Neonatal brain dynamic functional connectivity in term and preterm infants and its association with early childhood neurodevelopment

Abstract

Brain functional dynamics have been linked to emotion and cognition in mature individuals, where alterations are associated with mental ill-health and neurodevelopmental conditions (such as autism spectrum disorder). Although reliable resting-state networks have been consistently identified in neonates, little is known about the early development of dynamic brain functional connectivity and whether it is linked to later neurodevelopmental outcomes in childhood. In this study we characterised dynamic functional connectivity in the first few weeks of postnatal life and evaluated whether early dynamic functional connectivity: i) changes with age in the neonatal period ii) is altered by preterm birth and iii) is associated with neurodevelopmental and behavioural outcomes at 18 months. We used the Kuramoto Order Parameter as a metric of global brain synchrony and defined transient brain states (modules) using Leading Eigenvector Analysis (LEiDA) in a cohort of term-born (n=324) and preterm-born babies (n=66) scanned at term equivalent age from the developing Human Connectome Project. We assessed whether neonatal brain state features (mean synchrony, metastability, entropy, fractional occupancy, dwelling times) and state transition probabilities were associated with postmenstrual age at scan , postnatal days at scan and preterm-birth; and correlate with neurodevelopmental outcomes at 18 months measured using the Bayley Scales of Infant and Toddler Development, and atypical social, sensory and repetitive behaviours measured by the Quantitative Checklist for Autism in Toddlers (Q-CHAT). On a global scale, preterm-born infants had lower mean synchronisation and metastability, with reduced mean synchronisation associated with higher Q-CHAT scores at 18 months of age. On a modular scale, we identified six transient states of neonatal dynamic functional connectivity: three whole-brain synchronisation states and three regional synchrony states occupying occipital, sensory-motor, and frontal regions. Mean synchrony, metastability, fractional occupancy and dwelling times of these brain states were correlated with postmenstrual age and postnatal days at scan. Preterm-born infants had increased fractional occupancy of frontal and occipital states. Higher neonatal sensory-motor synchronisation was associated with lower motor and language outcome scores at 18 months. Lower frequency of occurrence of whole-brain synchronisation states and higher frequency of occurrence of the sensory-motor state were associated with higher Q-CHAT scores at 18 months. Thus, we have shown for the first time that a dynamic landscape of brain connectivity is already established by the time of birth in the human brain. This landscape is altered by preterm birth and its profile is linked to neurodevelopmental outcomes in toddlerhood.

Neonatal brain dynamic functional connectivity in term and preterm infants and its association with early childhood neurodevelopment

“…While this is a useful approach to ascertain "on average" characteristics of brain activity, connectivity of the brain is intrinsically dynamic, i.e., non-stationary 3,4 . Addressing this problem, dynamic functional connectivity (dFC) measures the constant neural adjustments needed to control different brain states, adapt to transient situations, and integrate information 5 . The dynamical properties characterising the continuous shifting between connectivity profiles or "states" have been linked to processes such as language 6,7 , cognition [8][9][10][11] , and motor function 12,13 .…”

Differential maturation of the brain networks required for the sensory, emotional and cognitive aspects of pain in human newborns

“…Regions related to the motor response to pain (cerebellum, supplementary motor area) were not included. The resulting pain connectome atlas defined on the dHCP 40-weeks PMA T2 standard ( Figure 1 ), was resampled to infant native fMRI space by combining (i) the 40-weeks PMA to each PMA week standard (publicly available, https://git.fmrib.ox.ac.uk/seanf/dhcp-resources/-/blob/master/docs/dhcp-augmented-volumetric-atlas-extended.md) and (ii) the standard to native fMRI warps (non-linear registration based on a diffeomorphic symmetric image normalisation method (SyN) 116 using ANTs, see França et al for details 117 ). The same atlas was also resampled to the adult MNI 152 standard space using dHCP warps (publicly available, https://git.fmrib.ox.ac.uk/seanf/dhcp-resources/-/blob/master/docs/dhcp-augmented-volumetric-atlas-extended.md).…”

Editorial Perspective: Bridging the translational neuroscience gap in autism – development of the ‘shiftability’ paradigm

“…(2021) identified that infants with elevated likelihood for autism show reduced tactile sensory gating. Using MRI indices of functional connectivity, we have reported that the local synchronicity within sensory systems is higher in newborns with a higher likelihood of developing autism (Ciarrusta et al., 2020) and the global network dynamics at full‐term predicts cognitive outcomes and the emergence of later autistic traits (França et al., 2022). We and others have reported similar profiles in autistic adults, namely reduced sensory suppression detected using EEG (Huang et al., 2022) and altered functional connectivity of sensory systems and whole‐brain networks using MRI (Holiga et al., 2019; Watanabe & Rees, 2017).…”