(40) Low Frequency Dual Target Deep Brain Stimulation May Be an Effective Treatment for Parkinson’s Disease

Introduction: Deep brain stimulation (DBS) is an effective treatment for motor symptoms of Parkinson’s disease (PD). DBS typically targets the subthalamic nucleus (STN) or the globus pallidus internus (GPi). Simultaneous DBS of both the STN and the GPi, or dual target DBS (DT DBS), outperforms DBS of either target alone. However, DT DBS effectively doubles the energy applied compared to single target DBS. Additionally, DT DBS introduces a new stimulation parameter: the pulse delay between the two targets. We therefore implemented a biophysical model to study the effects of DT DBS frequency and inter-target delay on beta oscillatory activity as a proxy for bradykinesia in PD.

Methods: We adapted a biophysical model of the rat cortico-basal ganglia-thalamo-cortical loop to human physiology [1]. We conducted three sets of simulations and measured changes in beta power (13-30 Hz). First, we swept the frequency of DBS applied to STN or both STN and GPi. Second, we applied GPi DBS phase-locked to the STN beta oscillation. Finally, we applied DT DBS at a fixed frequency and examined the effect of inter-target delay wherein STN DBS matched the endogenous beta frequency (22 Hz) and swept the delay of the GPi pulses.

Results: In the frequency sweep experiments, we observed a steep decrease in beta power between 70 and 110 Hz compared to DBS off. Additional increases in DBS frequency did not further reduce beta power. In Experiment 2, beta power in the STN was very dependent on the target phase of the STN beta oscillation at which the GPi pulse was applied. Pulses applied between -3π/4 and -7π/12 reduced beta power to 70% of DBS off, while pulses between π/6 and π/3 reduced beta power to 20% of DBS off. In Experiment 3, pulse delays between 8 and 30 ms increased beta power to up to 164% of DBS off. However, delays greater than 30 ms decreased beta power, reaching a minimum at 40 ms, which reduced beta power to 12.1% of DBS off.

Conclusion: Our results suggest high frequency DBS is most effective at reducing beta oscillations. However, low frequency DBS using appropriate inter-target delays could be an effective treatment for PD. Such delays can cause destructive interference of endogenous beta oscillations. Our modeling framework provides a platform for the optimal selection of experimental DT-DBS parameters, which is otherwise complex due to the large number of degrees of freedom and the complexity of response to different parameters.