Motivation: Deep brain stimulation (DBS) is a well-established therapy for reducing the motor symptoms of Parkison’s disease. The mechanisms by which it affects motor function and more specifically motor unit activity and recruitment patterns, however, are not well understood. It has been established that DBS antidromically stimulates pyramidal tract neurons projecting from the cortex to the subthalamic nucleus (STN) (Li et al., 2012). Motor evoked potentials may also be observed during low frequency, high amplitude stimulation. The effect of DBS at the motor unit level at clinically relevant frequencies and amplitudes, however, has not yet been investigated. To investigate this, high-density EMG was used to examine the effect of DBS on motor unit firing patterns during isometric contraction of the first dorsal interosseous (FDI) muscle.
Methods: Following institutional ethical approval, 17 (4 females, aged 60.8 ± 7.4 years, Time since Diagnosis: 11.3 ± 3.7 years) participants volunteered for the study. Isometric index finger abduction was performed at 10%-30% of maximum voluntary contraction (MVC) in 4 conditions: DBS ON and OFF, with and without medication. High-density EMG was recorded from the FDI using a 126-electrode array and the data were decomposed into the constituent motor unit firing trains. Power spectra of the EMG signals, individual motor unit spike trains and cumulative spike trains were computed using Welsh’s method. Synchronization was estimated from the cross-correlogram between pairs of motor unit discharges, with short- and long-term synchrony defined as an increased probability of motor units firing within 10ms or greater than 10 ms of each other, respectively.
Results: DBS led to entrainment of a subset of motor unit firing times at the stimulation. This was observed as a significant increase in power in the EMG power spectra at the DBS frequency and higher harmonics during DBS ON (p< 0.001). A significant increase was also observed in the power spectra of the cumulative spike trains of the decomposed EMG data during DBS ON (p < 0.001). Eight participants exhibited a peak in the power spectra of both EMG and cumulative spike trains. In four participants, peaks in the power spectrum of the spike trains of individual motor units were observed, with secondary peaks consistent with pulse frequency modulation of the motor unit mean firing rate at the DBS frequency. In these participants, motor unit synchronization of units was increased (p = 0.038) during DBS ON with cross-correlogram peaks at multiples of the DBS interpulse interval.
Conclusion: The presence of peaks in the EMG and motor unit power spectra, and increased motor unit synchronization, are consistent with entrainment of motor unit firing during high frequency DBS (79-130 Hz). The results provide evidence of orthodromic activation of pyramidal tract neurons during STNS DBS at clinically relevant amplitudes and frequencies. In addition to providing insight into changes in motor unit firing during DBS, the findings suggest potential new biomarkers for optimization of DBS parameters based on motor unit activity.