Hitting the brakes: pathological subthalamic nucleus activity in Parkinson's disease gait freezing.

Abstract

Gait freezing is a complex and devastating paroxysmal motor arrest commonly suffered in Parkinson's disease that causes significant impairment to mobility, commonly resulting in falls and subsequent injury. The neurobiological basis of gait freezing in Parkinson's disease is poorly understood and thus, currently available therapies are partially effective at best. We used a validated virtual reality gait paradigm to elicit freezing behaviour intraoperatively in eight patients undergoing subthalamic nucleus deep brain stimulation surgery while microelectrode recordings were obtained. This allowed us to directly test the hypothesis that increases in pathological multi-unit activity in the subthalamic nucleus are associated with freezing onset in real time, manifest as dysfunctional firing of lower limb muscles typical of freezing that were detected by EMG. We present evidence that freezing is related to transient increases in pathological subthalamic nucleus activity. We performed time-frequency analysis to characterize the oscillatory dynamics of subthalamic nucleus activity coincident with freezing onset, demonstrating an increase in pathological beta and theta rhythms that are followed by a temporal chain of activity culminating in characteristically abnormal lower limb muscle firing detected by EMG. Finally, we interrogate the potential clinical utility of our findings by contrasting the subthalamic nucleus activity signature during pathological freezing against purposeful stopping. These results advance our understanding of the neurobiological basis of gait freezing in Parkinson's disease, highlighting the role of the subthalamic nucleus and emergent synchronous activity in basal ganglia circuits in driving non-purposeful motor arrests in individuals with Parkinson's disease. Pathological subthalamic nucleus activity identified in association with freezing is discernible from that of volitional stopping, paving the way towards more effective therapeutics such as adaptive closed-loop deep brain stimulation protocols.