Subthalamic Nucleus Deep Brain Stimulation Enhanced Long‐Latency Reflex II in Parkinson's Disease Through Rapid Subthalamic–Cortical Modulation

Background The long‐latency reflex (LLR), particularly LLR II elicited by peripheral stimulation, is considered a transcortical reflex and may reflect cortical–subcortical excitability. Its relevance as a biomarker for subthalamic nucleus deep brain stimulation (STN‐DBS) optimization in Parkinson's disease remains unclear. Objective To examine whether LLR II responses are modulated by STN‐DBS stimulation parameters and dopaminergic medication, and to explore the potential mechanisms underlying STN‐DBS–related motor improvement. Methods Thirteen Parkinson's disease patients with STN‐DBS underwent median nerve stimulation to assess LLR II, quantified as the peak abductor pollicis brevis electromyography (EMG) response at 50–70 ms, normalized to prestimulus baseline EMG, across different stimulation intensities, contacts, and frequency conditions, including 60 Hz, 130 Hz, and an energy‐matched condition. A single DBS pulse was also paired with the cortical N20 response at N20 – 2 ms (T1) or N20 + 10 ms (T2) to assess timing‐dependent effects. Sixteen Parkinson's disease patients without DBS were tested in medication‐ on and ‐ off states. Motor severity was assessed using tested‐side Unified Parkinson's Rating Scale Part III (UPDRS III) subscores. Results LLR II ratio was higher at 70% and 100% of clinical stimulation intensity than during DBS‐ off ( P = 0.0029 and P = 0.042, respectively). LLR II was higher at the most effective than at the least‐effective contact ( P = 0.042). The clinically most‐effective frequency showed higher LLR II than DBS‐ off and the least‐effective frequency (both P = 0.048). In the single‐pulse DBS experiment, LLR II increased only at T2 compared to DBS‐ off ( P = 0.047). No medication‐state difference was observed. Conclusions LLR II was modulated by clinically effective STN‐DBS parameters but was insensitive to dopaminergic medication, supporting its utility as a biomarker of DBS‐related sensorimotor circuit modulation, potentially involving the hyperdirect pathway.