Motor, sensory and executive function brain networks in children with unilateral cerebral palsy

Abstract

Background: Cerebral palsy (CP) is an umbrella term for a clinical picture caused by antenatal or perinatal insult resulting in non-progressive motor dysfunction. Secondary impairments in executive and social functioning are also evident. Unilateral spastic CP is defined by predominant motor spasticity impacting only one side of the body, and may be due to either unilateral or bilateral lesions. Underlying pathology may be lesions impacting primarily either periventricular white matter or cortical and/or deep grey matter, malformations during brain development, or be non-specific. MRI imaging has become the standard for assessing brain pathology. Advanced MRI processing techniques have been developed allowing groupwise detection of changes beyond those visible on gross assessment of structural images. These techniques include automated identification of intracerebral surfaces to allow volume estimation; or registration to a standard space to allow voxel-wise comparison between groups. In addition, diffusion MRI allows measurement of water molecule diffusion, giving insight into the integrity of organised tissue such as axonal fibres. Complex mathematical models are applied to the raw data to allow clinically relevant information; significant advances in these models has allowed improved accuracy and clinical utility of results, however, these advanced models have thus far been infrequently applied in CP research. Method: A literature review was initially performed to summarise findings, assess methodology, and identify areas in which further research was needed. Subsequently several of the techniques described above were applied to subsets of a large cohort of children with unilateral spastic CP. Using high resolution T1 weighted images, volume of subcortical structures in children with cortical or deep grey matter lesions and periventricular white matter lesions was analysed, along with a whole brain voxel wise grey matter analysis. Subsequently using diffusion MRI with diffusion tensor imaging (DTI), a whole brain white matter analysis was performed in children with periventricular white matter lesions. In both of these studies comparisons were made between children with bilateral and unilateral lesions. Correlation between imaging parameters and motor function were also assessed. In an aim to assess the executive function network, a tractography study using constrained spherical deconvolution (CSD) was performed, seeded in the anterior cingulate cortex. Correlation between diffusion metrics in these tracts and executive function was assessed. Finally, an updated literature review was undertaken (four years following publication of the initial review), making specific recommendations for future studies using diffusion MRI in CP. Results: In the initial review (2012), the corticospinal tracts as well as posterior and superior thalamic radiations were the most widely studied tracts in CP, with minimal and often conflicting results in other tracts or regions. Subcortical analysis showed bilateral reduction in thalamic volume as well as unilateral reduction in volume of several basal ganglia structures across both grey matter and white matter lesions, alongside extensive cortical changes in children with grey matter lesions. Subsequent whole brain white matter analysis of children with periventricular white matter lesions showed children with unilateral lesions had focal unilateral changes in the vicinity of the corticospinal tract; while children with bilateral lesions had diffuse bilateral changes extending into all lobes. Diffusion parameters of the posterior thalamic radiations significantly correlated with hand function. Tractography seeded in the anterior cingulate cortex revealed altered diffusion parameters of the connection to the precuneus in children with periventricular white matter lesions. Additionally, performance in the flanker task significantly correlated with diffusion parameters of the connection to the superior frontal cortex. Finally, repeat review of the literature showed improved knowledge around multiple brain regions with an increased focus on prediction of clinical outcome and response to rehabilitation. To improve homogeneity of methods and avoid common study limitations in future studies a set of nine specific recommendations was produced. Conclusions: Several advanced MRI techniques were applied to contribute to a rapidly advancing understanding of the pathophysiology underlying CP. Whole brain grey and white matter analysis allowed differentiation between unilateral and bilateral lesions, as well as identifying common features between primary grey or white matter lesions. All analyses showed evidence of changes in at least one subgroup in parts of the executive function network. Specific study of the anterior cingulate demonstrated that damage to tracts involved in the executive function network is related with clinical outcome. This rapidly expanding field holds promise for ongoing advances in the understanding of motor, sensory and executive dysfunction in this cohort, as well as prediction of outcome and response to therapy.