The effects of hemorrhagic parenchymal infarction on the establishment of sensori-motor structural and functional connectivity in early infancy

Abstract

Introduction: The objective of the study was to characterize alterations of structural and functional connectivity within the developing sensori-motor system in infants with focal perinatal brain injury and at high risk of cerebral palsy.

Methods: Functional magnetic resonance imaging (fMRI) and diffusion tensor imaging (DTI) data were used to study the developing functional and structural connectivity framework in six infants born prematurely at term equivalent age. This was first characterised in three infants without focal pathology, which was then compared to that derived from three infants with unilateral haemorrhagic parenchymal infarction and a subsequent focal periventricular white matter lesion who developed later haemiparesis.

Results: Functional responses to passive hand movement were in the contralateral perirolandic cortex, regardless of focal pathology. In infants with unilateral periventricular injury, afferent thalamo-cortical tracts appeared to have developed compensatory trajectories which circumvented areas of damage. In contrast, efferent corticospinal tracts showed marked asymmetry at term equivalent age following focal brain injury. Sensori-motor network analysis suggested that inter-hemispheric functional connectivity is largely preserved despite pathology and that impairment may be associated with adverse neurodevelopmental outcome.

Conclusion: Following focal perinatal brain injury, altered structural and functional connectivity is already present and can be characterized with MRI at term equivalent age. The results of this small case series suggest that these techniques may provide valuable new information about prognosis and the pathophysiology underlying cerebral palsy.

Fig. 1

Functional activation and probabilistic tractography in a control preterm infant without focal brain injury and imaged at term equivalent age. Following passive sensori-motor stimulation of the right hand, clusters of functional activity were identified in the contralateral (left) perirolandic region and supplementary motor area (z-score threshold 2.3). The afferent thalamo-cortical tract (yellow) was identified using probabilistic tractography and the fMRI cluster (orange) as a target mask. Symmetrical efferent corticospinal tracts (blue) were identified using anatomical regions of interest

Fig. 2

Functional activation and probabilistic tractography in three cases with focal periventricular brain injury, studied at term equivalent and 1-year corrected age. A unilateral periventricular white matter lesion can be seen arising from the lateral ventricle at the site of the previous haemorrhagic infarction on the right (cases 1 and 3) and left (case 2) sides. Following passive sensori-motor stimulation of the contralesional hand, clusters of functional activity were identified in all infants at both time-points in the ipsilesional perirolandic region (z-score threshold 2.3). The afferent thalamo-cortical tracts (yellow and green) developed altered trajectories which circumvented the periventricular white matter lesion to meet the identified fMRI clusters (orange/red). Efferent corticospinal tracts (blue) showed marked asymmetry, with a decreased volume in the lesional hemisphere evident at both time-points

Fig. 3

Corticospinal tract volume and microstructural integrity. a In comparison to control infants (crosses), corticospinal volume in case infants was clearly asymmetrical at both term equivalent and 1-year corrected age. b, d Markers of microstructural integrity (fractional anisotropy (FA) and radial diffusivity (RD)) were also asymmetric at term equivalent age in cases, although this was not sustained at 1 year of age. c There was no difference in asymmetry in axial diffusivity (AD) at either time-point

Fig. 4

Functional connectivity in the sensori-motor network. In an example control infant (inset box), a clear pattern of functional connectivity can be seen between one brain region and its homotopic counterpart in the opposite hemisphere. In case infants, functional connectivity between the lesional (grey) and non-lesional hemisphere (white) is largely preserved even in the presence of focal brain pathology. The exception is case 1 at 1 year of age, who has lost inter-hemispheric connectivity between the perirolandic regions and has functionally ‘disconnected’ ipsilesional basal ganglia. Intra-hemispheric connectivity between the perirolandic regions and the supplementary motor area (SMA) is absent in the lesional hemisphere in all cases at term equivalent age and the majority at 1 year of age. Node sizes are scaled by degree (the number of connected edges), node colour is scaled by betweeness centrality (a measure of the amount of control that a node exerts over the interactions of other nodes in the network), and edge thickness is scaled by the pairwise Pearson’s partial correlation coefficient. Only edges which survived a false discovery rate correction (FDR) correction of p < 0.05 are shown