Motor evoked potentials as markers of internal capsule current spread during deep brain stimulation for Parkinson's disease

Objective: Deep brain stimulation (DBS) modulates neural elements in the target region but can cause involuntary muscle contractions through unintended internal capsule activation. We characterized upper and lower limb motor evoked potentials (MEPs) produced by DBS to determine how stimulation location and orientation affect responses. Methods: Low-frequency DBS was delivered in the globus pallidus internus (GPi, 17 hemispheres) and subthalamic nucleus (STN, 8 hemispheres). MEPs were identified by stimulus-triggered averaging of surface electromyography (sEMG) from nine contralateral muscles. Results: MEPs were most robust in distal upper limb muscles, with larger magnitudes and higher incidence than proximal or lower limb muscles (p < 0.001). Onset latencies did not differ between GPi and STN targets (p > 0.14). MEP incidence decreased from distal (ring 1) to proximal (ring 4) lead contacts (p < 0.003). A consistent recruitment hierarchy was observed: the thenar muscle was the most sensitive responder. Thenar MEP absence was associated with MEP absence in all other muscles in 96 % of cases. In GPi, larger MEPs correlated with posterior-medial stimulation and proximity to the posterior border. Conclusion: sEMG-recorded MEPs are a robust indicator of stimulation spread to internal capsule and provide the means to observe and quantify unwanted stimulation effects. Significance: This study establishes a predictable hierarchy of MEPs during deep brain stimulation, identifying distal upper limb muscles, specifically the thenar, as the most sensitive indicators of internal capsule activation. These findings provide a robust physiological framework for quantifying and avoiding stimulation-induced side effects in patients with Parkinson's disease.