LAUSR

Deep brain stimulation (DBS) is now a common treatment for moderate to advanced Parkinson’s disease (PD), yet its mechanisms of action and how it interacts with PD pathophysiology are not well understood. It has been proposed that DBS modulates the pathological synchrony of oscillations in the cortical– basal ganglia loop. DBS of the subthalamic nucleus (STN) decreases the excessive beta (13-30 Hz) oscillations in the basal ganglia and suppresses exaggerated phase-amplitude coupling between the phase of beta oscillations and amplitude of higher frequency gamma oscillations (50-200 Hz) in the motor cortex (M1). However, conventional methods that characterize oscillations by amplitude and phase assume sinusoidal waves, whereas brain rhythmic activities are nonsinusoidal. Indeed, nonsinusoidal features such as sharpness or asymmetry of beta oscillations reflect the strength of input synchrony onto cortical pyramidal neurons, representing a tool to understand cortical functions. The present study examined the waveform characteristics of beta oscillations recorded by electrocorticography from the M1 in 23 PD patients undergoing DBS surgeries. The authors found that cortical beta oscillations were sharp and asymmetric in the absence of stimulation and STN DBS made the oscillations less sharp and more symmetric, indicating that DBS may suppress abnormal input synchrony to M1. Also, these temporal features correlated with cortical beta-high gamma phase-amplitude coupling, which has previously been shown to index PD motor defiicits. These findings offer a novel way to study how DBS modulates rhythmic brain activities. Temporal characteristics of oscillations could be used as a control signal for the development of closed-loop DBS. Closed-loop DBS could sense signals from the cortex or basal ganglia and dynamically adjust parameters, which overcomes the limitations of current DBS practices with fixed stimulation parameters irrespective of the fluctuating clinical states and the labor-intensive programming to find the best stimulation parameters. It remains to be determined which control signals, such as the nonsinusoidal features of cortical beta oscillations or the previously used beta amplitude of the basal ganglia, track more efficiently the clinical states of PD. Further studies are needed to determine how cortical or basal ganglia oscillatory features change over time and how they vary according to the clinical states of PD such as on or off medications, dyskinesias, and nonmotor symptoms of cognition or mood. To develop closed-loop DBS for maximum benefits and minimum side effects, we need to understand how DBS changes PD pathophysiology. Studying the nonsinusoidal features of pathological oscillations will be an important step in this direction.