activity parameters and theta power increase, with a similar trajectory in preterm and full-term infants. Aperiodic activity is initially more pronounced in anterior regions, whereas posterior regions show more pronounced changes between EAT and 2mCA. In preterm infants, EEG measures at EAT are associated with several clinical factors (sex, gestational age, and birth weight, among others) and with brain microstructure assessed by diffusion MRI. Higher theta power is notably correlated with more advanced cortical maturation.

These findings indicate that EEG spectral parameterization, combined with spatial analysis, provides a sensitive approach for characterizing early brain maturation and vulnerabilities associated with prematurity. These results are consistent with a previous study by the team based on an EEG microstate approach, aimed at revealing the maturation characteristics of emerging brain network activity in preterm infants (reference 2).

Joliot' institute contacts: Jessica Dubois (jessica.dubois@inserm.fr / jessica.dubois@cea.fr) ; Parvaneh Adibpour (parvaneh.adibpour@inserm.fr / parvaneh.adibpour@cea.fr)

​