Early Pain Exposure Influences Functional Brain Connectivity in Very Preterm Neonates

Abstract

Background: Early exposure to nociceptive events may cause brain structural alterations in preterm neonates, with long-lasting consequences on neurodevelopmental outcome. Little is known on the extent to which early pain may affect brain connectivity. We aim to evaluate brain functional connectivity changes in preterm neonate that underwent multiple invasive procedures during the postnatal period, and to correlate them with the neurodevelopmental outcome at 24 months.

Methods: In this prospective case-control study, we collected information about exposure to painful events during the early postnatal period and resting-state BOLD-fMRI data at term equivalent age from two groups of preterm neonate: 33 subjected to painful procedures during the neonatal intensive care (mean gestational age 27.9 ± 1.8 weeks) and 13 who did not require invasive procedures (average gestational age 31.2 ± 2.1 weeks). A data-driven principal-component-based multivariate pattern analysis (MVPA) was used to investigate the effect of early pain exposure on brain functional connectivity, and the relationship between connectivity changes and neurodevelopmental outcome at 24 months, assessed with Griffiths, Developmental Scale-Revised: 0-2.

Results: Early pain was associated with decreased functional connectivity between thalami and bilateral somatosensory cortex, and between the right insular cortex and ipsilateral amygdala and hippocampal regions, with a more evident effect in preterm neonate undergoing more invasive procedures. Functional connectivity of the right thalamocortical pathway was related to neuromotor outcome at 24 months (P = 0.003).

Conclusion: Early exposure to pain is associated with abnormal functional connectivity of developing networks involved in the modulation of noxious stimuli in preterm neonate, contributing to the neurodevelopmental consequence of preterm birth.

Keywords: brain connectivity; fMRI; functional connectivity; neonatal neuroimaging; nociceptive modulations; pain; preterm neonates; resting state.

FIGURE 1

Flow-chart summarizing the fMRI analysis. Control group, neonates not requiring invasive procedures; IP group, neonates requiring invasive procedures; NICI score, neonatal intensive care invasiveness score; 0 category, no painful events during the early post-natal period or <5 skin breaks (i.e., heel lances, central line insertion, and intramuscular injection); I category, >5 skin breaks or neonatal endotracheal intubation; II category, >5 skin breaks and neonatal endotracheal intubation; III category, surgical interventions (i.e., patent ductus arteriosus ligation and surgical necrotizing enterocolitis).

FIGURE 2

Differences in brain connectivity profile indexed at a voxel level using multivariate pattern analysis (MVPA). MVPA results are overlaid on a representative neonatal axial T2-weighted image. Blue-violet clusters indicates brain regions (thalami and right insular cortex) that show different pattern of functional connectivity between IP neonates and controls. Colorbar indicates F-statistic of between group differences with regard to the spatial maps of the four principal components. Three clusters were identified: in both thalami and in the right insula. The F-maps are threshold voxelwise at a p < 0.001 FWE-corrected and at p < 0.05 FWE-corrected at cluster-level.

FIGURE 3

Post hoc Seed-to-Voxel analysis using seed regions at level of thalami at the peak coordinates from MVPA. This analysis identified decreased functional connectivity between thalami and somatosensory areas (blue-violet clusters) in the IP neonates when compared with controls. Colorbar indicates T-statistic of between groups differences (IP neonates < controls), the T-map is thresholded at voxelwise p < 0.001 FWE-corrected, and p < 0.05 FWE-corrected clusters level and at voxel level.

FIGURE 4

Post hoc Seed-to-Voxel analysis using seed regions at level of thalami at the peak coordinates from MVPA. This analysis identified increased functional connectivity between right thalamus and ipsylateral superior temporal gyrus (red-yellow clusters) in the IP neonates when compared with controls. Colorbar indicates T-statistic of between groups differences (IP neonates > controls), the T-map is thresholded at voxelwise p < 0.001 FWE-corrected and p < 0.05 FWE-corrected clusters level and at voxel level.

FIGURE 5

Post hoc Seed-to-Voxel analysis using seed regions at level of right insular cortex at the peak coordinates from MVPA. This analysis identified decreased functional connectivity between right insular cortex and ipsilateral amygdala/hyppocampus (blue-violet clusters) in the IP neonates when compared with controls. Colorbar indicates T-statistic of between groups differences (IP neonates < controls), the T-map is thresholded at voxelwise p < 0.001 FWE-corrected and p < 0.05 FWE-corrected clusters level and at voxel level.

FIGURE 6

The association between altered functional connectivity and locomotor subscale score. The strength of functional connectivity between the right thalamus and somatosensory cortex and between the right insular cortex and the ipsilateral amygdala and hippocampus correlates positively with neonates’ performance on the locomotor task (r = 0.642; P < 0.001 and r = 0.510; P < 0.001, respectively).