Lesion distribution and network mapping in dyskinetic cerebral palsy

Objective Dyskinetic cerebral palsy (DCP) encompasses a group of predominantly perinatally acquired complex motor disorders that present with dystonia and/or choreoathetosis and are frequently associated with brain lesions in neuroimaging. Recently, lesion network mapping provided a tool to redefine neurological disorders as circuitopathies. In this study, we aim to assess lesion distribution in DCP and identify a DCP-related network derived from lesions. Methods Here, we review the literature of MRI findings in DCP and perform literature-based lesion network mapping (LNM). Imaging findings and their anatomical distribution were extracted from literature and quantified according to an established MRI classification system for cerebral palsy. Whole-brain functional connectivity from lesions causing DCP was calculated using a pediatric resting-state functional MRI connectome. Results were contrasted with two control datasets for spatial specificity. Results Review of 48 selected articles revealed that grey matter injury predominated (51%), followed by white matter injury (28%). In 16% of cases MRI was normal. Subcortical lesions affected the thalamus, pallidum and putamen in >40% of reported patients, respectively. Figures available from 23 literature cases were used to calculate DCP-LNM. The LNM revealed functional connectivity to a wide network including the brainstem, cerebellum, basal ganglia, cingulate, and sensorimotor cortices. Strongest connectivity was found for the motor thalamus. Interpretation The neural network of DCP identified with LNM includes areas previously implicated in hyperkinetic disorders and highlights the motor thalamus as a common network node. The effects of targeting motor thalamic networks with neuromodulation in DCP should be explored in future trials.