Subthalamic nucleus activity during cognitive load and gait dysfunction in Parkinson's disease

Background: Gait freezing is a common, disabling symptom of Parkinson's disease characterized by sudden motor arrest during walking. Adaptive deep brain stimulation devices that detect freezing and deliver real-time, symptom-specific stimulation are a potential treatment strategy. Real-time alterations in subthalamic nucleus firing patterns have been demonstrated with lower limb freezing, however, whether similar abnormal signatures occur with freezing provoked by cognitive load, is unknown. Methods: We obtained subthalamic nucleus microelectrode recordings from eight Parkinson's disease patients performing a validated virtual reality gait task, requiring responses to on-screen cognitive cues while maintaining motor output. Results: Signal analysis during 15 trials containing freezing or significant motor output slowing precipitated by dual-tasking demonstrated reduced θ frequency (3–8 Hz) firing compared to 18 unaffected trials. Conclusions: These preliminary results reveal a potential neurobiological basis for the interplay between cognitive factors and gait disturbances including freezing in Parkinson's disease, informing development of adaptive deep brain stimulation protocols.