Arm trajectories in dyskinetic cerebral palsy have increased random variability

Dyskinetic cerebral palsy results from injury to the basal ganglia early in life. Symptoms can include hyperkinetic or dystonic arm movements that impair function. It is not known whether these movements comprise a small number of specific abnormal motor patterns or whether they are random and variable. We hypothesize that injury to the basal ganglia leads to impaired filtering and removal of undesired neural signals and that lack of appropriate removal of noisy or irrelevant neural signals leads to random and variable arm movements. To test this hypothesis, we quantified the variability in arm trajectories while seven children with dyskinetic cerebral palsy between the ages of 4 and 13 years old made repeated outward reaching movements. We compared the results with those of 21 healthy children between the ages of 5 and 16 years. The best-fit trajectory to the set of reaching movements for each child was taken as the predictable component of movement. We calculated the ratio of the power in the best-fit trajectory to the total variance. This measure is the signal-to-noise ratio, and it quantifies the extent to which trajectories are predictable. We found that children with dyskinetic cerebral palsy had a significantly reduced signal-to-noise ratio compared with healthy children at similar ages. This result shows that there is increased movement variability, and it is consistent with the hypothesis that inadequate removal of noisy signals could be a cause of the movement disorder in dyskinetic cerebral palsy.